|
| Greg's diary |
| April 2019 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Monday, 1 April 2019 | Dereel | |
| Top of page | ||
| next day | ||
| last day |
|
Another power failure
|
Topic: general | Link here |
Another brief power failure at 7:34 this morning. That's an unusual time, and for some reason it tripped the non-UPS circuit breaker. Why? There's almost nothing on that circuit.
|
Understanding Google Maps messages
|
Topic: technology, language, opinion | Link here |
Peter Jeremy, a Google employee, noted something that I hadn't about yesterday's problems with Google Maps: I had selected public transport, and that was clear from the error message. Clearly there is no public transport between Lauchröden and Eisenach.
Is it? Isn't it? The error message was:
Sorry, we could not calculate transit directions from "Eisenach, Germany" to "Lauchröden, 99834 Gerstungen, Germany".
What is there in that message that says “public transport”? The only possibility would be “transit”, but that doesn't have any such meaning, outside Google at any rate. The OED states:
transit The action or fact of passing across or through a place; a passage or journey from one place or point to another.
And that applies equally to all the transport methods.
And is there really no public transport between Eisenach and Lauchröden? It's Germany, after all. And it didn't take long to find this page, which gives times and prices.
But in one point Peter is right: the bug is related to the choice of transport. Presumably Google Maps still doesn't have a good understanding of German public transport. And in passing I discover that I don't have a good understanding of Google's icons:
|
|
|||||||||
What does that symbol really mean? Is it supposed to be a train? A bus? I can't tell, but most likely it's a train. What else does it include? Buses, clearly, but what about boats? And to the left there's a symbolized car and a right turn symbol. Both seem to give the same result. I've been using Google Maps for nearly 15 years, and I still don't understand it.
|
Solar energy installation confirmed
|
Topic: Stones Road house | Link here |
Today I got an email from Effective Electrical confirming the installation date of the “solar system”: 14/04/19 and 18/04/19. Converting to canonical form, that's Sunday, 14 April 2019 and Thursday, 19 April 2019. Why that?
Ah, sorry, finger trouble. Should read 18 and 19 April. Only two weeks! I can hardly wait.
|
Goodbye, April Fools?
|
Topic: general, opinion | Link here |
In the past we've played various April Fool's jokes on each other, but this year, somehow, it didn't happen. We also didn't see any on the web. There seems to be a general feeling that it's an old, worn-out kind of joke.
| Tuesday, 2 April 2019 | Dereel → Ballarat → Dereel | Images for 2 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More solar energy information
|
Topic: Stones Road house, general | Link here |
Into town today to meet Tomas Kucera and Fyodor Torgovnikov to talk about details of the
solar energy installation—I thought. When I got there, the place was deserted except
for Fyodor Fred, who showed me a display of the two kinds of inverters that Tomas
sells:
The choice is either the two on the left or the one on the right. As I had guessed, Fyodor represents Ingeteam (the one on the right), but since I had already decided for them, that wasn't a disadvantage.
Much discussion, to the point that I was nearly late for my next appointment. Managed to get a feeling for how the inverter communicates with the world, via the Internet Protocol but not necessarily via the Internet, as they put it, using 802.11. It seems (I didn't follow through) that one mode of operation is to go via the Ingeteam web site in Spain. I wonder what kind of confusion gave rise to that model. Maybe to make it possible to access the data from a remote mobile phone? And there was certainly confusion about the IP addresses. He was able to communicate with the thing using his Microsoft “Windows” 10 laptop, a horrible thing that doesn't have a scroll function for the mouse substitute; instead you have to use the touch screen. It seems that the inverter uses the same RFC3927 range address that the Selectronic uses, in this case 169.254.1.1. But somewhere I saw an option to use DHCP, so hopefully I'll be able to sort that out.
And the second battery set? That's from the battery supplier, not Ingeteam. They're hoping for a supply in October, but Fyodor didn't sound overly optimistic. If for some reason they can't supply it, they can always take the existing battery back and install something from another supplier instead. It's interesting to note that the inverter uses high voltage Lithium-ion batteries, 256 V, but in fact it will run on anything with the correct voltage, including 21 lead-acid batteries. And that's how the defaults are set up, allowing for only about 50% discharge:
|
|||||||||||||||||||||||
One thing that he couldn't answer—surprisingly—was how to configure the thing to not discharge the batteries below a certain point as long as grid power was present, but to allow use of that charge when the grid fails. He noted that others had asked that too, and in principle the setup suggests that it should be possible. I got a mail message from him later in the day to confirm that yes, indeed, it's possible.
|
More physiotherapy
|
Topic: health, general | Link here |
Left Fyodor in a bit of a hurry to get to my physiotherapy appointment, and in fact made it just about on time. Things are looking better, and Heather gave me some exercises to do that should be more useful than the ones I had in the Dereel Hall, including one of dubious promise where I need to sit on a roll and move around:
|
|||||||||||||||||||||||
I don't know how much use that one will be.
While in town, also had a haircut.
| Wednesday, 3 April 2019 | Dereel | Images for 3 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Dollee Curry Laksa
|
Topic: food and drink, opinion | Link here |
One of the laksa pastes that I bought a while back carries the name “Dollee”:
|
|||||||||||||||||||||||
It looks genuine enough—the instructions on the back in Chinese and Malay are a good indication:
|
|||||||||||||||||||||||
But for some reason they have written, on the front,
|
|
|||||||||
The characters are set strangely, but I read Cari laksa. Cari laksa? The Malay term, written elsewhere, is “Kari Laksa”. In Malay, the letter C is now pronounced like a ch: chari laksa, which coincidentally means “Look for (the) laksa” in Malay. I wonder why they did that.
And the results? It's OK. It's surprising how little difference there is between the individual brands. Next time I'll try another one that looks decidedly dubious: apart from only having half the quantity, it wants condensed milk in the sauce. I don't think I'll even bother trying that ingredient.
|
Android: smart!
|
Topic: technology, language, opinion | Link here |
For reasons I don't understand, my Android phone (taskumatti.lemis.com) keeps waking up, turning on the display, sometimes making a noise. It's not annoying enough that I have bothered to try to find out why. It does enough more annoying things.
Today, though, it excelled itself. Noted a display “NOKIA” on the screen (it is, after all, a Nokia 3). And it alternated with various error messages, tastefully not left there long enough for me to get really worried. One suggested that Android had crashed, but that was clearly so worrying that it didn't come again. The others looked like this:
|
|||||||||||||||||||||||
How about that, a real crash. There must be some way of resetting the thing. In the Good Old Days, removing the battery would have done it, but this has a non-removable battery. OK, off to the web to look for wisdom: reset hung android phone.
Ah, “hung” is an old, worn-out (if strong) magic word. Nowadays it's “hanged”, clearly a substitution made by some modern person who doesn't understand the difference between transitive and intransitive verbs, nor the difference between weak and strong verbs. But I found a solution anyway: Vol- and PWR together.
Problem: it didn't work. Phone seriously braindead? Tried the search again with the mention of the phone model, and came up with this page: press Vol+ and PWR together. And that worked, though not quite as described.
Huh? What stupidity could lead to such a flagrant violation of POLA? You'd almost think they were trying to annoy people. And yes, the suspicion of brain damage wasn't that far from the mark, though it didn't relate to the phone itself. But later I realized that there was a different issue at hand. Vol- and PWR now has a different meaning: take a screen shot. I wonder when they'll change it again to mean “perform factory reset, no questions asked”. So yes, braindeath and POLA violation, just on whose part?
The phone came back up, but clearly some things weren't the way they should be. In particular, Yvonne called from town, and not only did I not get my custom ring tone, I got no ring tone at all, and the phone didn't recognize the number. Further investigation showed that it had forgotten my entire contact list:
|
|||||||||||||||||||||||
OK, how do you download them? Shouldn't that happen automatically? At some point I received the message:
|
|||||||||||||||||||||||
OK, sign in. This horrible glass keyboard! And I had to go looking for my password, which was suitably complicated. The usual messing around, then it asked me whether I wanted to use the account I had signed in with, or a different account. Why? Pressed “Next” to select the account, and got:
|
|||||||||||||||||||||||
I've seen that before. And I saw it again, and again, and again. I couldn't find any way to sign in, nor any reason why I was taken back to the login page (of course Android is too polite to give reasons). I did, however, get email from Google:
|
|
|||||||||
|
|
|||||||||
What does it mean, a new Nokia 3? It probably knows more about the device than I do. About the most interesting thing is that it now knows roughly where I am; previously it had me located up to hundreds of kilometres away.
OK, account recovery?
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Somewhere I got as far as entering my email ID:
|
|||||||||||||||||||||||
And, reminiscent of Schrödinger's cat, I was both logged in and not logged in:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
At least the explanation is different. OK, it's my guess that I'm not the first person who has been bitten by this. What advice do I get on the net? Many and varied, including the claim that I have disabled my Google account (wrong; it works everywhere else) and that I should delete the account and recreate it. And then suggestions that I should clear cache, along with incorrect paths through the maze of twisty little menus (presumably correct for a different but unstated version of Android).
Finally found out how to clear cache on my (Android 8.1) phone, rebooted (this is Microsoft, after all, isn't it?). No change.
OK, 30 minutes down, no closer to resolution. The cache issues, even if they're not what's biting me, point to bugs in Google's applications, and there's no obvious will to fix them. It would be quicker to reset the device to default and set up again. Did that.
|
Android from scratch
|
Topic: technology | Link here |
So I reset the phone and rebooted it. This time it seemed a good idea to write down the steps.
First, even before anything else, it asked me if I wanted to restore from the cloud. OK, if that works, it's probably the easiest way. But first it needed to connect to the Wi-Fi network, which, for some reason, took several minutes. Why? And then it found:
|
|||||||||||||||||||||||
OK, that looks good (and makes me wonder why it thinks that this device is an “unknown Nokia 3”). Selected that and got quite a reasonable selection of things to restore:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Why Telstra stuff? It can't have been bundled with the phone, which was supplied locked to Vodafone, and I certainly didn't install it. So why “Included”? And how so I say, “No, thanks, not with a bargepole”?.
Somewhere round here it wanted to know a phone number and get me to agree to the service terms—why now? And off it went with the information that I could continue setup.
Next was to set up the “voice assistant”. Oh, no, sorry, you've already done that. And in fact there wasn't much more to do except set the phone ring tone, not for the first time.
That was more complicated than I thought. It seems that I had downloaded them to the directory /storage/emulated/0/Music/, which is what's well hidden somewhere on the menus (I had expected something as simple as “Music”, but I don't have that; maybe I need to spend money to access it from the GUI). But in this case the directory was empty. It seems that I had had a problem with the microSD card:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Why that? It was working before. And why is it called “Portable storage”? What on this device isn't portable? OK, set it up again, and in contrast to last time (where it hung), this time it worked. OK, now upload the tones again.
And at this point the X display on eureka hung while trying to display the directory on the phone. Switched to the other server and shot down one firefox after another, before finding the real culprit:
Finally I was able to download the tones and confirm that the settings menu automatically finds them in that location. But what else uses them? All I can find are things like “Play Mus..”, which wants me to pay money to look at something:
|
|
|||||||||
I'll never understand this stuff.
| Thursday, 4 April 2019 | Dereel | Images for 4 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Flash for the kitchen
|
Topic: photography | Link here |
Sooner or later we'll finally have all the electrical updates done. One question that's open is whether we need to mount another flash unit somewhere to illuminate the kitchen better. One possibility would be on the corner from the hallway to the lounge. Today I tried that with a studio flash to match the existing unit on top of the (from this vantage point) left-hand fridge:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|
|||||||||
How do I match the illumination? Both units are adjustable, but how? The existing one, a Godox Smart 300 SDI, with a maximum output of 300 J, has 6 settings, marked simply with bars (or, as the web page puts it, “LED Precise Power Display”, “8 steps”). And the test one, a Godox DE 400 (already a couple of years old and thus disowned by its maker), has a maximum output of 400 J and a scale reading from 5.0 to 7.0 in steps of 0.1. It seems that 1.0 on this scale represents 2 EV (why do they do these things?), so it seems that the 0.1 steps are steps of the tenth root of 4 (1.148, or 0.2 EV). So how do I set 300 J output? 0.4 EV, it would seem; that gives about 303 J.
Tried that and measured illumination in various places with five combinations. It's not quite clear what the bars mean on the old flash, and it's difficult to get precise readings because of the shadows I cast (and which I need to take into account when taking photos), but here are the results:
|
||||||||||||||||||||||
Clearly version 5 was the best, and it gave relatively even lighting:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Only the critical area round the stove was slightly underexposed, but that's really only visible before postprocessing. Here exposures at f/11 and f/8 before and after postprocessing:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
So I think I can mount a flash unit there. It looks as if 200 J should be enough, and with that I can use f/9 or f/10 at 200/24° ISO, so the first stove photos was ⅓ EV to ⅔ EV underexposed anyway.
|
Air BBQ
|
Topic: food and drink, opinion | Link here |
Yvonne bought some chicken skewers for experimenting with our newest “Air fryer”. It comes with a “skewer rack”, a rod with two plates for mounting each end the skewers.
It also comes with metal skewers for use with the rack, but our skewers were already on wooden, well, skewers. Would they fit? Yes, with unrelated issues:
|
|||||||||||||||||||||||
The plates hold 10 skewers, so they're at 36° intervals. The central shaft has a square cross-section, and it's possible to mount the plates in any orientation. So I mounted them 90° apart, giving an 18° offset between the holes at each end. “Well Don't Do That Then”.
Then there was the issue of getting the thing into the “fryer”:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That proved soluble: just move the whole thing to the “top” end of the shaft:
|
|||||||||||||||||||||||
And the device has a special programme for “BBQ”, 12 minutes at 200°. At the end there wasn't much obvious difference:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
They were cooked, but only just, and not in the slightest browned. But the results tasted very good, and possibly this is a reason to keep the thing.
But why so pale? This was chicken, and the pieces were smaller than recommended. How would it work out with larger pieces of beef? I'm sure that things would be much better at a higher temperature. That's easy to do, of course, and we will next time, but once again I'm left wondering who comes up with these default settings.
| Friday, 5 April 2019 | Dereel | Images for 5 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Replanting garden beds
|
Topic: general, gardening | Link here |
To my surprise, I discovered that we've now been in Stones Road for half the time that we were in Kleins Road (here from 7 May 2015, in Kleins Road from 10 July 2007 until, well, 7 May 2015). You wouldn't have noticed from the garden, which still looks pretty terrible. But the flower boxes in front of my office are a mixed bunch: on the one hand I planted a ground cover there in the spring and watched it die, but it still has lots of bulbs in it. Soil problems? Today Mick the gardener came, and I got him to replant the one I suspect of having particularly poor soil. There were certainly enough bulbs in there:
|
|||||||||||||||||||||||
One looks and smells like garlic. Mick was convinced that it was garlic, but it doesn't have any outside skin:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
We replanted a number, including tulips, daffodils and others:
|
|||||||||||||||||||||||
Also removed a number of Mirabilis jalapa plants, some of which ended up in the north garden. At least they grow well with no attention. Petra Gietz took some of the bulbs with her, and so did Mick, and he planted more in the rose garden, leaving the rest behind.
|
Dangerous gymnastics
|
Topic: health, opinion | Link here |
Heather has given me some gymnastics to strengthen my gluteal muscles. One of them involves standing with one foot in front of the other and with eyes closed. That seems straightforward enough, but it's surprisingly difficult to keep your balance. I've taken to doing it when cleaning my teeth, but today something went wrong, and I found myself in the bath. Did I lose consciousness? No idea. I don't think so, but there's a fraction of a second when I'm not quite sure what happened. She was clearly right when she said that I should take precautions. Still, apart from a scratch on my shoulder and a bit of minor bruising, everything seems to be OK.
| Saturday, 6 April 2019 | Dereel | Images for 6 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Germinating seeds
|
Topic: gardening, opinion | Link here |
It's been 11 days since I planted my Epazote seeds. When are they going to germinate? Even parsley only takes a week. I've been looking carefully for some days, then this morning I saw some seedlings which look dried out.
Why? I've been watering them constantly. Dragged out my Olympus STF-8 flash unit and a macro lens and took some photos:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
That doesn't look that bad after all. They're just taking a very long time to germinate; they're still less than 1 cm high. By contrast, the Kniphofia seeds that I planted some time later have all germinated and are now about 3 cm in size:
|
||||||||||||||||||||||||
|
NiZn and STF-8
|
Topic: technology, photography, opinion | Link here |
I've had my Olympus STF-8 flash unit for over a year now, and I've hardly used it: it's such a pain to put on the camera. When I did so today to take the photos of the seedlings, the batteries were discharged.
Discharged? Pretty much dead. All of them showed a voltage of 0.3 V or so. They're all Nickel-Zinc (NiZn) batteries, which I've been experimenting for over 8 years, and they should have a voltage between 1.6 V (discharged) and 1.8 V (charged). What happened? Put them in the charger. One just died, but the others charged again.
My experience with NiZn has been that they discharge unevenly, so the voltages are very different on discharge. But the even discharge of these, especially after so relatively short a time, suggests to me that the STF-8 might be trickle draining them. I'll have to leave them out when I'm not using it.
|
Air-fried Bratwurst
|
Topic: food and drink, opinion | Link here |
Can you grill Bratwurst in an “air fryer”? Of course not. It's not a grill. But it has illusions: there's a “grill” setting on the control panel, which translates to 12 minutes at 200°.
That's a big difference from real grilling, where the rule of thumb is one minute per mm thickness. In practice I find that mine (28 mm thick) take about 20 minutes. Still, worth trying. Here before and after:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Once again there's not much difference to be seen. As with the skewers on Thursday, they're very pale, and they were just cooked enough. But again as with the skewers, they show promise.
In passing, it's interesting to note how the sausages “sweat” fat:
|
|||||||||||||||||||||||
The air current ensures that they stay dry; in the toaster oven they give off a lot of liquid.
| Sunday, 7 April 2019 | Dereel | Images for 7 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Bikers again
|
Topic: general, opinion | Link here |
While walking the dogs down Harrisons Road, saw a number of people and bikes in the distance. They appeared to have dogs with them, and Yvonne got the impression that they were about to start a race. OK, it would make a good photo, and if they misbehave, we have evidence:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
No, no misbehaviour there. Yvonne thought that it might have been a family outing. But what got me the most was that the bikes made almost no noise. Why are the typical bikes round here so noisy?
|
Port upgrades
|
Topic: technology, opinion | Link here |
Hugin version 2019.0 is out, time to upgrade the FreeBSD port. And while I was at it, also finally committed a patch to upgrade enblend.
And almost before I knew it, two email messages. First, errors:
Date: Sun, 7 Apr 2019 07:07:27 GMT
From: Portsnap buildbox
CC: grog@freebsd.org
Subject: INDEX build breakage
...
make_index: /usr/ports/graphics/hugin: no entry for /usr/ports/graphics/OpenEXR
Committers on the hook (CCed):
antoine
grog
And then a fix:
From: Antoine Brodin <antoine@FreeBSD.org>
To: ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org
Subject: svn commit: r498262 - head/graphics/hugin
Author: antoine
Log:
Unbreak INDEX
Modified:
head/graphics/hugin/Makefile
Modified: head/graphics/hugin/Makefile
==============================================================================
--- head/graphics/hugin/Makefile Sun Apr 7 06:39:26 2019 (r498261)
+++ head/graphics/hugin/Makefile Sun Apr 7 06:51:52 2019 (r498262)
@@ -10,7 +10,7 @@ MASTER_SITES= SF/${PORTNAME}/${PORTNAME}/${PORTNAME}-$
MAINTAINER= grog@FreeBSD.org
COMMENT= GUI for Panorama Tools, to stitch panoramic images
-LIB_DEPENDS= libIlmImf.so:graphics/OpenEXR \
+LIB_DEPENDS= libIlmImf.so:graphics/openexr \
In fact, the commit came before the error message. What went wrong there? I had no reason to change that line, but it seems that I had done. Why? Reversion to an older version? I have some vague recollection of issues with capitalization of port names, but I can't see anything in my diary. And why didn't I notice?
And then, why is graphics/OpenEXR wrong? It seems that the name of the port has changed some time in the past couple of years:
=== grog@eureka (/dev/pts/16) /usr/ports/graphics 16 -> l -d /home/src/FreeBSD/svn/ports*/graphics/OpenEXR
=== grog@eureka (/dev/pts/16) /usr/ports/graphics 17 -> l -d /home/src/FreeBSD/svn/ports*/graphics/openexr
|
Air-fried Tanduri
|
Topic: food and drink, opinion | Link here |
Tanduri chicken for dinner tonight, according to Yvonne's westernized version. In the past we've grilled it; how about trying it in the air fryer?
Set it to 230° and 15 minutes, and to my surprise it went very well; it even darkened as expected:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The normal problem, of course: the fryer is too small, and we were left with a couple of pieces that didn't fit. Barely cooked them in the microwave oven, then grilled them for a couple of minutes. It's difficult to tell the difference:
|
|||||||||||||||||||||||
The two on the right are the latecomers.
With tanduri chicken goes tanduri nan, of course. We cheat and buy partially cooked “naan traditional” from ALDI and heat it in the roti cooker. Problem: the cooker isn't big enough. Today I discovered an old sandwich toaster that Yvonne used to use. Looks like just the right thing. And how about that, it did the job.
| Monday, 8 April 2019 | Dereel | Images for 8 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Tortillas in sandwich toaster
|
Topic: food and drink, opinion | Link here |
Yesterday's success with the tanduri nan made me think: I've been cooking my tortillas in the roti maker, and I've had difficulties: either they dry out or they don't get cooked. Can the sandwich toaster do it better? At any rate it has the advantage that the top has a double mount, so it doesn't produce wedge-shaped roti or tortillas the way the roti maker does. Tried that today.
First question: how close do the two surfaces get? A distance of several millimetres is OK for a sandwich, but too much for a tortilla. Tried it with a chopstick in between and decided that it would probably be OK.
The first attempts were less than perfect:
|
|||||||||||||||||||||||
Not only were the two surfaces close together, the weight of the top squeezed the tortilla and made it wider, in the process tearing it apart. I'll have to think of how to avoid that; probably wait until it's partially cooked before closing it.
The results were OK, though one part was a little tough. But it's something to experiment with.
|
TTL studio flash?
|
Topic: photography | Link here |
After last week's experiments with flash in the kitchen, it's time to buy a new flash head. That's straightforward enough; go to eBay and find what they have. And I found one that looked surprisingly cheap:
|
|||||||||||||||||||||||
Only a little over $100 for a Godox DS400II, along with a flash trigger! Looking at the photos is even more exciting:
TTL flash control! And it's available for Olympus! Just what I always wanted!
Hah hah, only joking. For $104 you get only the flash trigger. If you want the item advertised, you first need to select it (another $237). And if you want both, you have to pay both prices, a total of $341. And if you read very carefully further, you'll discover that you also need a receiver, which they don't even offer.
Sorry, eBay seller qlians_au, you're not trustworthy. I'll read your descriptions, but I'll never buy anything from you. It's a pity that eBay hasn't stopped this kind of deception.
Still, does this really work? More investigation shows that Godox' web site is appallingly inconsistent, and though I found the documentation for the XProO (PDF), I couldn't find an up-to-date list of which models it supports, nor which of them work with TTL flash. My best guess so far is: some models have built-in receivers, others have triggering via an external receiver and a USB port, like the DS400 II for sale here, and others again, like my DE400s, have only a slave receiver and a conventional sync cable.
Clearly you can't use TTL without a receiver. And clearly the built-in receivers support TTL. But what happens with USB? My guess would be that USB wouldn't be fast enough to work with TTL. On the other hand, if not, what is the advantage? I have really cheap remote triggering hardware (transmitter and receiver about $15 together). Why use expensive stuff and USB connections unless it can do more?
As if that wasn't enough, it occurred to me how similar the DS400 looks to my DE400:
|
|||||||||||||||||||||||
It almost looks as if it should have a USB connector too. Yes!
|
|||||||||||||||||||||||
So: how do I find out if it supports TTL mode? Is there a (shudder) Facebook group? Yes. And of course I have to be approved. Applied, but didn't get any results before evening.
| Tuesday, 9 April 2019 | Dereel | Images for 9 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Dirty laksa
|
Topic: food and drink, opinion | Link here |
Last week I tried the “Dollee” laksa paste and found it barely distinguishable from the “Teans” laksa paste that I normally use.
Until today, when I ate the final portion. It was filthy! Well, no, but that was the first impression. Some of the spices (pepper?) clearly weren't ground enough, and they left a lot of sediment in the bottom of the pot. Nothing serious, but given that Dollee doesn't seem to have any advantages, enough reason to stick with Teans.
|
No entry, and please shut the gate
|
Topic: general, animals, opinion | Link here |
Since moving to Stones Road, we don't have a convenient circular route to walk the dogs: we have to go up to a point and then turn round and return the way we came, usually along a road. We could walk up to Progress Road to Rozenstein Road and then back via Bliss Road, but that's 3.3 km, rather more than we really want. We did have a way that cut across through bush from Progress Road to Bliss Road about halfway up, but that seemed to annoy the people who lived on the neighbouring block, so they blocked the route.
In fact, one possibility would be to go 300 m down Bliss Road and then along the back of the blocks facing Stones Road to the back of our block. Took a look today. No:
|
|
|||||||||
But that was only part of the truth:
|
|||||||||||||||||||||||
|
|
|||||||||
I'd like to understand what led to that combination.
|
Flare in wide angle lenses
|
Topic: photography, opinion | Link here |
In the past I've noted that my fisheye lenses have particularly little flare. Today a question came up on Quora: “Do certain cameras have better lens flare, if so which ones?”. Well, I wouldn't call any flare “better”, but certainly it was worth comparing a fisheye (the Olympus M.Zuiko Digital ED 8mm f/1.8 Fisheye PRO) with my widest rectilinear lens, the Zuiko Digital ED 9-18 mm f/4.0-5.6, with shots directly into the sun. The results weren't as pronounced as I had expected:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
To make things easier to compare, I “defished” the fisheye images, thus also showing how very much wider the angle of view is than the 9 mm.
| Wednesday, 10 April 2019 | Dereel | Images for 10 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Manipulating journaled soft updates
|
Topic: technology, opinion | Link here |
A while back I set journeled soft updates on teevee:/, which caused a rather unexpected problem:
What do I do there? I can hear “Use ZFS!” ringing in my head, but the obvious thing is to disable them again. But how? Tried today and ran into unexpected problems:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The correct thing to do (apart from installing ZFS) would be to fix dump to work with SUJ. But I suspect that that's a lot of work.
|
How to annoy customers
|
Topic: technology, food and drink, opinion | Link here |
So I've gradually come to terms with the wine that I bought online last month, and in the meantime the seller (McLaren Vale Cellars) has been sending me direct mail offering me what look like good choices of wine. OK, let's try some Shiraz. Put a dozen bottles in my “Shopping Cart” and try to pay for it:
|
|
|||||||||
OK, that's simple: do what it asks. But I couldn't! The field is set not to accept pastes: it wants me to enter the email address manually! Clearly a bug, so I selected the “Contact Us” page and had to jump through more hoops before I could finally send a message. And the reply I got really blew my mind:
Why? Why don't they get it fixed? Don't they want to sell anything? To quote Daniel O'Connor on IRC:
For me the decision was borderline anyway, but that has made the decision for me. There are other companies out there. Goodbye McLaren Vale Cellars.
| Thursday, 11 April 2019 | Dereel → Ballarat → Dereel | Images for 11 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Tidying up Carpobrotus
|
Topic: gardening | Link here |
Petra Gietz along with her granddaughter Kahlan (pronounced Kehlen) this morning to pick up some of the Carpobrotus that was overflowing the garden bed—about the only thing that has really grown well in the 2½ years since we planted it. Here when we planted it and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
They took a lot of stuff with them—Petra filled the boot and the back seat. And she barely made a dent.
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
Cleaning the uncleanable
|
Topic: food and drink, opinion | Link here |
In principle I'm relatively satisfied with my Saeco coffee machine that I bought 3½ years ago. As I noted at the time, the “documentation” is a catastrophe, and I won't forgive Philips for not giving my $50 cashback.
But even with that, it has probably made itself paid. Previously I had had a capsule machine that cost $0.375 (itself a good price) per cup. With fresh coffee that drops to about $0.09. For 4 cups per day, that makes a good $1 a day, so the $650 I paid for it were paid off well over a year ago.
But it has another really irritating “feature”: automatic descaling. What for? We use rainwater, which contains no minerals. How do I turn the ”feature” off? Ah, can't do that, might damage your machine. So every few months I'm greeted with this display:
|
|||||||||||||||||||||||
OK, how does that work again? RTFM? That way madness lies. Follow the menu? Potentially less mad. OK, fill the water container, put a pot under the output, and point the nozzle into it:
|
|||||||||||||||||||||||
Ah, right, put in descaling compound:
|
|||||||||||||||||||||||
I can lie, and off it goes, flushing the only water that ever goes through the nozzle, splashing water everywhere:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It used no less than four containers of water—more than I recall from previous occasions. On each case I had to swivel the nozzle to one side, giving it a chance to piss on the floor. At the end, after it had said “done” and asked for confirmation, it once again pissed on the floor.
You'd think it was trying to annoy me.
|
More dental stuff
|
Topic: health | Link here |
Into Ballarat today for my six-monthly dental check, which revealed nothing of interest.
|
Y2K for GPS navigators
|
Topic: technology, opinion | Link here |
My GPS navigator was behaving strangely today. The trip log to Ballarat showed the wrong date, 26 August 1999. And the time was off by a couple of hours. Hardware issues? Restarted the thing without any improvement.
Back home, discovered that it had had its Y2K event: the GPS week count had passed 1024, and so it had gone back 1024 weeks in time. That doesn't explain the 2 hour time discrepancy.
But what were the GPS designers smoking to limit themselves to 1024 weeks, somewhat shy of 20 years? Were they expecting the technology to be replaced by then?
| Friday, 12 April 2019 | Dereel | Images for 12 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
BSD man captured
|
Topic: politics, technology, opinion | Link here |
Who's the most well-known BSD personality? Jordan Hubbard? Theo de Raadt? That depends on whether he's known for BSD or something else. From the NetBSD project we have proff, now better known as Julian Assange. And he's certainly in the news. What got me is how old he looks:
He almost looks older than I, but he's only 47, young enough to be my son. I wonder what went on in the Ecuadorian Embassy. And will they reckon his 7 years' self-imposed imprisonment to whatever sentence he now gets?
While discussing proff's connection with BSD, went looking through the commit logs. He seems mainly to have been involved with a package called Surfraw, which he wrote:
And then there are a number of old fortune data files entries, finding things like:
Clearly a bridge builder. Some of the “fortunes” in netbsd-o seem offensive even to me; I wonder what a lawyer would make of them if the rape case in Sweden ever goes ahead.
|
Victorian Government: we don't need no steenking solar power
|
Topic: Stones Road house, opinion | Link here |
Call from Tomas Kucera of Effective Electrical in advance of our solar panel installation next week. Surprise, surprise! The Victorian Government has cancelled the solar energy installation rebate with immediate effect! I can now pay $2,250 more out of my own pocket!
Clearly there's no election coming up. But what are they thinking? It seems that they will reinstate it on 1 July, maybe. Or maybe I can get $4,000 for adding batteries to an existing installation. I can do that, but will they honour their obligations? Grrr!
|
Weekend clinic and visitors
|
Topic: general, animals | Link here |
This weekend Yvonne has organized a horse clinic, with Anke Hawke teaching. Yvonne had arranged that Anke will be staying here, and also Terry (surname unknown). That required tidying up the “library” opposite my office, which I had been using as a junk store, so that Terry could sleep there.
And then Yvonne got a message: Terry had had an accident, a tree trunk fallen on his back. We're still not sure exactly how bad it is, but it seems that he will recover. Clearly, though, he's not going to be doing much riding in the immediate future, so only one visitor.
|
More garden work
|
Topic: gardening | Link here |
Mick the gardener along this morning for more bulb transplants and odds and ends. For once there was barely any wind, ideal weather for spraying. Out to pick up the sprayer and... the battery was empty. Yes, I should have thought of that a couple of days ago. In fact, I did think of that, but that's as far as I got.
OK, find the charger, put it on charge. How much charge can I get in 90 minutes? Tried it out, and how about that, there was enough for a full load of spray. In fact, we ran out of weed killer first, after spraying 3 tankfuls of spray. I wasn't expecting that.
|
All bad things come in threes
|
Topic: technology, opinion | Link here |
Later in the morning Petra told me that she had heard a strange noise (click?) from my office, and thought that it might have been a power failure. Checked, and we were off the net! I didn't see any lights on the NTD. But while investigating, it was back again! No power failure after all, just a brief spell off the net.
Whose fault? As time went on, that became clearer. We had no less than 7 failures, a total of 14,221 seconds or nearly 4 hours:
| Start time End time | Duration | Badness | from | to | ||||
| (seconds) | ||||||||
| 1555032141 1555032624 | 483 | 0.003 | # 12 April 2019 11:22:21 | 12 April 2019 11:30:24 | ||||
| 1555034739 1555035952 | 1213 | 1.702 | # 12 April 2019 12:05:39 | 12 April 2019 12:25:52 | ||||
| 1555037101 1555037416 | 315 | 3.133 | # 12 April 2019 12:45:01 | 12 April 2019 12:50:16 | ||||
| 1555038451 1555048766 | 10315 | 3.478 | # 12 April 2019 13:07:31 | 12 April 2019 15:59:26 | ||||
| 1555049372 1555049396 | 24 | 5.941 | # 12 April 2019 16:09:32 | 12 April 2019 16:09:56 | ||||
| 1555053989 1555054531 | 542 | 0.784 | # 12 April 2019 17:26:29 | 12 April 2019 17:35:31 | ||||
| 1555055199 1555056528 | 1329 | 5.389 | # 12 April 2019 17:46:39 | 12 April 2019 18:08:48 | ||||
| Date | Outages | Duration | Availability | Date | ||||
| (seconds) | ||||||||
| 1554991200 | 7 | 14221 | 83.54% | # 12 April 2019 |
OK, time to download the Aussie Broadband Android app and painfully type in the passwords (which were not intended to be typed on glass), checking each individual letter before it turns into a pumpkin. And then I was able to lodge a fault. First, though, it had to check my connectivity. Success!
Success? What kind of nonsense is that? The NTD was clearly showing an ODU (outdoor unit) fault. There was no possibility of establishing connection. What a waste of time!
Entered the fault, and almost immediately I had connectivity again, so (with a real computer) reported the fact. But then it went down again. Tried calling up and was informed that I was in for a 21 minute wait. And that with expensive mobile phone calls! But at least that indicated that I probably wasn't the only one.
Tried to enter the fault again. Hang on “Analysing”. But then I discovered that I had run out of mobile credit. Simple, buy some more. How? I was doubly off the net. In the end got Yvonne to buy a voucher when she was in town.
How I hate network outages!
|
Borzoi ≠ watchdog?
|
Topic: animals, general, opinion | Link here |
Our Borzois may be big, but they're not exactly aggressive. In fact, they're far too friendly for any protective activity. But when Anke Hawke arrived this evening, I heard a loud bark and howl from the (dark) hallway.
Had she brought a dog with her? No, it was Nikolai! I can't recall having heard him bark before. It seems that he found it highly suspicious that anybody should come into the house in the dark. Once he saw that it was Anke, he was happy.
I suppose that's a good sign. If anybody comes into the house when we're not there, I don't think Niko would be as easily placated, and an aggressive-looking dog of his size would scare most people off.
| Saturday, 13 April 2019 | Dereel | Images for 13 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
ALDI induction saucepans
|
Topic: food and drink, opinion | Link here |
While in town on Thursday, I bought a couple of saucepans and frying pans at ALDI. They were cheap, and in contrast to the old ones that we have, they are intended for use with induction cooktops.
This morning I put one on the cooktop and turned on. Not recognized. One of the pans worked, but heated very slowly, and I had to use another one.
Why? They looked OK. But looking more carefully at the saucepan, the reason became clear:
|
|||||||||||||||||||||||
The dotted pattern on the bottom is one of the kinds of induction-compatible bases that are in common use. It's only 10 cm across, with a 5 cm gap in the middle. The gap in the middle is of no importance, since the induction coils don't heat there, but effectively this has made a 10 cm pan out of a 16 cm pan (rather less than 30% of the surface area), and none of my induction fields can talk to anything that small.
And the other pan? It's 30 cm in diameter, just what I need for bigger things. But the heating surface is only 17.5 cm in diameter, again only a third of the area. By contrast the existing 26 cm pan, also from ALDI, has a different heating surface 23 cm in diameter (78% of the area):
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
A pity. The pans look quite good otherwise.
|
More flash exposure issues
|
Topic: photography, opinion | Link here |
Taking the photo of the bottom of the saucepan wasn't easy. It took me five attempts. The first time (once again!) the flash didn't trigger, and the next one, taken with TTL flash and standard settings, looked far too dark:
|
|||||||||||||||||||||||
It wasn't until I gave the thing 2 EV overexposure, both in the camera and with the flash, that things looked normal:
|
|||||||||||||||||||||||
But it seems that the reflections of the base confused the issue: they showed up as overexposed even in the first image.
|
Incompatible goals
|
Topic: photography, opinion | Link here |
Seen at https://www.canon.com.au/camera-lenses/ef-70-200mm-f-2-8l-usm:
|
|
|||||||||
Japanese Craftsmanship & Innovation Collide in the making of EOS R
So which one won?
|
Forced system upgrade
|
Topic: technology, opinion | Link here |
While processing my house photos this afternoon, the screen suddenly went black, to be replaced by a constant stream that I could only decipher by taking a photo:
|
|||||||||||||||||||||||
Ugh. "vm_fault: pager read error" doesn't look good. Some things still seemed to be working (the weather application beeped to say that it had once again crashed), but clearly it was reboot time.
After reboot things still didn't look good:
ada0 is the first disk, the one with the root partition /dev/ada0p2. Without a root file system I'm dead in the water. But why the detach? There were no previous messages. The error = 6, on the other hand, is completely understandable: “Device not configured”.
Rebooted and considered whether this wasn't forcing my hand to finally upgrade the system. Yes, indeed: it happened again only a couple of hours later:
OK, nobody's home (Yvonne and Anke were at Chris' place with Anke's clinic), so I had several hours to migrate to the new system. I needed them, and more.
First I discovered that I still hadn't enabled journaling on /home and /Photos. Setting on /Photos was strange:
=== root@eureka (/dev/pts/3) /home/grog 58 -> tunefs -j enable /Photos
And then it hung. ps -l showed:
That's reading something from disk. What? And why? Finally it completed:
Is that serious? Looks like another UTSL moment. My guess is that tunefs wants to put the journal file somewhere specific with contiguous block locations, but that it couldn't find anything suitable, and that noncontiguous locations might have a slight performance impact.
/home also came up with a couple of strangenesses while running fsck:
|
|||||||||||||||||||||||
OK, that's not too much. But at the end, without any obvious reason, fsck wanted to run again. OK, do that. And then I got:
|
|||||||||||||||||||||||
Those are the same two files. The first run had set EOF values that the second run didn't like, so it had to be done again.
Booted into the new system, which was painless enough, but X wouldn't start: “no screens found”. I've been fighting X configuration for nearly 30 years now, so it wasn't exactly unexpected. OK, check the /var/log/Xorg.0.log: I was using version 304.137 of the nvidia driver. Which version did I need?
For that I needed web access. That brought me back to where I was three weeks ago: install BIND. This time I had made sure to add the external name server to my /etc/resolv.conf. pkg search is my friend. Is it? “Permission denied”. Build from source? First I had to find it. What's the port called? What directory?
OK, without X it was difficult to do much, so went to teevee and tried from there. Ah, BIND is so important that it's not under net or net-mgmt or such, but dns. And there I had
Which one? Why isn't there one just called bind (possibly with a symlink to make things clear)? bind-devel suggests that it's a little too bleeding edge. What about bind914? Checked bind914/pkg-descr, no warning about being unstable or whatever. Back to eureka, tried to build. Bloody EPERM again!
Back to teevee, did a make fetch. Success. Back to eureka:
Dammit, back to teevee and fetched it. What's BIND doing with Python anyway?
Finally got BIND built, watching it joyfully print:
So is it stable? Isn't it? I have more important things to worry about.
But after starting it couldn't resolve anything. The overly voluminous log messages told me that I didn't have any zones. Grrr. I've been here before: BIND has had a long history of migrating its configuration files, from /etc/namedb/ to /var/named/etc/namedb/ and (currently) to /usr/local/etc/namedb/, where it wanted them now. OK, put the info there, remember (forcedly) to change /etc/namedb to /usr/local/etc/namedb in /usr/local/etc/namedb/named.conf, and finally it was working.
Round about here (rather late) it became clear that the EPERMS were due to firewall issues. For some reason I had in my ipfw rules:
How did that ever work? Removed it and finally had connectivity.
Gradually time was getting on, and my main concern shifted to giving other computers in the house access to the outside world. It seems that squid wasn't doing what it should, so I had to set the browsers on teevee and lagoon to connect directly to the Internet, and also had to mess around with email, which hadn't come back the way I expected. And MySQL and httpd weren't doing what I want either. The latter somehow got started with the flag -DNOHTTP. Are they now mandating https? Where do I get a valid certificate? Dammit, this is all more pain that I need.
In the evening—Yvonne had outdone herself and taken 24 GB of videos and photos—made a photo backup. That normally takes about 20 minutes. Today it took 2 hours! Why? Something to do with my file system parameters? One of many questions to put off until tomorrow.
|
After dinner photos
|
Topic: general, photography, opinion | Link here |
Dinner with Chris Bahlo (as usual on a Saturday) and Anke Hawke this evening, Anke noting that she had a tendency to leave her wine glasses half full. I promised to write it up in my diary, but when I checked I found that she had drunk them after all (or maybe Yvonne, trying to keep the peace, had poured out the rest).
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Somehow our after dinner photos aren't as interesting as they were 10 years ago.
| Sunday, 14 April 2019 | Dereel | Images for 14 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Cat's breakfast
|
Topic: animals, food and drink | Link here |
While setting the table for breakfast this morning, saw this on the floor:
|
|||||||||||||||||||||||
|
|
|||||||||
That's a multivitamin pill, one that I take every morning, and which Yvonne had put on the table an hour or so before. What's it doing on the floor?
|
|||||||||||||||||||||||
Piccola must have tried them out. The pill that she bit through (and obviously found not to her taste) was Allopurinol, which doesn't seem to be particularly toxic. And she doesn't seem to have swallowed any of it, so hopefully no harm was done.
|
eureka upgrade, day 2
|
Topic: technology, opinion | Link here |
Started pretty early in the morning trying to get eureka back up and running. Given the issues I had, started copying the old root file system to the spare /destdir partition on the new disk, so that I would have the files to compare, and if everything went wrong I could boot the old system from it.
The obvious next thing to do was to get X up and running again. Which driver version did I need? That's described on the list. I have a GeForce 9500 GT and a GeForce GT 640. Which version of the driver do I need?
That's straightforward enough: this page, with the rather misleading title “What's a legacy driver?”, tells me that I need driver version 340.xx for the 9500 GT, and driver version 390.xx for the GT 640.
Huh? There can only be one, and it used to work. Is this only half the story, or have they changed things so that I can't run both cards with the same driver?
Ok, let's try. Deinstall the current driver and...
I've seen problems with pkg upgrade before, but this is ridiculous. I had carefully upgraded packages only a month ago, before my first system upgrade attempt. Why this ridiculous number of changes? Something else must have gone wrong. I'm reminded of Aussie Broadband's abortive attempt at a network upgrade a few weeks ago.
OK, revert to the old system again, not helped by the fact that I had thrown a switch in /etc/rc.conf that changes its configuration:
# Decide whether this is a build system or a "for real" system.
# This is set to 0 during build, and 1 during normal operation
FORREAL=0
if [ $FORREAL -eq 0 ]; then
hostname="stable.lemis.com"
hostip=192.109.197.192
defaultrouter="192.109.197.137"
else
hostname="eureka.lemis.com"
hostip=192.109.197.137
fi
OK, fix that, reboot, and I was back where I was yesterday afternoon. The weather observations database shows the real outage: last record yesterday at 15:19:26, first record today at 09:57:43. Total outage a little under 19 hours.
OK, now lick my wounds. What went wrong? First, which X drivers did I have? From /var/log/Xorg.0.log, the old system has:
And the new system has:
The same driver! Why didn't it find anything? A bit of searching brought me to this page which, after selecting “supported products”, confirms that version 304 supports both cards. So it's just a documentation error: the first page just shows the most recent driver that supports a card.
So: what went wrong? It seems that my configuration was correct, and yes, I had the module loaded. More head-scratching. Sometimes I wonder if I made the right choice in buying nvidia cards. My one recent experience with Radeon was surprisingly easy, and it didn't require any driver downloads. Maybe it's time to investigate them more closely.
Back to other issues. What was wrong with my ipfw firewall configuration? A diff between the old and the new configuration shows a regression:
I had a specific rule to bypass setup from the gateway address, but that had changed some time in the past, and I hadn't updated it on the new machine. Another of those irritating upgrade issues. I've been trying to find a comfortable upgrade path now for something like a quarter of a century, and I don't seem to be much closer.
One thing that did clear itself up: why did the photo backup yesterday take so long? Because I had deleted a whole lot of work files on /Photos since last time, and the backup had to get rid of them. In fact, the average transfer rate was 33 MB/s, much higher than the normal 10 MB/s or so:
| Monday, 15 April 2019 | Dereel | Images for 15 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Solar power? No solar power?
|
Topic: Stones Road house, opinion | Link here |
Call from Tomas Kucera of Effective Electrical first thing this morning, on my mobile phone; I had asked him not to, but he didn't seem to understand, and claimed to have called the same number that he called me on on Friday. He wanted to know whether to continue with the installation on Wednesday or not, maybe without the batteries.
No way. I want those batteries, preferably both of them. He talked to somebody behind him (his boss?), and I suggested that they work things out between them and that he then call me back.
What does that mean? Maybe the whole issue wasn't so much the solar energy rebate as his supply issues. I'm not happy about having to wait until October, and if he can't deliver now, what are the chances of getting the second battery then? Time to investigate others; at least I now know what I'm looking for, so it's not quite so painful.
First tried EverSolar, who had installed a system for ibiza, a nick on IRC without a name. He was happy with the installation, but EverSolar doesn't venture out into the open countryside, so I was outside their installation range. Adelaide yes, Brisbane yes, but not Dereel.
Filled out some web forms and later got a call from Jim of AriseSolar, who accordingly calls himself Jim AriseSolar <jim@arisesolar.com.au> later on. Sydney number, not quite the local presence I was looking for. But he came up pretty quickly with a competitive quote: $19,220 with 10.8 kW panels and (only) 9.8 kWh batteries. It's not clear why there was this limitation, but he also planned 2 5 kW inverters, which seems sub-optimal to me.
Later a call from Tomas, on my mobile phone, but I didn't get there before it ended. Couldn't be bothered to call back, and sent him a mail message instead expressing my concerns. No reply.
|
Garden flowers in mid-autumn
|
Topic: gardening | Link here |
Middle of autumn today, time for my monthly garden flowers photos. Gradually the weather is getting milder, though it's still relatively warm. Today the temperature reached 28°.
The Hebes are a mixed bunch. The ones I was worried about are almost certainly dead, but some of the others have started flowering again.
|
||||||||||||||||||||||||
Some are even doing both:
|
||||||||||||||||||||||||
The Petunias are similar. The ones in the hanging baskets have suffered badly, while the ones in the garden beds are doing well:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
I think they're too sensitive to drying out. Next year they'll all be planted in beds.
Since being freed from their big brothers, the small Carpobrotus seem to be feeling a lot better. This one in particular, opposite the house entrance, has developed a number of flowers, though it's not really due until November:
|
|||||||||||||||||||||||
The Hibiscus bushes are all still alive, though their states are different. The “Uncle Max” Hibiscus rosa-sinensis looks happy enough, but it's been a while since it has a flower:
|
|||||||||||||||||||||||
The other Hibiscus rosa-sinensis has numerous buds but small flowers:
|
||||||||||||||||||||||||
The Hibiscus syriacus are similar: one (the one that Mick trimmed with a whipper-snipper) has grown happily over the year, but has no buds, while the other has developed multiple flowers:
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||
And the Alyogyne huegelii that I planted in the spring has fought its way through the tomatoes to produce a few blooms as well:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
In the north beds, the succulents are doing well:
|
||||||||||||||||||||||||
And the Banksia integrifolia is finally producing lots of flowers:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Unfortunately, the other plants in the north beds aren't doing as well. My Salvia microphylla only has a couple of flowers, and the blue ones are dead—for this year, anyway:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
And the Eremophila nivea that usually flowers so profusely is also very shy:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Surprisingly, the Gazanias are also not doing as well as I had expected:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
I had thought that I could just plant them and ignore them, but they're not quite as resilient as I thought.
The Clematis Edomurasaki is still flowering, though I think the wind has given it some problems:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
And the mistreated Clematis “General Sikorski” (last month I had assumed that it was dead) seems also to be on the road to recovery. Here on 11 March and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And somehow the roses are once again doing well, despite only so-so care:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
Canon EOS RP: Avoid?
|
Topic: photography, opinion | Link here |
An interesting review summary about the Canon EOS RP from DPReview today, brought out after the real review. I'm not that interested in the camera, and the real review is too detailed for me, but the title of this one (“Is the Canon EOS RP right for you?”) sounded interesting.
And it was. Apart from the lack of useful lenses and the abysmally slow frame rate (only 4 frames per second! Even my old Olympus E-PM1 does better than that!), the really interesting thing was:
On the downside, if you like to mix stills and video in your travels, the RP won't be of much help for the latter – if you have a reasonably modern smartphone, chances are its 4K video will be leagues ahead of what the Canon is capable of in most lighting conditions.
So what good is it? I wonder when Canon will come out with a competitive mirrorless camera.
|
Battery charging revisited
|
Topic: technology, opinion | Link here |
I've had a surprising number of issues with mobile phone batteries lately. Two different old Samsung phones needed new batteries. In particular, Petra Gietz' I9210 died a couple of months ago, though the battery wasn't that old. A replacement was dead on arrival. And then Petra recently told me that the third was also dead.
There's something wrong there, and it's difficult to blame it on the batteries. Kept the phone and put it on charge, and by elimination discovered that one of the USB charge cords—and only one of them!—was defective. And the first of the three batteries wasn't so dead that it couldn't be recharged. It does seem, though, that this particular phone is very fussy about the charge current.
Only: I can't turn the bloody thing off! When I power it down and it's connected to the charger, it automatically restarts. Without the charger, it works fine. That only started happening with the old battery. Why? And how can I stop it?
|
Where's Chris?
|
Topic: general, opinion | Link here |
Chris Bahlo is a typical modern person: her main form of communication is via Facebook, and she's pretty good at checking that mess at frequent intervals.
So when Yvonne didn't get a response from her for some hours, she got worried. Call her on the phone. Voice mail. No call back.
Chris lives on her own, and she has a number of animals to look after. What if something has happened to her? Over to take a look, and found that she had left by car with Rev, one of her stallions. Did she have problems on the way? Call up Amber Fitzpatrick and Cliff Marisma? Sorry, don't have phone numbers.
Somehow this is all wrong. Facebook isn't the centre of the universe, and if you have a mobile phone, you should be able to answer it, at least if you're not at home, and people should have the numbers. What about the animals if she ends up in hospital?
Things turned out OK. She had been with Cliff (for whom we still don't have a phone number), and had just been busy. But we should work out an emergency schedule for her animals in case she's incapacitated.
| Tuesday, 16 April 2019 | Dereel | Images for 16 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More solar energy stuff
|
Topic: Stones Road house, opinion | Link here |
Still no reply from Tomas Kucera of Effective Electrical by this morning, so I called him up to find out what had gone wrong. He had tried to call me, he says, on my home number, the one stored in his phone. No evidence of that at my end. Maybe the number's wrong? That's the one he called me on on Friday. And why didn't he leave a message? And why didn't he answer my mail message? As he said, he's a Microsoft certified something.
Ah, he never got the mail message. The problem must be at my end. OK, here's my server log... Oh:
Damn! Forgot to restart my mail tunnel after the last reboot. OK, one point to Tomas. But he seemed to be excited because on Friday I had asked for time to think over what to do: he had to schedule his sparkies, something that he hadn't mentioned earlier. And he couldn't quote any other batteries because there are no other ones that would work with the Ingeteam inverter. I told him about my discussion with Fyodor two weeks ago, but he didn't know any of that information. Left him to discuss the matter with Fyodor and get back to me.
In any case, the good news is that he will, indeed, be able to deliver the first battery pack tomorrow. And what happens if the second one isn't available on time? Dammit, this is all too difficult, I have a "get out of jail free" card. Why don't you just take your deposit back and we'll forget the whole thing?
Somehow he doesn't understand what a contract is. He referred to the terms and conditions in his quote, but there were no such conditions in the contract that I signed. In any case, that's not in anybody's interest. I asked to speak to his boss, but no, his boss isn't interested in speaking with me—says Tomas. I'm sure that would change if we really had to go to VCAT about the matter.
So we're going ahead with the installation tomorrow. Tony, the electrician, showed up and did some more measurements, apparently related to the way the inverter needs to be wired (in groups of three, it seems). We'll get 24 on this section of roof (facing 10°):
|
|||||||||||||||||||||||
And 6 each on these two sections, facing 280°:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
He also confirmed that Tomas can get “a bit excitable”, and made the relatively obvious observation that if the battery coupling (basically just cabling) is not available on time, it's relatively trivial to do some custom wiring. But he had heard from Fyodor that the coupling is not just on time, it may be available earlier than expected. Preferably not before 1 July, so that I can get the state battery rebate—maybe.
Roll on tomorrow at 7:30.
|
Bushfire!
|
Topic: Stones Road house, photography, opinion | Link here |
Message from Chris Bahlo (on Facebook, of course) to tell us of two bushfires: one in Mount Clear, and one in Grassy Gully Road. The latter was supposed to be small.
And indeed it must have been. The sorry excuse for an emergency map showed only:
|
|
|||||||||
|
|
|||||||||
No further information whatsoever. Earlier it has at least said that one vehicle had attended the incident. But the map shows that it was about 500 m from hour house. Anything to be seen? Nothing.
Later we went walking with the dogs, and of course looked for any signs of the fire. None. But there was something there in the distance, a couple of hundred metres south of the specified location:
|
||||||||||||||||||||||||
That's taken at focal length 100 mm (comparable to a 200 mm “full frame” focal length). What's there? What can a crop in postprocessing achieve?
|
|
||||||||||
Later we were closer, and got:
|
||||||||||||||||||||||||
Clearly nothing to do with the fire, but interesting because it shows so much more than we could see with the naked eye. Something like a delayed action telescope.
| Wednesday, 17 April 2019 | Dereel | Images for 17 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
World jousting tournament
|
Topic: general, animals | Link here |
Interesting local news on the (state-wide) radio news broadcast: apart from the bushfire in Mount Clear, there was a mention of the World Jousting Tournament [dead link] in Federation Square, Melbourne. Jousting? That's what Chris Bahlo does, with her mates Cliff Marisma and Phillip Leitch. Today's news report gave two names: the current world champion, Phillip Leitch, and his Australian contender Cliff Marisma. It seems that nobody has twigged to the fact that Phil is also Australian (he runs the show at Kryal Castle). And the page above shows a particularly poor photo with the text “Reigning world champion Phillip Leitch (red) and Australian team mate, Cliff Marisma of Australia perform a medieval [sic] solid lance joust...”:
Dead image link: http://www1.pictures.zimbio.com/gi/Medieval+Jousting+Display+Melbourne+CBD+TW9nxvBtaXtl.jpg
Only three years later, these links are dead. There's plenty to be found about jousting on Google, but I no longer know what the contents of the links and images above are, and exceptionally I didn't keep a copy. Possibly this one could be related:
But I identify them more with paintings in Yvonne's office:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
The first painting is of Cliff, the second of Phil (painted a long time ago).
The photo on Federation Square was particularly mediocre, but it's not the only one we saw. Here's Phil again with Andrew McKinnon:
Is that genuine? It's particularly impressive, anyway. Phil on horseback, Andrew on the ground.
|
Solar electricity installation, finally
|
Topic: Stones Road house, general, opinion | Link here |
The electricians from Effective Electrical along first thing in the morning with a surprisingly small pack of 36 solar panels, and set to making a fiendish noise attaching clamps to the roofing and rails to the clamps.
|
||||||||||||||||||||||||
By midday they were ready to mount the panels, which I saw for the first time close-up:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
The wires seemed to be very thin. The label shows that the panels can deliver up to 9.72 A, and there 12 of them connected together. But as Tony told me, they're in series, so that's the total current. But the open circuit voltage of the panels is 40.1 V, so the array has a voltage of 480 V! That's much more than I expected.
And by evening all but two of them had been mounted:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
In the course of the day, Tony and Fyodor came along and brought in the battery and the inverter. It seems that Tony was wrong about the “coupling” just being cables. It seems that there's a lot of electronics involved, and that's why it has taken so long. But they're doing a field test soon, so I asked Fyodor to put me down as a tester. That could still happen.
|
Leo does a runner
|
Topic: animals | Link here |
With all the things going on in the front of the house, let the dogs out to have a look. But somebody had left the gate open, and despite the interesting stuff going on in the driveway, Leonid decided that Grassy Gully Road was more interesting, and set off by himself. By the time I caught up with him he was half way along. I really hadn't expected that.
|
GPS time rollover revisited
|
Topic: technology, opinion | Link here |
Last week I discovered that my “new” GPS navigator (less than 5 years old, anyway) had had a Y2K event and reset its date to August 1999. And that made me think: what would my ancient Garmin GPS2 GPS receiver do?
Went looking, but couldn't find it. But coincidentally, because of the solar panel installation, I had to move the fridges out of the garage and into the shed. OK, take a look in the old cartons that have been there since we moved in... GPS receiver, a digital level and a whole lot of old important CDs that, for some reason, I haven't needed in 4 years.
OK, fire up the GPS 2. It works, though a row on the LCD screen has died. And it gave me the correct time! It must have been through two W1K rollovers. How did it manage?
| Thursday, 18 April 2019 | Dereel | Images for 18 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
What flower?
|
Topic: gardening | Link here |
One of the more rewarding (and long-suffering) flowers in my garden is this one:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I've forgotten who gave it to me, but it must have been well over 10 years ago. I had been told that it was a Peruvian lily, but that seems to be Alstroemeria. And Laurel Gordon told us that it was a Mirabilis jalapa. But I had already noticed discrepancies: Mirabilis cross-breeds, while our yellow plant remains yellow and the red one remains red. And they call it the “four o'clock flower” because it doesn't flower until 16:00. These photos were taken at 7:30. And finally Mirabilis is a bush, and these are tubers. So what are they?
|
Solar electricity installation, day 2
|
Topic: Stones Road house, general | Link here |
The electricians—today 5 of them—arrived at 7:30 again, and got going quickly in the garage:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Fyodor was there again supervising the installation of the equipment. He later took some photos with his phone, and told me that this was of great interest because it was one of the first installations of this size and in this area.
It's surprising how much wiring is needed for these systems. They told me that they had to install a number of oscillators, but I later discovered that they meant isolators—it's clear where I'm coming from (and, interestingly, where my spelling checker comes from: it flags “isolator” as a spelling error). But what a number of isolators! One at the end of each row of panels (the orange ones) and a whole lot more in the garage:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The one on the right of the inverter (second photo) is double-throw Turning it off isolates the inverter and feeds grid power to the house. The ones on the top left are for the three individual PV arrays, and the one underneath is for the batteries. The pipe on the right is part of the vacuum cleaner system, nothing to do with the wiring.
Round 14:00 they rewired the main switchboard to go from main switch (left) to the “Inverter in” switch (second from right), then back from the inverter to the remainder of the switchboard (far right). Hopefully for the last time we needed the UPS to bridge the 10 minutes that that took.
|
|||||||||||||||||||||||
And there are warnings everywhere. Here the switchboard three years ago and today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
They were done round 15:00, and Fyodor showed me how to access the configuration from a web browser. Lots of nice knobs to turn. As discussed, he had set the thing to stop using battery when it dropped to 20%, as long as the grid was connected; otherwise it would continue using it down to 5%, the minimum practicable (and not 0%, as Tomas had claimed). He had also set it to charge from the grid at up to 500 W when it reached 20%. It's not clear that that's a good idea, but hey, I can change it if I don't like it.
Spent some time setting up an account with Ingecon, with some rather interesting details. Clearly you need to enter a device ID and password, and clearly you need to select “Verify connection”. But this photo is after successful verification!
|
|
|||||||||
That's what the little green tick means. How modern!
And then it discovers your location—after a lot of mistaken guesswork:
|
|
|||||||||
|
|
|||||||||
By nightfall the battery was about 60% charged. Spent some time watching the display from the inverter, which looks like it could do with improvement:
|
|||||||||||||||||||||||
This was done the following morning. I think that it purports to show solar panel input on the left (2335 W), consumption in the house (7306 W), power from the grid (4971 W) and power charging the battery (2838 W). How does this add up? And why is the battery being charged from the grid? Fyodor put a 500 W cap on that, and as I discovered during the evening, after dropping to 20%, it charged back up to 30% from the grid. Does that make sense? I need to think about it.
In addition, the somewhat meaningless usage statistics are stated for Saturday and Sunday, without any date. Does the inverter not know the time? It has a configuration entry for NTP, which I set. It should know that the days are Thursday and Friday. And sure enough, it shows the time correctly, in emetic US American date format: 04/18/2019. The only other language that the device does is Spanish, but maybe that's preferable to that horrible date format.
|
Rod for hanging pots
|
Topic: gardening | Link here |
Three months ago CJ Ellis and I tried to mount a rod between the wall and pillar in the house entrance. We failed: the bricks were too hard. So today Yvonne asked the apprentice sparky if he could finish the holes. Yes, in record time, with a bit of advice from his foreman. It was so fast that I didn't have time to get a photo of him.
Was that good luck or good equipment? I had started the first hole in January, and he finished it in a second or two, breaking through to a hole in the brick. Might I also have done that? And was the second one softer, maybe? I had had one like that on the other side. Still, it's done.
|
Failed inverter
|
Topic: photography, technology, opinion | Link here |
We talked a lot about inverters on IRC, of course. And Daniel O'Connor came up with a photo of a failed inverter, taken with his iPhone 6S:
|
||||||||||||||||||||||||
Where's the failed component? With that gradation it's hard to say. Got Daniel to tell me where it was and put it through DxO PhotoLab:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Yes, nicely bent out of shape. It seems that both transistors had been destroyed, and I'd guess that more are as well. But doesn't it make things easier to have well-graded photos?
|
Apple: all your photos is belong to us
|
Topic: technology, general, opinion | Link here |
Getting hold of Daniel O'Connor's photos wasn't as easy as it sounds. In principle if you have an image in a web browser, you should be able to save it. But Daniel's photo is https://www.icloud.com/photos/#04YWjMDeSreIQ-Fi7InYEGESw. Apple. No, you can't just save it, you have to ask nicely by signing in with your Apple ID. OK, I can do that, since I couldn't find a way to delete it.
Not so fast. First we'll send you an SMS to * ******70. Ah, that must be my normal phone that is always on auto-answer. Forget it.
Apple, why do you always appear to me like a shyster after a quick buck? It reminds me of an article I saw earlier: Apple Unveils Plans To Revitalise Notre Dame Precinct With New Flagship Store
|
X screen blanking again
|
Topic: technology, opinion | Link here |
Once again, eureka can't keep quiet. Blank the screen and it's back on again within seconds. What's causing that? I had had various programs under suspicion, but this evening I couldn't be bothered. Switch to X server 1, which doesn't have anything unusual running. But that didn't stop it.
So what is causing this? I can't see any way that a program connected to one server can want to influence another server; there are few enough computers that run more than one X server. More head-scratching.
| Friday, 19 April 2019 | Dereel | Images for 19 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Understanding the solar electricity installation
|
Topic: Stones Road house, technology, opinion | Link here |
So now I have a correctly functioning solar electricity system, and it seems to be doing its job quite well. All I need to do is tweak it a bit.
One thing's clear: the web display provided by the inverter is Just Plain Broken.
|
|||||||||||||||||||||||
After some comparison, it seems that the “PV Generation” and “Grid Consumption” figures add up to the number in the middle with the house symbol. And the “battery charge” figure below is not directly related; to get the house consumption you need to subtract that number (7306 - 2838 = 4468 W). The dates are ridiculously wrong. It's even counting in the wrong direction: yesterday was “Saturday”, but now it's “Sunday”, and today is “Saturday”:
|
|
|||||||||
And there are no useful statistics. How much PV power did I generate today? How much did I feed into the grid? How much did I take out of the grid? Nothing there, just meaningless comparisons to some arbitrary concepts.
“Daily consumption of 1 home”. What kind of home? “0.00 barrels of oil”, saved by not drawing current from a coal-fired power plant. “4.07 kg of CO2 [sic]”. How was that calculated?
Instead I got a report from Spain with marginally more information:
|
|||||||||||||||||||||||
Once again dates in wrong-way-round format, at least part of the time, and pretty useless graphics. But there's a little more information there:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
“Consumption 15.02” (we're too leet to specify units; everybody knows that we mean kWh). What does that mean? The other second explains: “From PV: 15.02”. So that's what we used directly? And then there's “Battery Charge” and “Grid Feed-in”. At first I thought that they matched “From storage” and “From public grid”, but no, I think the first are what we transferred from PV to these destinations, and the second is where the power came from that we used. But that's all still pretty much bare-bones.
But wait, there's more! Go to your web site in Spain (this is clearly too important to keep on your site) and see:
|
|||||||||||||||||||||||
Potentially that's more useful. It's interesting to see the tooth-like pattern during the night where the battery alternately charges and discharges. And round 7:00 the big spike is turning on the air conditioner to warm the place up. The next one is presumably cooking breakfast. I need to analyse this stuff (and find a way to fix the broken text rendering), but it could give insights.
But is it correct? The day before shows a peak usage of somewhere between their chosen grid values of 3.7 and 4.6 kW:
|
|||||||||||||||||||||||
Round 19:00 I noted a consumption of over 7 kW. Who is lying? It certainly could be the inverter, which shows a lot of to-and-froing between exporting to the grid and importing from the grid in a time frame of 5 seconds or so. Is it really doing that, or just reporting it incorrectly? The display on the inverter itself shows the same thing: here I'm running off the PV array and charging the battery (at 7.7 A), but it claims to be feeding power to the grid as well (admittedly only 11 W):
|
|||||||||||||||||||||||
After throwing the main switch it showed a warning light and no further connection to the grid:
|
|||||||||||||||||||||||
About the only thing that's clear is that I didn't get the promised alarm that the system was off the grid.
Still, what I'm seeing here is a total power generation round 31 kWh for today. Given that we're in the winter semester, and that I had expected the equivalent of about 3.5 hours “sunshine” on average for the year (37.8 kWh) that's not too bad. But of course today was sunny; we'll need to factor in the cloudy days.
Still, I want to tweak the system. Where's the configuration? It seems to be hidden where I can't get at it. I have a password for the inverter, but it doesn't seem to apply to the web interface. The documentation talks of a “password” (« contraseña ») 0332, which I would call a PIN. I need to tread carefully here; there's every possibility of locking myself out, and I don't want to do that.
| Saturday, 20 April 2019 | Dereel | Images for 20 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More photovoltaic software investigations
|
Topic: Stones Road house, technology, opinion | Link here |
Yesterday's investigations brought me to two partial conclusions: the web application on the inverter has serious bugs, and I don't have access to the configuration menus. I could continue trying to break in, but it's probably a good idea not to do anything until the people come to inspect it. That should happen this coming week, but considering that Monday and Thursday are public holidays, I'm not betting on it.
So what can I do? Clearly there's a lot of information going out of the inverter. How about firing up a wireshark and seeing what's going on? First stop all the web browser sessions and then start one and watch what happens.
First problem: after stopping them all, there was still traffic. Where was it coming from? I couldn't find out, and in the end I had to run the capture to ignore its port number. And then, sure enough, there was plenty of stuff to see. The most interesting URLs were http://inverter.lemis.com/ems/summary and http://inverter.lemis.com/ems/last7Days. OK, I can get that too. The information looks like:
JSON, clearly. Daniel O'Connor pointed me at a pretty printer:
=== grog@eureka (/dev/pts/25) ~/solar 35 -> python -m json.tool < summary.last
That's interesting, but it doesn't make much sense. Why is the StartDate the way it is? On 11 March 2019 I didn't even know about the Ingeteam offerings. And why is there no time? I've already established that the inverter has NTP and had it set correctly.
And the other values? The identifiers don't make much sense. And it doesn't seem to be updated very often. At first I thought it might be a web cache thing, but that shouldn't happen for local sites, and adding the -n option to fetch didn't change anything.
And last7Days? There's more information there, multiple records, probably the background for this broken display:
|
|
|
|||||||||
Formatted, today's record looks like:
=== grog@eureka (/dev/pts/25) ~/solar 41 -> python -m json.tool < last7Days.last
At least the date is correct (though in a different format from the summary record), so there's really no excuse for the broken bar chart. But it's still not clear what the numbers mean. More head-scratching needed.
| Sunday, 21 April 2019 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Still more PV software investigation
|
Topic: technology, opinion | Link here |
Yesterday's network sniffing gave some insight into how the photovoltaic inverter communicates, but not enough. Spent some more time sniffing with ngrep and found:
Ah! That's the real data, not the locations I found yesterday. And this one does have time stamps that make sense. OK, take a look at http://inverter.lemis.com/ems/sse/stream.data:
=== grog@eureka (/dev/pts/29) ~ 1 -> fetch http://inverter.lemis.com/ems/sse/stream.data
Hmm. What's this stream stuff? Must be something that I don't understand. But there's an alternative, the dirty one that Daniel O'Connor suggested: sniff the data stream and collect what I'm looking for. Here I have clearly identified source and destination IP addresses, source port, and the beginning of the packet data.
So how do I do that? BPF, of course, and more specifically pcap. Is there a program that will do this for me? You'd think that ngrep would do it for me, but it wasn't obvious from the man page.
Further searching brought me to this page, a tutorial with source code. Played around with that for a bit. The first program, ldev (source in the text), compiled out of the box and ran:
=== root@eureka (/dev/pts/4) /home/grog/src/pcap 107 -> ./ldev
Next came testpcap1.c. Again no issues compiling it, but then...
=== root@eureka (/dev/pts/4) /home/grog/src/pcap 108 -> ./testpcap1
OK, what's that? The code is:
Where do I go from there? The best guess from the code is that it doesn't like -1 as a timeout value. But do I really want to get involved in debugging? I just want a filter. Maybe I should learn about streaming after all.
|
Irrigation woes again
|
Topic: gardening | Link here |
Yvonne told me what I should have known today: time to clean the bore water filter. I had decided that it needed cleaning every week rather than every few months, probably as a result of the low water level due to the dry summer. But I had neglected it for a couple of weeks, and it was very much in need of it.
That wasn't the only issue. It's been weeks (months?) since I decided I need to look at circuit 4, which has only too clearly been leaking:
|
|||||||||||||||||||||||
Somehow I'm just too lazy.
|
Solar electricity progress
|
Topic: Stones Road house, technology, opinion | Link here |
We've been planning the solar electricity for well over 4 years, and much of the delay was due to procrastination on my part, coupled with a complete inability to establish contact with Simon Reid of BREAZE. Chris Bahlo didn't have that luxury: she has no grid connection, so she had to have it installed for her new house (also four years ago).
Today she came along and we compared notes. She paid about the same amount of money as we did. For that she got 7 (count them, seven) solar panels and a lead-acid battery of unspecified dimensions. But since she's only generating 1.75 kW, and the average hours of sunshine here are round 3.5 hours, it's unlikely that the batteries would store more than 7.5 kWh. And of course her inverter is completely dumb.
That's quite a difference, though even at the time I thought that the price was excessive. Maybe it wasn't such a bad thing after all that I didn't go with BREAZE.
|
Das rechte Osterlamm
|
Topic: food and drink, language, music, opinion | Link here |
Easter Day today, traditional time for eating lamb. And we have had a boned leg of lamb in the deep freezer since last Easter (Why? We ate lamb then too. Did we buy two of them?).
As mentioned at the time, the thing was so big that I cut it in two.
OK, thaw it out and roast it.
Not a success. I've been trying to work out roasting times for meat, including lamb, for some time. But one thing I really must do is ensure that the start temperature is the same each time. It should be room temperature (say 22°). If you take it straight out of the fridge (say 4°), like I did today, and you want to cook to 60°, you have a temperature difference of 56° instead of 38°. That's nearly 50% more, and it makes a big difference in the result. In fact I took it out when the thermometer was showing 52°, and it was still marginally overdone.
Of course, tradition has a lot to do with the timing. I'm reminded of Martin Luther's hymn „Christ lag in Todesbanden“, with the modifications in BWV 4, which renders verse 5:
Hier ist das rechte Osterlamm,Davon Gott hat geboten,Das ist hoch an des Kreuzes StammIn heißer Lieb gebraten,
There are various translations of this text, the most literal of which (mine) is:
Here is the right Easter lambFrom which God has ordainedThat is high on the cross's stemFried in hot love
And somehow that, too, belongs to Easter.
| Monday, 22 April 2019 | Dereel | Images for 22 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Cracking the inverter data
|
Topic: technology, Stones Road house | Link here |
Yesterday I investigated sniffing the traffic between the inverter and a web browser, in the assumption that it would be easier than recreating the requests needed for communicating directly with the inverter. But Jamie Fraser and Daniel O'Connor (on IRC) saw it differently. Jamie suggested using the developer tools (F12 on most browsers, it seems). I've seen them before, but never used them in earnest, especially not to analyse HTTP streams. OK, try it out:
|
|||||||||||||||||||||||
|
|
|||||||||
That's not immediately obvious, but none of the traffic I'm looking for appears there. Ah, says Daniel, works on Chrome and Safari. OK, find a Chrome and try again. Great improvement. Now I can see the data coming in, and as expected without any GET requests:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Next, click on the left hand (name) column in the network tab. What column? I don't have no steenking columns. But Daniel did:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Ah, the console is hiding it all. Remove that and we have:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That looks a lot better. In particular, the progress bar at the right shows that it's continuous. And sure enough, select it as a curl command, and we're away!
=== grog@eureka (/dev/pts/30) ~ 11 -> curl 'http://inverter.lemis.com/system/events/sse/stream' -H 'Accept: text/event-stream' -H 'Referer: http://inverter.lemis.com/dashboard/main.html' -H 'Cookie: __utma=51177012.766199467.1486957619.1542435223.1545448630.4; __utmc=51177012; __utmz=51177012.1542145939.2.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); adblk=adblk_no' -H 'Connection: keep-alive' -H 'Cache-Control: no-cache' --compressed
Wonderful!
What next? Pipe the data into a database, of course. Daniel has been there before as well, and he has an application, written in python, which collects the data, puts it in a database and even generates graphics. Took a look at it and was discouraged: it's relatively long, and my Python-fu is not really up to it. There must be a simpler, if kludgier, way.
|
Autumn lily
|
Topic: gardening | Link here |
Walking the dogs today, found this flower:
|
||||||||||||||||||||||||
I'm sure I've seen this one before, but it's worth identifying: it's a time of year when not much is flowering. Maybe I should find a few autumn-flowering bulbs.
| Tuesday, 23 April 2019 | Dereel | Images for 23 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Australian building standards
|
Topic: Stones Road house, opinion | Link here |
Coming out of the bedroom this morning, caught my sleeve on the door catch, tearing it off the door frame:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's not the first time. Last time I tore my shirt apart. The other side just sticks out into the room, asking to catch on something:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The real surprise is that it doesn't happen more often.
Why are Australian doors of such poor quality? They don't shut properly, the joiners install them with a 3 cm gap at the bottom, and they have these positively dangerous fittings, though I suspect that JG King have chosen a particularly poor example.
|
More solar power software insights
|
Topic: technology, Stones Road house, opinion | Link here |
More playing around with the solar power data stream today, and managed a partial conversion to an SQL INSERT statement, complicated by my lack of recollection of the syntax. But I'm getting there. The question remains whether I should do a proper parse of the JSON data, but I think I can get by without it. Other questions are more interesting: what do I do with the data? There's one record like this every second:
All those numbers appear to fit into 16 bits, but there are quite a few of them. 86,400 of those a day? I have a comparison with my weather database, which has been collecting one record a minute for the last 9 years. I now have 3,800,000 records, and the observations table is 280 MB in size. I can live with that, but with one record a second I would reach that size in a couple of months. For the time being I suppose I should just collect them, but probably I'm going to have to remove records older than a week or so.
And what do the fields mean? I think I have cracked most of them. These ones appear in the once-per-second records:
This interpretation is only partially correct. Don't use it.
| Identifier | Meaning | Units | Comment | |||
| Id | ? | IP address? | ||||
| Pac | "Power AC" | W | Current house consumption | |||
| SetPoint | ? | numeric | Always 0 | |||
| Alarms | String | |||||
| Status | Inverter status | String | Only ever seen "On-grid" | |||
| SOC | "State of Charge", battery | % | ||||
| VBat | Battery voltage | V | ||||
| PacGrid | Power from grid | W | Does not show power to grid | |||
| PacBat | Power from battery | W | - means charging | |||
| PacPV | Power from PV array | W | ||||
| FromPV | ? | Boolean | Always false, even when drawing power from PV | |||
| Codes | ? | Never seen any set | ||||
| W1 | Power to grid | W | One of three (possible) phases | |||
| W2 | Power to grid | W | One of three (possible) phases, always 0 | |||
| W3 | Power to grid | W | One of three (possible) phases, always 0 | |||
And these appear about once a minute:
| Identifier | Meaning | Units | Comment | |||
| Id | ? | IP address? | As before | |||
| Date | Date and time | String | Format "2019-04-22 13:57:00" | |||
| Phase | Phase number on grid | numeric | Always 3 of these, even on a single phase connect | |||
| W | Grid power | W | Negative → grid | |||
| Pac | "Power AC" | W | PV power input, not current house consumption! | |||
| PacDischarge | ? | ? | Always 0 | |||
| PacCharge | ? | numeric | Digit between 0 and 3 | |||
| PacPVBatt | ? | numeric | Appears to be close to Pac (power from PV array) | |||
| EMS_VBAT | Battery voltage | V | VBat in other record | |||
| EMS_SOC | State of charge | % | SOC in other record | |||
| WatDigitalInput | ? | numeric | ||||
| PacRev | ? | numeric | ||||
| HWOutputs | ? | numeric | ||||
Clearly there are a number of contradictions here. Different programmers, or inadequate planning? And does the thing really always output for three phases, or is this a matter of the inverter configuration?
|
Getting the most from solar power
|
Topic: general, technology, opinion | Link here |
I'm quite happy with the amount of power we're saving even at this time of year; the daily reports in triplicate from Spain show that I'm generating between 53% and 66% of my total usage. But it's worth thinking about how to ideally use the power. The algorithm used by the inverter is:
And then there's the other view:
All this appears to make perfect sense. But already there's a first problem: what if the battery is charged and the grid is being supplied at full rate, and there's still a reserve of energy? It gets lost. And from my observations, that happens quite a bit of the time. The second battery, when it comes, will reduce that time, but not eliminate it. So clearly there are things that should be done when the sun is shining, like running dishwashers and washing machines.
And then there's another situation that is even more difficult to understand: should I charge the battery first and only then feed to the grid? Today the battery started charging round 8:00 and was done by 12:00. Then I fed to the grid for a total of about 3 hours, in the process running into the limits. If I had started feeding in to the grid earlier, I would have been able to feed in more, possibly double the quantity, and still charge the battery.
The problem is: I know that now. How could I have known it this morning?
|
Air-fried Bratwurst again
|
Topic: food and drink, opinion | Link here |
Another attempt at “air-fried” Bratwurst today. Based on previous experience, decided to “fry” them hotter and longer, 210° and 15 minutes instead of the 200° for 12 minutes last time. The higher temperature seems to have been a good idea, but it seems that the original time was sufficient. Here before and after:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That seems to be a reasonable time from now on. That's a left-over Wiener Schnitzel on the left. Despite the appearance, it tasted very good.
| Wednesday, 24 April 2019 | Dereel | Images for 24 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Solar energy: enough!
|
Topic: technology, Stones Road house | Link here |
I've spent the best part of a week thinking about how to process data from my solar inverter, and I think I'm on top of how to get it into a database. But that's only part of the story. What do I do with the data in the database? Do I want to create another page like the weather page? In principle the pretty displays that come with the inverter do most of that. So maybe it's time to step back and do something else.
One thing that strikes me: what is using all the power? The daily reports (the ones that come in triplicate) confirm what Powercor keeps telling me: we're currently using about 38 kWh a day. That's over 1.5 kW average. How? At first I thought that it was the computers and displays, but I've seen values as low as 500 W with the big TV display going, and I estimate that to use 200 W by itself. The fridges? The newest freezer claims to use 616 kWh per year, which corresponds to about 1.7 kWh per day. Multiply that by 4 (two freezers, one fridge, one combination) and we're still only at 6.8 kWh per day. What else? The air conditioner is an obvious culprit, but I see this consumption when it isn't running.
I have some power meters that plug into the power points, but they're inconvenient: to read them I have to kneel on the floor and press buttons, far too much trouble. Isn't there anything that will connect with a network?
Yes, of course. eBay is full of offers for “10XSonoff Power 10A Smart Home Automation Measuring Monitor Wifi Switch ControYO” from unscrupulous sellers like share-happy, who offer them for $3.20. And they don't seem to be available elsewhere. $3.20? Only joking, for that you get an only marginally related “Board DIY Kit”. The item itself costs $11.45, or $121.42 for 10. It's available much more cheaply from other sellers. High time that eBay put paid to this kind of deception.
Still, what can the thing do? The most important thing in the description is that they can be controlled (only!) with a mobile phone. Mobile phone! Dammit, what happened to interoperability? Discussed it on IRC, and Daniel O'Connor pointed me at the DLink DSP-W215, which proved to be much more expensive, and some hacking instructions, which proved to be for the Sonoff (and not 10XSonoff) switch. As I feared, it notes:
The problem is, that this nice device is only cloud-dependent. Meaning, you cannot run it without internet connection on your local network.
Dammit, what happened to interoperability? Once upon a time we were on course to have a nicely integrated world network. Now marketeers and short-sighted product managers have built a Tower of Babel!
Still, the article goes on to describe how to connect to it anyway. I suppose I should get hold of one and play around with it.
| Thursday, 25 April 2019 | Dereel | Images for 25 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Shepherd's pie: the pain
|
Topic: food and drink, opinion | Link here |
On Sunday (Easter Day) we ate a boned leg of lamb, not as good as we have had it. Today we had the rest, and traditionally we make shepherd's pie out of it.
OK, that's straightforward, right? I can do it after the bread I baked.
No. I must finally realize that shepherd's pie is more work than a roast leg of lamb. And once again, working with Yvonne, I ran into problems:
It requires mincing the meat, of course, along with a significant quantity of bread (English-style soggy stuff). Today I forgot the bread and had to put it in later; fortunately it crumbles easily.
How many herbs? I chickened out and used dried herbs, because I was in a hurry.
How big a dish? I need to relate that to weight. Based on today's experience, I'd say between 1.2 and 1.5 g per cm² of dish.
How much mashed potato? That's a double problem. The easy part is: what weight? Based on today's experience, I'd say 1.6 times as much as the meat. But that's quite a guess:
How do you make the mashed potato? Simple:
|
|||||||||||||||||||||||
Bloody cups again! Why can't they give you a weight? But OK, we can calculate. The whole package weighs 350 g, and that's enough for 8 serves. So with a bit of calculation we can convert that table:
| Serves | “DEB” | water/milk | total weight | Proportion | Proportion | |||||
| - | - | DEB:liquid | total weight | |||||||
| 1 | 43.75 g | 170 ml | 190 g | 25.7% | 88.9% | |||||
| 2 | 87.5 g | 310 ml | 380 g | 28.2% | 95.6% | |||||
| 4 | 175 g | 625 ml | 740 g | 28.0% | 92.5% | |||||
| 8 | 350 g | 1500 ml | 1500 g | 23.3% | 81% |
Do these numbers make sense? Barely. I get the impression that whoever created this table didn't know what he was doing. Maybe he was confused by the strange units.
But it gets worse. Yvonne mixed "4 serves" according to the instructions and came up with something that was far too thick. She had to add another estimated 300 ml of water to make something useful. So at a guess that was 1 kg of mashed potato, and still not really enough. There were 700 g of meat, so the 1.6 times as much sounds right for the next try. Last time I wrote twice as much, but I don't think that is necessary.
How much gravy? Last time I wrote: as much by weight as the meat. That's what I did this time, and it was too much. Maybe 80%?
How simple a leg of lamb is by comparison: put in the oven, cook until 60°, and you're done!
| Friday, 26 April 2019 | Dereel | Images for 26 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
A week of solar electricity
|
Topic: Stones Road house, general | Link here |
We've now had the solar electricity system up and running for a week, time for a first summary.
Wouldn't it be nice if there was some way of getting this from the various software supplied with the hardware? But no, all I get are daily reports in PDF form, each in triplicate, along with bar charts of weekly use:
|
|||||||||||||||||||||||
Powercor also shows daily usage, though they know that a week starts on Monday, and nothing will change that:
|
|||||||||||||||||||||||
From last Friday until yesterday things look good, power usage down by round 50%. But the blue line shows the “suburb” average, still less than we use with solar help. And yesterday was not as good, and today was close to the same as before the installation. What went wrong there?
One thing, two results: the weather has changed. Today's temperatures ranged from 5.5° to 13.5°, with an average of 10.1°, and there was little sunshine. So we had little solar electricity (only 20.18 kWh), and needed more to heat the house. The big issue was the heating; yesterday we also only had 22.18 kWh, but total consumption was 47.21 kWh instead of today's 59.53 kWh. That shows in the black line (battery charge level):
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The maximum was only 69%, compared to 100% on all other days. The usage information is misleading; the graphs have been scaled to match usage. The first graph goes to 6.8 kW, the second to 8.2 kW.
It'll be interesting to see how things progress. Maybe last week was unusually mild and bright, and that today's results are closer to what it's going to be like through the winter. Still, it's interesting to note that even today we got more than a third of our power from solar energy.
Apart from that, I'm still observing the data coming from the inverter. My guesses a few days ago appear not to have been completely accurate. My guess now is that the meaning changes depending on the relationship between the values, so I'm collecting various screen shots and corresponding data. The screen shots are, as ever, on my daily photos page.
|
Fake Boursin
|
Topic: food and drink, opinion | Link here |
We used to eat a fair amount of Boursin cheese, a soft fresh cheese with herbs. Today Yvonne tried her own version based on Philadelphia cheese:
|
|||||||||||||||||||||||
It certainly looks better, but of course it tastes different. Certainly a useful addition to our cuisine.
|
STF-8!
|
Topic: photography, opinion | Link here |
I've been keeping the Olympus STF-8 flash unit on my old OM-D E-M1 (Mark I) lately: it's quite useful, but it's a pain to put on the camera, so effectively the E-M1 is just for it, at least for the moment. And so it was trivial to pick it up and take a photo of the cheese this evening:
|
|||||||||||||||||||||||
But what's that shadow on the right? This is a dual flash from both sides. Did I have the intensity set wrong? No:
|
|||||||||||||||||||||||
That wasn't deliberate. The diffusers fold forward very easily. There are so many things to take into account with this unit!
| Saturday, 27 April 2019 | Dereel | Images for 27 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Hugin pain
|
Topic: technology, photography | Link here |
House photo day again today, and once again I ran into great difficulties getting Hugin to join the house-s panorama:
Why? I can't see any obvious issue with that view, but I'm continually having difficulty with it.
And to make my day, got a problem report: Hugin gets a SIGSEGV and crashes on start. That's about all he said. Clearly I need more information, but what? He shows a whole list of error messages, but for the most part this is just hugin whingeing when it thinks nobody is looking. It tails off like this:
So far, so good; this is exactly what I get too. But then for me GTK carries on whingeing:
And the bug report states:
What can that mean? All GTK's fault, of course, and I know nothing about GTK. A quick web search showed nothing enlightening, and it's not even clear that the crash is related to the message, given the amount of grumbling it does. We have a core dump. Would it help to take a look at it? Potentially.
And what else? Poke around in the dark?
|
Muscle pain improvement?
|
Topic: health, opinion | Link here |
It's been over 3 weeks since Heather told me about massage balls and things, and they seemed to work well. Initially. But things still aren't much better, though they're changing. Gradually the pain is working its way down my leg, not quite what I would have expected from issues with the gluteal muscles.
Still, sleeping with a ball against the muscles seems to be improving things. The only issue is the shape. I call the ball a blowfish, for what seem to be obvious reasons:
|
|||||||||||||||||||||||
But in fact it appears that this is a porcupine fish. I wonder if Theo knows.
| Sunday, 28 April 2019 | Dereel | Images for 28 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More solar power insights
|
Topic: Stones Road house, technology, opinion | Link here |
The weather continues to be cool and dull, and it shows in the solar power generation and consumption. It wasn't as bad as yesterday, and in fact 12.66 kWh don't look that bad, but it's still only 38% of the total need. And the more I look at it, the more this “charge battery from grid” doesn't make sense. I currently have the system configured to discharge the battery to 20% while on the grid, and down to 5% when off the grid. But when I reach 20%, it's set to charge back to 30% from the grid. That makes particularly little sense early on a sunny morning. What I need is to charge back to 20% when the charge is below that level, like on power restoration after a failure.
And how do I detect the failure? I've been looking through all the data that I have collected over the last few days and noted that a lot of the fields never change. What's “Alarms”? It always comes across as 0. But that, it seems, is because I defined it as a smallint; it should be a string. Something to fix before the next experiment.
And what other Status can I have than “On-grid”? Is Fyodor right that I will be warned of a grid power failure? Time to test it.
Into the garage and threw the main switch. It took the inverter about 2 seconds to notice and put on a warning light. Left it for a while, then turned it on again. It didn't recover!
Aaah, yes, it did; it just took a long time. Back to see what it had told the world. Here a limited view of the table, since the other columns didn't have any useful information:
There are a surprising number of things to note here:
Power off at 14:33:36. The inverter goes into a “waiting to connect” state for 3 seconds before deciding that it's off the grid. That's the delay I saw with the alarm LED.
It's not clear what happened at 15:33:37. It seems that up to that point the battery had been supplying 540 odd watts, but then went into a "charge" status. What happened to the house power (roughly Pac)? That doesn't make sense, but maybe some chance change in load happened to confuse me. A few seconds before the load went up from 981 W to 1317 W.
After another 15 seconds it starts generating error codes. I have no idea where I will be able to find them out. It's also possible that the codes relate to the restoration of grid power, which it doesn't acknowledge until 2 seconds later.
It remains in “Waiting to connect” status for 1 minute, maybe to ensure that the grid has settled; potentially there could be unwelcome surges during that time.
At no point was any alarm raised. I also didn't get any message about the outage. So I'll need a program to warn me. That's of particular concern if the power fails in the middle of the night; as it is currently, the air conditioner could drain the battery in a matter of minutes.
I had guessed char (20) for the string fields. Status, at any rate, seems to be longer, and potentially Codes is too. More changes to be done.
|
Drought!
|
Topic: general, opinion | Link here |
We've been living in Stones Road for nearly 4 years, which means we must have walked the dogs to the “schoolyard” on the corner of Bliss Road about 700 times. They have a pond with waterlilies, and it never dries up.
Well, not quite. Today it looks as low as I have ever seen it:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
But then, we have had really little rain, rather less than 0.5 mm per day this year:
| Monday, 29 April 2019 | Dereel | Images for 29 April 2019 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Insurance claims: the pain
|
Topic: general, technology, opinion | Link here |
I'm finally getting round to making an insurance report for the accident I caused on 5 March. Partly the delay was due to the other side not reporting it until recently, and partly because the email was lacking a Subject: line. And, of course, my usual procrastination.
OK, off to the RACV web site, which offered online reporting. Sort of:
|
|
|||||||||
“RACV Online is currently undergoing maintenance. Please try again later.”
Dammit. Let's talk on the phone. Conveniently they have a real (not “1300”) phone number to call, 9790 2211. Called that and got put through to a voice menu with all of 6 or 7 different choices. The first was the one I wanted, but it didn't listen to my key press. I had to go through the entire list. And still it didn't listen!
Dammit. Let's ask for a direct connection. That worked marginally better—I still had to make an input, but at least it worked—but when I got to a real live person, she couldn't hear me.
What's wrong with these people? I'll be in town tomorrow, so I'll try with a live person in front of me.
|
Tracing stray dogs
|
Topic: animals | Link here |
Chris Bahlo over today with a horse to go riding. First, though, she needed Yvonne's help to put on her armour. I suppose it's a sign of the times that I didn't even bother to take a photo of her in her armour.
But she came back with more than she had when she left, a dog:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Took some photos of him to be able to report his presence.
|
Facebook beats conventional photos
|
Topic: technology, photography, opinion | Link here |
Went back inside and finished watching the TV news before processing the photos I had taken of the dog. Too late. Chris, outside and still in armour, had braved the glass keyboard of her mobile phone and published a photo of him on Facebook:
She even got some kind of reply. But that reply seemed rather like my abortive attempts to contact RACV: no two-way communication. No response to the phone (just Facebook; what kind of world are we living in?), so Yvonne guessed that he lived up at the vineyard at the other end of Stones Road, and took him up there.
| Tuesday, 30 April 2019 | Dereel → Ballarat → Dereel | Images for 30 April 2019 |
| Top of page | ||
| previous day |
|
Pay now, get later
|
Topic: Stones Road house, opinion | Link here |
The terms of my quotation from Effective Electrical were interesting: pay $2,000 deposit (already done), and the balance a week after “Commission”. No mention of the fact that they planned a partial installation this month (done) and planned to do the rest in October. Clearly commission can only occur when they have fulfilled the entire contract.
So I was wondering what they would do about getting paid. They had sent me two different quotes, one for a single battery and one for two batteries. I accepted the quote for two batteries, but what they have installed so far corresponds exactly to the “single battery” quote. An obvious solution would be to ask the money that they had quoted for the single battery.
But no, today I received a bill for the full sum. Sorry, can't do that, don't owe you anything. But it seems unfair to stick to an obviously botched quote. On the other hand, I don't want to pay the full price for one battery: then it would be easy for them to put the second battery into the “too hard” basket and call it quits. So I offered $1,500 less as a voluntary progress payment, leaving a bit of space for negotiation. No, no worries; within 30 minutes I had a revised invoice for the price I offered. Only fly in the ointment was that it also contained the text:
Inspection and Compliance will be billed when completed, as will the second Battery
No information about how high that billing will be. But probably it's another case of Hanlon's razor.
|
More physiotherapy
|
Topic: health, opinion | Link here |
To Ballarat again today for another physiotherapy appointment. Spent a lot of time talking about the exercises I need to do, what we're trying to achieve, and why she's not so interested in how the pain seems to happen at night (it seems that that's only because I'm not moving, and that the cause of the pain happens during the day). And then the time was up; no massage.
|
Insurance claims, try 2
|
Topic: general, opinion | Link here |
As planned yesterday, into the RACV shop in the middle of town after the physiotherapy appointment. I hadn't expected to find a queue:
|
||||||||||||||||||||||||
It took me something like 8 minutes before I could talk to anybody, and then she told me that they don't do insurance claims in the shop. She did, however, confirm that they've been having web server issues for the last 2 weeks, and that they're ongoing. In that connection I'm glad to see that they're no longer running FreeBSD.
She took me to a phone and got me connected to Craig pretty quickly. Craig took details for a “notification claim” and suggested that I first get two written quotes for the repair, and that they could get more deeply involved if necessary. That includes paying the excess ($700!) up front and getting it back if they don't need to pay.
Damn, these little scratches are more pain than they're worth!
|
Parking fines
|
Topic: general | Link here |
One of the nice things about Ballarat is that you can usually find a free parking space. But the RACV is pretty much slap bang in the middle of town, and I had to buy a ticket with my remaining coins. To my surprise it was only enough for 15 minutes. And I spent a long time there; by the time I got out, the ticket had been expired for about 10 minutes.
And then I saw somebody looking at a car, what was under the windscreen wipers, taking photos. Poor bloke—had he really been issued a parking fine? I asked him, and he looked confused. No, he was issuing parking fines.
Somehow I managed to make it out of there before he got to my car.
|
Insecticides
|
Topic: gardening | Link here |
I have to admit that treating the mites on my curry tree with Pyrethrum has not been successful. On the way home, dropped in at Formosa in Leith Street and was given some very expensive insecticide, Natrasoap, which on closer examination appears to be conventional soap (“Potassium salts of fatty acids”; my recollection is that soap is effective Sodium stearate). We'll see how we get on with that.
Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
