Test Equipment Anonymous (TEA) group therapy thread

[mnementh]

CANbus setup short. You'll need to know which vehicle library and which CANbus decoder brick your particular car needs.

mnem
 

[AaronLee]

I bet there a few mouse nests in there as well 

And given Murphy's Law, for whoever buys the lot, the mice made a nest out of just the one manual you needed the most.

[TERRA Operative]

(1)  Steering Wheel KEY 1
(2)  Steering Wheel KEY 2
(11) CAN BUS TXD
(12) CAN BUS RXD
(14) BRAKE.  I guess it means brake pedal...
(17) ILL ...like in "Illumination" / backlight ? I see that this wire goes straight to the external can bus decoder box...
(18) ACC ... accelerator ? They monitor the brake pedal so why not the Accelerator pedal
(19) ANT .... "Antenna" I guess ? Power feed to extend a telescopic antenna ? Or power a roof mounted amplified antenna ?
(20) REVERSE :  I guess it means reverse GEAR, probably used to automatically switch to the rear view camera video feed...

(1)  Steering Wheel KEY 1
(2)  Steering Wheel KEY 2
These will either go to two buttons, or to multiple buttons with a resistor network so each button press creates a different resistance that is read by the unit to know what button is pressed.
This may be a standard, or may be configurable in the unit for different car manufacturers.

(11) CAN BUS TXD
(12) CAN BUS RXD
Self explanatory. Some CAN bus stuff, not sure if it connects to the car's CAN bus, or to add-on modules.

(14) BRAKE.  I guess it means brake pedal...
I assume it connects to the brake switch or brake lights for some function.

(17) ILL ...like in "Illumination" / backlight ? I see that this wire goes straight to the external can bus decoder box...
Illumination, connect it to the wire that switches to +12V when you turn on the headlights.
It'll make the buttons light up, dim the screen etc for 'night mode'.

(18) ACC ... accelerator ? They monitor the brake pedal so why not the Accelerator pedal
Accessories, Connect it to the wire that switches +12V when you turn the key to the first or 'ACC' accessory position.

(19) ANT .... "Antenna" I guess ? Power feed to extend a telescopic antenna ? Or power a roof mounted amplified antenna ?
A +12V output that activates when you turn the unit on to provide power for electric antennas/amplifiers/etc

(20) REVERSE :  I guess it means reverse GEAR, probably used to automatically switch to the rear view camera video feed...
You guessed correctly. Connect it to the reverse tail lights.

[beanflying]

Would need to strip some of the guts out of it before adding the lightbulb  eBay auction: #304179850148

[med6753]

Would need to strip some of the guts out of it before adding the lightbulb  eBay auction: #304179850148

Couldn't afford a range switch? That would be some pain to use.

[beanflying]

The terms are 'eccentric' or 'delightfully quirky' 

[med6753]

The terms are 'eccentric' or 'delightfully quirky' 

I think 'fucking useless' is more appropriate. 

[cyclin_al]

I need your help people... lots of you are into audio stuff so I am hoping one of you will know the answer.

I have just been given a car stereo for repair. First circle of friend and he is not a time waster and first time he ever asks me to fix anything... so I agreed to " have a look " at his piece of junk. It a double DIN / height modern thing that does everything and anything, with a big LCD screen, can display video from front or rear camera on the vehicle, TV, wireless or wired, USB, all sorts of I/O. It's less than a year old, still under warranty but the seller apparently is so great that he won't take responsibility for it. Only matched by the quality of the unit itself : 100% chinesium, the worse kind : there isn't even a BRAND name or model written on it, nada ! You just don't know what it is, nor who made it. Only thing you have is a sticker with a wiring diagram/ pinout for the very many connectors this thing carries at the back !!!! 

Owner said the problem is taht the screen is dead, black, no picture no matter what.. BUT that the unit still powers up and that if he blindly plays with the knobs, unit seem to operate normally and he can listen to radio and what not. So I thought OK looks like it's just a dead display, might be worth having a look at it, especially since unit is so new and cost so much money, 300 Euros... a fortune for chinese crap that's not even branded/identified !   

So I have this thing on the bench, actually very service friendly can't complain. Easy to split and the mother board is perfectly accessible with all the components on the top side, great. Front panel assembly show you the component side of the LCD board, where I can see the connector tghat goes to the backlight, and the board carries a couple SMPS. So I can easily test the backlight and make sure the SMSP work as well.

So I give the unit power and ground... press the pwoer button but... it does not wake up ! And the ammeter shows that the unit draws 0.000mA, not a mA ! Weird....

So I am here to ask those in the know here, who are used to wiring modern car stereo... is there some other input I need to wire up other than ground and battery / +12V, the twxo beefy wires, to get the thing to power up ??   I see the unit is connected to the freaking CAN Bus of the car ! 
There is an external little box supplied with the stereo, that reads " CAN Bus decoder ". Whatever that does. I plugged it in, but no joy either.

Could a problem with the can bus, or lack there of, prevent the unit from powering up ? Do I really need to have CAN working to xwake up the stereo ? That means I can't debug the unit on my work bench then ?! 

Also there are other inputs whose name sounds mysterious, maybe they need to be wired as well, I don't know. See close up picture of the pinout diagtram below :

(1)  Steering Wheel KEY 1
(2)  Steering Wheel KEY 2
(11) CAN BUS TXD
(12) CAN BUS RXD
(14) BRAKE.  I guess it means brake pedal...
(17) ILL ...like in "Illumination" / backlight ? I see that this wire goes straight to the external can bus decoder box...
(18) ACC ... accelerator ? They monitor the brake pedal so why not the Accelerator pedal
(19) ANT .... "Antenna" I guess ? Power feed to extend a telescopic antenna ? Or power a roof mounted amplified antenna ?
(20) REVERSE :  I guess it means reverse GEAR, probably used to automatically switch to the rear view camera video feed...

Pedals, antenna, reverse gear, I guess I can safely ignore.

However ACC not sure, ILL not sure either, CAN bus I am scared, and steering wheel KEY 1 and 2 scared too...

Please help poor Vince, how the fuck can I wire this thing on my bench to get it to power up and draw some current ! If AT ALL possible ?! 

My first GUESS, and it is just a guess with an unknown unit, is that the CAN bus is used to communicate with volume control buttons on the steering wheel, and simple user interface things like that.  Yes, it would be hooked up to the BCM (body control module) in order to get those communications.  I doubt CANbus functions are in any way required for the unit to function.
EDIT:  Vince has said that there is no power being consumed by the unit.  CANbus is not a worry at this point.  The CANbus transceiver and a microcontroller in the unit need to be active in order to communicate with the car.  Until that hardware has power, there will be no CANbus.

One thing I suspect, and again guessing, is that the steering key stuff has to do with anti-theft or immobilization functions.  I know car stereo theft was a huge deal in the GWN back in the '90s.  Maybe that compulsion has held over.  My guess is those signals might have to do with enabling the unit to turn on or not.  However, 0 mA meaning no phantom power being consumed makes me think this guess is not a very good one.

ACC being accesory power is the most likely case, as the dwagon and several others suggested.  Certainly worth a try.

[cyclin_al]

Happy Thanksgiving !  The date may be different elsewhere, but it is this weekend in the GWN.

It is actually a pretty good weekend.

SWMBO's nephew got married yesterday.  I missed out on it. 
Instead, I took the kiddo for a COVID test.
Results cam back negative 
School announced during the weekend that there are more cases ...    I figured that they should have already closed the school for a 14-day quarantine 

I am running on severe lack of sleep.  Last night was spent giving the dog a de-skunking bath with special shampoo.  This is where a smaller dog is easier to deal with ... dog still smells somewhat.  Garage, especially around the tractor door, stinks terribly.

Despite all this, I am making progress re-wiring a panel.  I will not be testing or plugging anything in this evening.  Turkey is in the oven for tomorrow, and I am not going to take anything to chance with the electrical supply.

[vk6zgo]

The terms are 'eccentric' or 'delightfully quirky' 

I think 'fucking useless' is more appropriate. 

That sort of thing was pretty much confined to the "economy" sector of analogue VOMs, back in the day.

Full size VOMs, had full switching, including in most cases, the change between Volts, Ohms  & Amps ranges, so that you didn't have to unplug one lead & change it to another socket to measure current, as is common with DMMs.

The "trap for young (& old) players" of  leaving the leads in the current position, switching to volts, testing the latter, & blowing things up, was not possible with such switching,& you had to be "double dumb" to hang the things across volts with the switch still on the current range.

[vk6zgo]

I need your help people... lots of you are into audio stuff so I am hoping one of you will know the answer.

I have just been given a car stereo for repair. First circle of friend and he is not a time waster and first time he ever asks me to fix anything... so I agreed to " have a look " at his piece of junk. It a double DIN / height modern thing that does everything and anything, with a big LCD screen, can display video from front or rear camera on the vehicle, TV, wireless or wired, USB, all sorts of I/O. It's less than a year old, still under warranty but the seller apparently is so great that he won't take responsibility for it. Only matched by the quality of the unit itself : 100% chinesium, the worse kind : there isn't even a BRAND name or model written on it, nada ! You just don't know what it is, nor who made it. Only thing you have is a sticker with a wiring diagram/ pinout for the very many connectors this thing carries at the back !!!! 

Owner said the problem is taht the screen is dead, black, no picture no matter what.. BUT that the unit still powers up and that if he blindly plays with the knobs, unit seem to operate normally and he can listen to radio and what not. So I thought OK looks like it's just a dead display, might be worth having a look at it, especially since unit is so new and cost so much money, 300 Euros... a fortune for chinese crap that's not even branded/identified !   

So I have this thing on the bench, actually very service friendly can't complain. Easy to split and the mother board is perfectly accessible with all the components on the top side, great. Front panel assembly show you the component side of the LCD board, where I can see the connector tghat goes to the backlight, and the board carries a couple SMPS. So I can easily test the backlight and make sure the SMSP work as well.

So I give the unit power and ground... press the pwoer button but... it does not wake up ! And the ammeter shows that the unit draws 0.000mA, not a mA ! Weird....

So I am here to ask those in the know here, who are used to wiring modern car stereo... is there some other input I need to wire up other than ground and battery / +12V, the twxo beefy wires, to get the thing to power up ??   I see the unit is connected to the freaking CAN Bus of the car ! 
There is an external little box supplied with the stereo, that reads " CAN Bus decoder ". Whatever that does. I plugged it in, but no joy either.

Could a problem with the can bus, or lack there of, prevent the unit from powering up ? Do I really need to have CAN working to xwake up the stereo ? That means I can't debug the unit on my work bench then ?! 

Also there are other inputs whose name sounds mysterious, maybe they need to be wired as well, I don't know. See close up picture of the pinout diagtram below :

(1)  Steering Wheel KEY 1
(2)  Steering Wheel KEY 2
(11) CAN BUS TXD
(12) CAN BUS RXD
(14) BRAKE.  I guess it means brake pedal...
(17) ILL ...like in "Illumination" / backlight ? I see that this wire goes straight to the external can bus decoder box...
(18) ACC ... accelerator ? They monitor the brake pedal so why not the Accelerator pedal
(19) ANT .... "Antenna" I guess ? Power feed to extend a telescopic antenna ? Or power a roof mounted amplified antenna ?
(20) REVERSE :  I guess it means reverse GEAR, probably used to automatically switch to the rear view camera video feed...

Pedals, antenna, reverse gear, I guess I can safely ignore.

However ACC not sure, ILL not sure either, CAN bus I am scared, and steering wheel KEY 1 and 2 scared too...

Please help poor Vince, how the fuck can I wire this thing on my bench to get it to power up and draw some current ! If AT ALL possible ?! 

Three questions;

Does the car still have a real ignition key, or does it just have a silly keycard +"on" button?

If the former, & the ignition switch is on the side of the steering column, maybe the key switch has a second pair of contacts which are used to loop between (1) & (2).

Alternatively, is there an "entertainment centre" control button either on a short stalk, or one end of the steering column "spoke"?

If you know which plug connects to each thing on the stereo, can you probe the likely ones to see if they do anything if you turn the ignition on & off, etc?

[capt bullshot]

Hey Guys 'n Gals,
[...] Fluke 199C's. Very nice machines, [...] started charging it. [...] Was charging fine, at some point I was happy with how much was in the battery and turned of the power supply.

R4101 blew up pretty much instantly.

[...] New Unit,  start with the original, pretty dead battery, at least it's 100% sure compatible. These things have a battery recycling option so I hit that.

R4101: from glow to blow in about 2 seconds.

At this point, I obviously start to doubt everything. The dead units I have? May well have blown when I turned them on. Have I been sending out crap units? And more importantly: WTF?? So I started looking at the schematic. R4101 is a 2010 0.15 resistor. For it to blow up that fast it must have been pulling, what, 3A?

To me it seems the likely current path has to be from the batteries.
...

+1 on that. Looks like something in the power circuitry has shorted. There's the LCD backlight converter directly connected to VBAT, V4000, the flyback converter and the PM Asic itself. V400, the flyback transistor (V4001) and the backlight converter all have some kind series impedance (L4201, R4000, R4003) that might show some evidence s of short circuit current happened. Worst case would be Pin 3 of N4000 , or anything else related to this PM Asic.
I'd try to rule them out by applying power from a lab PSU instead of the battery (which can have quite high short circuit current output) and try to measure what happens ...

[Robert763]

Thanks for all the good info ! 

1) I wired +12V to the ACC wire. That meant I had to unplug the CAN bus box because the ACC wire goes straight to that box !
Result = no change, still no power, still does not draw even ONE mA... nothing at all.

2) Visual inspection leads nothing, damaged components or dislodged flat flex cables.

3) Probed main board for voltages... checked all electrolytic caps, all "SMPS" looking SMD coils, and all discrete components, diodes and what not, in the immediate vicinity of the main/power connector at the back of the unit. ZERO voltage everywhere. If the screen/front panel was at fault, the mainboard would still be running so I would be able to see supply VOLTAGES and theunit would draw at least SOME current ! But it draws absolutely zero, and zero voltage anywhere. So there is no point worrying about the display or flat flew cables at this point I think ! 

4) REGARDLESS of whatever magic the unit might perform to "decode" its numerous inputs, to decide if it will be so kind as to power up the unit... this smartness is embodied by circuitry, a CPU, or a dedicated power management chip or discrete circuitry or SOMETHING.... this means that even if the conditions for the unit to power up were not satisfied, the smarts inside the unit would be powered up anyway and therefore draw some current and need voltage to run.... but since I can't even witness the slightest current draw... it means the main board is just not "upset" with the way I wired it... it's more like it's simply not powered up AT ALL !

So I am now shifting from a "No display" fault, to a "No Power" fault.... not at all the same thing...

Zerto votlage and zero current... it's just like the main fuse was blown, as simple as that ! Yeah, I did check the fuse... it's fine...

Next step is to unscrew the main board from the chassis, so I can inspect the underside of it, and access the pins of the power connector. I will trace the +12V pin to see where it lands on the top/component side, then trace it there.

But it's too late now, going to bed.  Before I waste time on this thing, I will the owner who actually lives just 500m from me, so he can come with his car. I will ask that we  reconnect the stereo to his car, so he can show me if he really can get this thing to work.... as a quick sanity check. If it does power up and work, then I don't understand how that's possible 

More tomorrow then, good night ! 

Only one other thing to ask...
Have you connected the chassis to the supply 0V?
If you really have 0mA current draw there must be an open circuit somewere. Likely either a bad joint (soldered, crimped or loose connector) or a track on PCB or ribbon having been taken out by a short circuit.
If it poweres up with the chassis connected look to the 0V wiring.
Generally these do not need CAN for basic operation. CAN can save a bunch of auxillary wiring like the steering buttons, reversing and brake light. Some CARS need the CAN to stop warnings about the radio or make things like nav system work.
The CAN box with the AUX connetion is beause the car it's in does not have a switched 12V AUX to tell the unit the car is "on". The  CAN box decodes the signals and provides a switched 12V AUX supply. Commonly needed on Fiats for eample.

More generally, this unit is clearly a steamng turd. As it has no makers name, no E approval and an incorect CE mark it is clearly ILLEGAL to ue in the EU or install in a car. Technically fitting it invalidates the car's type approval and probably the insurance.* (Yes I know there is zero enforcement) Tell your friend to demand his mony back. If he paid with a credit card or PayPal they are also liable in most cases. If no joy tell him to talk to consumer protection or whoever enforces CE compline in France.

* Any electronic item that is installed in a car in the EU/UK that may affect the safety of the vehicle must be properly approved and "E" marked. Items that have displays or controls for the driver are explictly included as affecting safety.  CE marking does not apply to automotive parts.

[capt bullshot]

Sold another one last week, so I put in a new battery and with the unit still open started charging it. The adapters are UK style so I took a plug and hooked it up to a Rigol DP832. Was charging fine, at some point I was happy with how much was in the battery and turned of the power supply.

...

R4101 blew up pretty much instantly.

...

Fact is I just can't come up with a mode that would lead to 3A being dumped through the battery unless the charging circuitry has some readlly bad design flaws. Can't imagine Fluke would have let that go out the door, can't imagine it not being a widely known and discussed fact. So, before I start risking more units, I'd like to understand...

Another wild guess: Your lab PSU went into CC mode while charging (or for some strange reason at the end of charging) and produced some overvoltage as it recovered from CC to CV mode when the Fluke stopped to draw charging current. Some PSUs can have major flaws here - don't ask me how I know

[Robert763]

Hey Guys 'n Gals,
[...] Fluke 199C's. Very nice machines, [...] started charging it. [...] Was charging fine, at some point I was happy with how much was in the battery and turned of the power supply.

R4101 blew up pretty much instantly.

[...] New Unit,  start with the original, pretty dead battery, at least it's 100% sure compatible. These things have a battery recycling option so I hit that.

R4101: from glow to blow in about 2 seconds.

At this point, I obviously start to doubt everything. The dead units I have? May well have blown when I turned them on. Have I been sending out crap units? And more importantly: WTF?? So I started looking at the schematic. R4101 is a 2010 0.15 resistor. For it to blow up that fast it must have been pulling, what, 3A?

To me it seems the likely current path has to be from the batteries.
...

+1 on that. Looks like something in the power circuitry has shorted. There's the LCD backlight converter directly connected to VBAT, V4000, the flyback converter and the PM Asic itself. V400, the flyback transistor (V4001) and the backlight converter all have some kind series impedance (L4201, R4000, R4003) that might show some evidence s of short circuit current happened. Worst case would be Pin 3 of N4000 , or anything else related to this PM Asic.
I'd try to rule them out by applying power from a lab PSU instead of the battery (which can have quite high short circuit current output) and try to measure what happens ...

I think the key is the lab supply. Was anything else connected to or touching the 'scope at the time? The OEM supply has high levels of isolation and low leakage current for safety reasons. It's possible the bench supply may be providing a sneak path or somehow damaging the charging circiut.
Take the battery out of the faulty scope and measure the impedance / volt drop between te Batt + connecton and circuit common (low end of R4101) of the 'scope. Use a DMM on diode test mode. This will tell you if it is a hard short circuit or something being switched e.g. by charge controller.
 

[Ice-Tea]

I've seen what ammonia from farm biogas reactors does to copper bus bars, and we had early death of resistors in S7-200 automation controllers used there - only one type of resistor failed, a tiny smd 4 resistor network used for led current limiting. Given the use and size I would say it had a high current density per area of resistive element.

Didn't know that was a thing. Will check on others, at least that's guaranteed non destructive

but I would suspect something in the power supply afterwards taking lot of current?

That was my original thought as well, but with two diodes in series, I just don't see how... But if/when I continue powered tests, will have more monitoring for sure.

Another wild guess: Your lab PSU went into CC mode while charging (or for some strange reason at the end of charging) and produced some overvoltage as it recovered from CC to CV mode when the Fluke stopped to draw charging current. Some PSUs can have major flaws here - don't ask me how I know

That's interesting. That would make a lot of sense, and, well, it's Rigol. It would also be my absolute worst nightmare as that increase the chances of more than a sense resistor being cooked.

I think the key is the lab supply. Was anything else connected to or touching the 'scope at the time? The OEM supply has high levels of isolation and low leakage current for safety reasons. It's possible the bench supply may be providing a sneak path or somehow damaging the charging circiut.
Take the battery out of the faulty scope and measure the impedance / volt drop between te Batt + connecton and circuit common (low end of R4101) of the 'scope. Use a DMM on diode test mode. This will tell you if it is a hard short circuit or something being switched e.g. by charge controller.

A lot of folks cycling back to the PSU. Figured it was a good strategy as it would allow the protection of a settable current limit, starting to doubt now. Maybe I should take an HP PSU out of stock for this...

[Ice-Tea]

Anyone with access to a N8974A, N8975A or compatible noise figure analyzer? Would like to test something, quite willing to pay...

What do you need exactly ?

Heh... wasn't exactly expecting that to be picked up anymore but essentially: I have a measuring head for it, don't know if it's good or not. Makes it a tough sell. If anyone could verify and provide a stamp of approval preferably with a sexy impressive picture of the thing in action that would help a lot. And again: quite happy to pay for the privilege...

[srb1954]

I need your help people... lots of you are into audio stuff so I am hoping one of you will know the answer.

This YT video might provide some guidance on how to hook up one of these units.

https://youtu.be/TeXCC0P_iqM

If you fast forward to 3:50 it shows the wiring to the unit. Bear in mind this installation is for a Lamborghini Gallardo so YMMV with a different vehicle.

Good luck!

[Ice-Tea]

Smells to me like the batteries are trying to charge a large capacitor bank;

Well, yeah, but diode and I don't see any cap banks that size on the board, the PSU one is protected by diode and that would also mean it would have to sustain current for a long time, not just peak.

maybe some weird interaction between that particular bench PSU and that particular BMS in the Fluke. Could be you PSU is using a FET-based crowbar that is allowing some feedback.

Bench PSU again... Really starting to think replacing it with an HP unit. Or TTi.

I would replace the resistor in the 2nd unit with OEM battery, then try again with a small fuse and diode inline in your charge cable and see if it continues to misbehave. Alternately, look for a proper wall-wart for one of these and see if it does in fact play well with the aftermarket battery. If so, your customer may have been at fault for using a wrong power source.

I like the fuse idea.Might put one in the battery lead and one in the PSU lead. Can't really blame the customer, I think. He got either the original wall wart or a replacement I provided. Also puts the bench PSU in question as main culprit. Just too much coincidence around this...

[Neper]

Heads up for all German members...

The German Institute for Broadcast Technology (IRT) in Munich is being disbanded which in itself is a shame and a sin but there's nothing we can do about it. They're currently auctioning off all of its assets, mostly office furniture and computer/network equipment but also various electronics and TE. The list includes things like Rohde & Schwarz ADS with a starting bid of 200 euros, HP Universal Counter, Modell 5314 A, various decoders, encoders and analyzers, a Tektronixs (sic!) 2430 scope starting at 50 euros, Weller WECP 20, 3D printers, Rohde 6 Schwarz FSQ8, CMW500, Tektronix TDS 724A, Audio Precision System One +DSP, Rhode & Schwarz SAF... over 1800 items.

https://www.reich-bayern.de/de/objekte/au-116/liquidationsversteigerung_der_irt_institut_fuer_rundfunktechnik_gmbh_teil_2_-_ha?Lstatus=1

Do note: +18% fee, +19% VAT, no shipping, bidder collects.

[psykok]

Originally not my suggestion but for sure you will enjoy this channel.

[dl6lr]

Another wild guess: Your lab PSU went into CC mode while charging (or for some strange reason at the end of charging) and produced some overvoltage as it recovered from CC to CV mode when the Fluke stopped to draw charging current. Some PSUs can have major flaws here - don't ask me how I know

Yeah, got a Hameg HM7042 Triple Power Supply for almost nothing. Blown up pass transistor, shunts etc. on one channel. Repaired to some degree and observed some nasty behaviour. Asked in de.sci.electronics and got confirmation, that those huge overshots when switching from CC to CV are not a specific problem of my blown device...

[mansaxel]

Heads up for all German members...

The German Institute for Broadcast Technology (IRT) in Munich is being disbanded

Aw, fuck! I have friends there!

[mnementh]

Smells to me like the batteries are trying to charge a large capacitor bank;

Well, yeah, but diode and I don't see any cap banks that size on the board, the PSU one is protected by diode and that would also mean it would have to sustain current for a long time, not just peak.

maybe some weird interaction between that particular bench PSU and that particular BMS in the Fluke. Could be you PSU is using a FET-based crowbar that is allowing some feedback.

Bench PSU again... Really starting to think replacing it with an HP unit. Or TTi.

I would replace the resistor in the 2nd unit with OEM battery, then try again with a small fuse and diode inline in your charge cable and see if it continues to misbehave. Alternately, look for a proper wall-wart for one of these and see if it does in fact play well with the aftermarket battery. If so, your customer may have been at fault for using a wrong power source.

I like the fuse idea.Might put one in the battery lead and one in the PSU lead. Can't really blame the customer, I think. He got either the original wall wart or a replacement I provided. Also puts the bench PSU in question as main culprit. Just too much coincidence around this...

Have you already tried with a diode inline between the unit and the PSU...? Did I miss that?

Also... unless these have some oddball connector, you cannot be sure the customer used only the wall-wart you provided. I'm currently having a bit of a tiff with my son over this:



...which, even after we built it together, and I explained exactly how the 5V-powered BMS etc worked and repeatedly stressed that he could not charge it with anything but the USB charger/cables I gave him with it, he still plugged it into the 12V charger for his BT speaker because he couldn't find the correct cables after the move.
And when it didn't charge (probably also let out some magic smoke), he then plugged its brother into the same charger as well.

mnem

                               Is it time for THIS yet...?

[Zucca]

The German Institute for Broadcast Technology (IRT) in Munich is being disbanded

of course, few months after I left Munich.