Crosspost from: https://slrpnk.net/post/25006653

So my Siemens Fridge has been acting weird lately. The freezer under the fridge has randomly started to cool to -30°C. The temperature is set to -18°C and the freezer actually gets back to that after reaching -30 for a while and then warming back up again. It sounds like the compressor is working almost the entire day and the temperature at the top seems to almost always display the correct temperature. Ive yet to check wether it actually reaches -30, but my vegetables in the fridge have started to freeze 🙃 Ive noticed that manually rebooting the fridge helps to get it started and I plan on doing it automated until I can fix it myself. Siemens wants 120€ just for a technician to come look the the problem and because I am a cheap college student, I'd really like to fix it myself. Is there any steps I can take to further locate the problem or should I let a technician come look at my fridge?

Its a Siemens KG25FA40/01 fridge and freezer.

Thanks in advance :)

Repair cafes are great for getting appliances fixed. But I think they could improve from volunteers just doing the work to a more instructional format. It’s great that you can watch them work and ask questions. It’s almost a teachng experience in that regard, but teaching is merely incidental. The repairer probes around with the DMM¹ quickly because they want to get answers quickly. Understandably so, as there is usually a line of people waiting to get stuff repaired. They don’t generally have time to explain everything.

But what if instruction were part of the goal? I would like to hang out in the workshop and watch other repair jobs and ask questions. But I get the feeling I would be in their way and slow them down. It feels like it would be unwelcome.

In principle, they could have one repairer who welcomes an audience where he describes his every move. He works slower, but ~5 or so people could learn from it. It could even be recorded and posted on peertube (not Youtube!).

Repair cafes do not accept large appliances because they are working out of classrooms and community centers on weekends, which don’t accommodate bulky things. So I have a broken refrigerator and washing machine that will not get repaired. In principle, a repairer could have a planned session and meet “students” outside to demonstrate and teach large appliance repair.

¹ digital multimeter

I need to connect a PC to my washing machine. The washing machine has a bank of pins labelled like this:

• 0v
• tx
• rx
• 5v

The microcontroller is an ATmega32L, which has specs for the serial connection as follows:

My USB to TTL serial adapter is apparently based on a ch340 chip. It looks almost exactly like the pic I attached, except mine does not have a crystal on it because I think the chip has an embedded clock. The important thing is the pins match my adapter.

My knee-jerk thought was to connect it as follows:

adapter → washing machine PCB

gnd → 0v
rxd → tx
txd → rx
3v3 → (nothing)
5v → 5v ← bad idea?

(with s1 jumped to 5v on the adapter)

Someone told me I should not connect 5v to 5v. I was assuming one connection needed 5v and the other supplied it, but I was told they are both supplying 5v, but not perfectly 5v, so the difference will strain something and cause damage.

So how should I hook this up?

update (I’m stuffed?)

I heard washing machine manufacturers often sabotage the serial ports before shipping as an anti-repair tactic. I thought my old machine might pre-date that practice, but I might be wrong. I metered TX voltage against 0v using a crappy cheap DMM. Results:

0.01, 0.00, 0.01, 0.00, 0.01, 0.00, 0.01, 0.00, … etc, every second or so.

Looked encouraging, as if there is activity. Then I metered 0v against 5v:

~~0.01, 0.00, 0.01, 0.00, 0.01, 0.00, 0.01, 0.00, … etc, every second or so.

Yikes. I was expecting that to read a steady 5v. Due to bad wiring in the house, I think that is just noise on the ground wire. And apparently the serial port is dead.

I had the two 5v lines connected to each other for a while, so it’s possible I damaged it, if not the manufacturer.~~

Update 2: I have 5v, so the port may work

It turns out I had a bad alligator clip, so of course I got a flat reading. The 5v pin on the serial port gives 5v. Thus I might have a live port. Now how do I use it?

Unfortunately I have excessive battery drain in the left earbud of my Sony WF XM4 set. I really like these earbuds and I am going to at least try to fix them. Has anyone here done any thing like this before?

It makes sense, honestly, that the left runs down faster than the right, I use just the left at work all day everyday and only use the right at home maybe once a week. The left lasts 2.5 hours, the right lasts six or more.

cross-posted from: https://lemmy.dbzer0.com/post/38015770

A washing machine is trapped in a fault state even though all the components function (AFAICT). The controller board has two ports:

• ISP (to attach an ISP programmer to flash new software)
• USART (4-pin serial port: 0v, TX, RX, 5v)

I’m guessing the ISP port is useless without whatever proprietary software is needed. But what can the USART do for me? Can that be used to obtain the error code and clear it, or reset the board to the factory state? Has anyone done that, without documentation?

I was never able to obtain the service manual (thread).

When I called them for repair info, they asked me if I was a consumer or a pro repair tech. I find the question disturbing. If I lie then I risk them making assumptions about my tools and perhaps willingness to pay for wiring diagrams and service manuals. They probably assume I would have access to some kind of subscription to a pro repair platform that has that info. So I admitted to being a consumer at the risk of being told “call a pro”. They asked what was wrong with the machine, then I got cut off because I ran out of phone credit (after being on hold for a long time). A glimpse of their answer before the call dropped sounded like incorrect speculation.

Google as a repair gatekeeper

I’ve watched a lot of Youtube videos. Also quite disturbed that Google is the gatekeeper of educational repair videos which it blocks me from saving a local copy of. I can only watch at the public library because I do not have an unlimited connection. So I have to memorize steps or take notes.

I don’t give a shit if Google jails viral cat videos. But Google jails knowledge, and it’s knowledge that ecocide avoidance depends on. Maybe we need some #Greenpeace sponsored peertube instances to become a go-to for repair instructions.

Error codes concealed

One of the youtubers said Beko’s PCB stores the error internally but it will not allow consumers to see the error code. Does anyone see a legit reason for this? Even if they intend to block consumer repair, consumers would still have a legit use for the code -- to have an idea of the cost of a problem before calling out a costly technician. It would also be sensible to to give the code to the technician in advance.

The Youtuber (fuck I hate that term) gave a secret sequence of steps to get the PCB to disclose the error code. The special sequence has no effect on my particular model. Different models have different secret sequences to reach the diagnostic mode, and no youtuber has covered my exact model. Similar models from roughly the same time period are covered in various non-English languages, but when I repeat the motions I do not reach diagnostic mode.

Do I have a serial port?

The PCB has unused pins labeled 5v, TX, RX, ~~9v~~ 0v. Apparently this is a serial port for an RS232 cable. What info likely comes from it? My boiler PCB has the same thing and in that case there are mere hints in the docs that it is to attach a device that transmits telemetry data over GSM. Would a washing machine have that? Or is it to attach a PC or proprietary device to get detailed telemetry data?

Secret steps to tell the machine that the fault is gone

A youtuber gave the impression that after fixing the problem (whatever it is, I still don’t know), the PCB will still think there is a fault until steps are taken to clear the error. The video ended before he explained how to do that -- although it was not my model and would not have worked for me anyway. But I have to wonder if my problem is merely a false detection of a problem. Or if it’s a real problem that I fix, is it likely that I have track down a special sequence of steps to tell the PCB the fault is fixed?

I think if I could go back in time and mention this experience to Twilight Zone writers in the 1980s, they would simply write this into an episode as-is.

cross-posted from: https://lemmy.dbzer0.com/post/36272192

My washing machine is dying in stages. It started with the same symptoms as this thread. Specifically, after filling the tub for a wash cycle, it would go straight into a high-speed spin (full of water!) for a second or two (instead of the expected slow tumble), then quit. The speculation is that the tachometer is failing.

Then the machine got worse. I now cannot even start any program. No matter what program I select, I press start and after a few second pause the start button LED just blinks. It’s a generic blunt signal of a fault. The blinks are evenly spaced non-stop, so there is no error code of any kind.

To test the universal commutative motor, I followed the linked video and took resistance measurements. All seems okay in that regard (but this is based on vague resistance ranges that are not specific to my machine). Test results:

tachometer

Expectation: any reading that is not infinite/disconnected is fine.

• (video): 70 Ω
• (my motor): 52 Ω

carbon brushes:

Expectation: should be in the range 1—7 Ω

• (video): 5 Ω
• (my motor): 3.45 Ω

field windings:

Expectation: all combinations should be 1-7 Ω

• (video): 3.5 Ω
• (my motor): pin1-2: 1.35Ω, pin1-3: 1.35Ω, pin2-3: 1.9Ω

I do not have whatever model is in that video, but my readings are in the range suggested by the video presenter, fwiw. I believe my testing is incomplete because I was expecting the tachometer to be bad based on the behavior.

Motor spin test (hotwiring)

My next move was to try to make the motor spin. There is no service manual or wiring diagram for my Beko. So I inspected the motor and derived these pins (quotes are labels on the PCB):

1 (field) brown socket → brown+white “stator M” 2 (field) black socket → purple+white “stator 1” 3 (field) blue socket → red+white “commu” (commutator) 4 (brush) white socket → purple+yellow “rotor 2” 5 (brush) red socket → green+red “rotor 1” 6 (tacho) yellow socket → green “tacho” 7 (tacho) yellow socket → green “tacho” 8 (ground) green+yellow socket → green+yellow

Someone suggested this wiring:

L → pin 1
N → pin 5
jumper connecting pins 3 & 4

I did not connect direct to the wall because I wanted to use the mechanical power button of the machine to turn on and off the motor (so I could quickly cut power if needed). So power took this path:

wall (220 VAC) → safety capacitor → mechanical button → motor (wiring redirected to motor instead of control panel)

When I switched it on, the motor spun for 1 or 2 seconds and I saw a white flash (I think) and the motor quit. I turned it off. Then tried to switch it on again. No response.

220 VAC quit coming out of the safety capacitor. Instead the voltage jumped around between 10 VAC and 20 VAC. So I thought I fried the capacitor or resisters therein. I checked the motor to see if any of the pins connected to ground (answer: no, so the motor was not harmed). Then I disconnected the safety capacitor and connected it just to mains and ground. 220 VAC was output (WTF.. why does it magically work again?)

I think I’m back to the state it was in before I tried to power the motor. But I want to understand why the safety capacitor apparently flashed white and temporarily died with only 10—20vac output. I need to get to the bottom of this because I still need to test the motor for more than 1 second in a way that doesn’t cause more white flashes. Is it a bad idea to have the safety capacitor in the circuit?

I cannot repair my washing machine without documentation. I have no idea how to use my multimeter to check the components. There are parts dealers for Beko in my area, but none of them have the service manual.

The parts shops all say go to the website for the manual as a flippant off-the-cuff answer. There are no service manuals on the Beko website -- at least not for my model. The navigation of the Beko website does not even have a path to docs. And worse, my model is treated as non-existent by the website.

What would I do if I were a professional repair service? What is the official channel?

I am open to “piracy¹” but it would be a long shot to scour all the dark web for a manual for a specific washing machine. It’s not the type of content people have a strong interest in spreading/trading.

¹As RMS says, it’s not a just and appropriate term for it (but “sharing” is awkward too).

(update) Added frame from Youtube video t1XaUolbjLY which shows that service manuals exist for at least some Beko models. As we can see in the snapshot, Beko wants to restrict who is servicing their machines. (btw the video covers a very different model than mine).

I probably need to find the test mode for my machine, comparable to YT video cq_uSyghZC0.

new problem

The machine reached a new low. Now there actually is a problem with the water valve, it seems. When running a program, it pauses then the start button just blinks. (It previously started by pumping then at least filled the tub). So I followed this video from the 5m10s position. I do not get 220v on either valve. But certainly I can see that 220V is getting to the control panel. So 220v goes into the control panel, but does not make it out of the control panel leads where the water valve connects.

(edit) The water valves themselves are fine (I connected 220v directly to the water valves and water flows). I guess I should suspect the pump now. The pump was actually the very first task back when the machine worked. So I should not have even been looking at the water valves which only start after the initial pumping ends.

I really enjoy the arcade blogger for the arcade cabinet raid writeups he does, and his step-by-step repair posts. The history aspect is neat too.

This is a repair post with a bit of history.

Decades ago, to combat ROM-cloning piracy, Capcom started adding a chip to their PCBs that stored encryption keys in memory backed up by an onboard battery. You may see the problem here - batteries are not meant to last forever, and if the chip lost power AT ANY TIME the keys were lost and the game was unplayable.

This feels like yet another example of the total disregard corporations hold for the media they own the rights to, in favor of short term profits. We've seen before that many works produced by entire teams would have been lost if not for the efforts of pirates and amateur archivists.

To quote the blog:

Its hard to say if Capcom knew this would happen, but then again, the shelf life of most arcade games was months, or at most a couple of years – I guess it wasn’t something they planned for.

Fortunately, this is a well established problem with a motivated, technically-minded community looking for a solution, so this early DRM has already been circumvented. The article doesn't go into detail on how they researched and reverse engineered this sabotage, but I might do a little reading and edit the post if I find anything cool.

Edit: this seems to have more details and is an interesting read so far: https://arcadehacker.blogspot.com/2014/11/capcom-kabuki-cpu-intro.html?m=1

The gist is that the problem is well solved at this point and there's a small industry of aftermarket components out there that are nearly plug and play. The version the author went with works like this:

You desolder the dead battery and replace it. Then desolder the blank sabotage chip. You swap in the aftermarket one and configure it (by using tiny switches) to inject the correct set of encryption keys. Then you slot the blank sabotage chip into the aftermarket one.

When the game is powered up, the aftermarket chip restores the encryption keys, the PCB looks for the keys, then successfully used them to run the game files.

The cool thing is that the sabotage chip is now functional again on its own.

The author spent some time restoring the board to stock, by keeping the sabotage chip powered with another battery while removing the aftermarket chip so it could be used elsewhere. I should appreciate the effort at not wasting any resources but I think it makes sense to keep the de-sabotauge chip as a permanent addition, as it automatically prevents the kind of data loss the company intended.

Either way, it's a neat article and I'd recommended reading it. He does a lot of arcade cabinet restorations, but generally sends the electronics away for repair, so this was a neat one.

I know tech has come a long way since these were made but there's something to be said for these big, chunky, through-soldered components and the well-documented wiring instructions that often came with them.

Decided to restore this old thing from out in the shed. It was in the weather for a couple of years, and I think it's 20+ years old.

Replaced the pump with a USB powered 5 watt silent one.

Replaced the halogen lamp with an LED one.

Cleaned as much as I could.

Seems to be working well. How long the $12 ebay pump lasts remains to be seen.

More pics at https://photos.app.goo.gl/HDmVb2K1MeUQTfuG8

Hi,

Is there good literature on how to repair stuff? Just general things, not specific appliances. How to repair wood, how to properly sand wood and metal, how to replace a tile, how to read and analyze circuits, identify faulty parts and correct replacements, etc. I just want to become better at repairing stuff.