Good news! It seems my Raspberry Pi isn’t dead after all, though the problem is a little odd.
I ordered some new SD cards so I had some spares, and out of interest put a new copy of Raspbian on one of them, put it in the Pi and it booted happily first time. Yet it wouldn’t boot from the old SD card that was fine in another Pi! I’m not going to complain, but I can’t claim to understand why…
My first ever Raspberry Pi has died 🙁
It’s been out with the weather sensors in my Stevenson screen for at least a year, but we had a power outage a short while ago and it wouldn’t reboot. I brought it back inside to check over and found that it was reporting faults on the SD card when trying to boot. I reformatted the card and reinstalled the OS, but it still wouldn’t boot, yet swapping the card into a different Pi gave no errors. A different card wouldn’t read at all.
I know there can sometimes be problems with the SD card socket, but actually it appears completely undamaged and the contacts look clean. I’m wondering if it is perhaps the SD card interface that is becoming unreliable.
For the time being I’ve replaced the unit with another and if it’s convenient at some point I might give replacing the socket a go just in case, but it must be about ten years old now, is slow compared with newer models and doesn’t really owe me anything, so I might just give up on it.
I was a bit pushed for time today so I had to sacrifice some of my 25m reps, but that perhaps wasn’t such a bad thing as I was struggling with them a bit after managing to do all my 50m reps under 43 seconds again 😀 In the end I think I only managed ten twenty-fives before I had to get out.
Very pleased with the performance on the 50m set though, so next time I shall drop the interval again and see how much pain that causes 🙂
I had a couple of days off last week thanks to not feeling great and missing quite a bit of sleep, so today was my first swim for six days. I was expecting it to be fairly tough, but actually it went quite well.
My 50m reps were all below 43s except for two and the 25m reps were acceptable as well. I’m still looking for another set where the fifties are all below 43s though. Hopefully that will come soon.
This is a photo of some of this winter’s purple sprouting broccoli plants taken on 26th November.
They’re starting to flower!
The plants weren’t sown or planted out particularly early. In fact possibly later than the instructions on the packet suggest. But because of the strange way our weather is now behaving, they’ve started producing flower spikes several months earlier than we’d normally get them appearing.
Now the weather actually does appear to be turning colder for the winter having spent most of November above 10°C, perhaps they’ll calm down a bit and wait until the new year before making any more.
Forgot to post this a couple of days ago…
As I’ve already written, I’ve been getting ahead on compost for next winter, filling one of the bins with shredded material from some tidying up of the garden. There’s no grass in this. It’s mostly hedge prunings. Certainly nothing I’d have been too keen to put in unless it went through the shredder first. Three days after I added the last material I put the compost thermometer in to find out how it was going. And it certainly was going!
I’m surprised yet pleased to see that the temperature can get so high without the need for stuff like grass.
We’ve still not pressed the apples (this coming weekend, hopefully) and the spent pomace will definitely be going in on top. After that I’ll be moving on to the next bin. This time last year I think we’d only just started filling this bin and I think it took until some time in May to completely fill it, so we’re way ahead there. I may need to make another compost bin though 🙂
Well, I hope so, at least.
A new reed switch arrived and has duly replaced the old one. That turned out not to be too tricky. After unsoldering the original switch I bent the leads so the body of the switch would be as close as possible to the wall of the compartment for the electronics and soldered it back in.
I’d also bought a holder for two AAA batteries. There’s plenty of room for it inside the battery compartment. In the long run I might actually move it elsewhere, but it will do for now. I unsoldered the “leads” used to hold the original 3V battery in place and replaced them with the leads for the holder making sure to maintain the same polarity. An equally easy job.
Finally, because I wasn’t really sure if the original magnet was up to snuff any more, I ordered some round neodymium magnets, 2mm diameter by 1mm thick. I was originally thinking that I might have to fix them to the opposite side of the tipping buckets from the original magnet and turn the buckets around. However, it then occurred to me that there was room to put some on the end of the existing magnet, so I decided to give that a go:
A quick test suggested that would work, so I put a couple of batteries in the holder, reassembled everything to the point where I could test it and plugged it in. Initial test suggest the sensing of the bucket tipping works fine. I can actually even here the reed switch actuating. Unplugging and the unit and reconnecting it several minutes later also shows the counters maintaining their values when there’s no external power, so that’s looking good too.
Tomorrow I shall finish reassembly and then check the bucket sizes by dripping water in with a syringe. Once that’s done I think it can go back up 🙂
I ran out of time yesterday so wasn’t able to swim and in fact I felt exceptionally tired too. I can’t claim that I felt that much better today, but off I went anyhow. It’s been the first really cold day of the winter and usually the pool is quite chilly when that happens, but for the first time in quite a few years I was pleasantly surprised to find the water no colder than normal.
The 50m reps were quite tough and I struggled to keep them below 43s, ending up with four or five over. I’m not very happy about that. Hopefully I can get it back under control when I swim later in the week.
The 25m reps weren’t so bad though. Not quite as fast as I’d like to have seen them, but easily within half a second. Given that I wasn’t really feeling completely on it I think I can take that for today.
I’m not going to claim this is absolutely spot on as I’ve reverse-engineered it from the PCB traces which aren’t always easy to make out, but below is my best current idea of the circuit diagram for the PCB inside the AAG rain gauge. The board is labelled TAI-8575 and I assume that’s a part number for the entire thing, but it’s mostly based around the now-defunct Dallas DS2423 1-wire counter chip.
The terminals for the 1-wire bus are on the left. The terminals on the right are for an additional switch that might be used for testing.
I don’t understand the numbering of the resistors: there only appear to be three, for a start, and I can’t see the screen-printed label for the 1M resistor. U2 is, I believe, a three-pin surface-mount device that contains two Schottky diodes with a common cathode. I’d assume it’s there to protect the two potential power sources from reverse voltages.
It’s also odd that the Dallas rain gauge which may actually be what this design is based on (or vice-versa) doesn’t use the resistor bridge or transistor. Effectively the reed switch is just across Vbat and the two inputs on pins 5 and 6.
I’ve now got the rain gauge down and taken it apart. Possibly I’ve posted photos before, but this is what the guts of it look like:
After ungoo-ing the PCB, I’ve come up with a first approximation of the switching part of the circuit which I have drawn, in true engineering fashion, on the back of an envelope.
The reed switch fell apart as soon as I touched it, so that’s definitely one problem to sort. The fact that there’s a dead battery welded to the PCB isn’t that helpful either. I may clip that off and replace it with something more maintainable. There are perhaps some other things that could do with tidying up too. I’ll ponder on those. Meanwhile, as I have most of the circuit worked out (I think), perhaps I should finish the job and draw it properly.
