Technical Engine dies and loses ECU connection (Punto mk2, 2004, 1.2 8V)

[Ksport472]

Hi!
I have a 2004 Punto with the 1.2 liter 8V engine. Recently I did some maintenance work (timing belt, spark plugs etc.). The car was working fine before that (except for occasional power steering isues). Now after the maintenace work the car runs only for few minutes until the engine suddenly dies. When that happens, there seems to be no connection to the ECU. The engine doesn't start (cranks fine, but probably no spark and/or fuel) and I cannot connect to the ECU through the OBD2 port. If I wait for a while, or if I disconnect and reconnect the battery, the ECU connection seems to work again and the engine starts fine and runs again for few minutes max. This happens over and over again. It seems like the ECU just shuts down suddenly or is unable to communicate when the issue triggers. All the accessories like blower motor, lights, radio etc. continue to operate normally.

I have tried to:

• Read the fault codes (with a generic reader). Usually there isn't any codes at all. Maybe once per ten attempts it might throw some codes, but hard to say if they are related. One time I got codes U1600, U1602, P1325 and one time I got some coil-related codes. Dash seem to also lose connection to the ECU, because it might blink the odometer, blink the ABS and brake indicators or burn the yellow immobilizer light. When the engine starts and a connection to the ECU exists, there are no error indications on the dash.
• Monitor live data using a generic scanner. The connection to the ECU drops immediately when the engine dies, so nothing abnormal can be seen there. At least in the basic values that I could see.
• Disconnect and reconnect most of the wire harness connectors and clean the battery cables and terminals.
• Clean and measure the ground cable to the throttle body / ECU.
• Measure the voltage when the engine is running. Alternator is working and the voltage seems normal.
• Add extra ground wires to the engine and throttle body / ECU.
• Poke the cables and connectors when the engine is running to see if it has any effect -> nothing.
• Swap some random relays with identical ones from different places. Most likely I swapped all of those ECU relays too, but I'm not sure anymore.
• Remove the battery. It was actually removed for couple weeks because of the maintenance work.
• Warm or coold down the engine. Exact engine temperature does't seem to have any noticeable effect, so most likely not a thermal expansion issue.

My primary suspects:

• Wire harness. I had to disconnect most of the connectors and I moved the harness a bit during the maintenance work. I don't see any clear damage, but maybe there could be a short circuit somewhere and the ECU goes to some kind of safety mode for a while because of that? Or maybe there is a voltage drop which causes a brownout for the ECU?
• Broken ECU. It would be bad luck if the ECU happened to break when it was almost untouched when the car sat in the garage. However, this would explain the sudden deaths of the engine and also the lack of connection to the ECU when the issue is active.
• Broken ignition components. I have read that the coils are a common fault in these engines and they might kill the ECU. Not sure if a broken coil, a spark plug or a spark plug cable could "tease" the ECU so badly that it shuts down after a while. I changed the spark plugs during the maintenance, but they should be similar to the old ones.
• Immobilizer. I would assume that the engine would not start, or at least not run for so long time, if this was the case. Also I don't think it should prevent any connection to the ECU even if the immobilizer triggers. Also the yellow immobilizer light only comes (not every time) if I try to start the engine again immediately when the issue occurs, but that might be because of the missing ECU connection.

Do you have any ideas what could cause this? I would not like to randomly throw a bunch of new parts into the car, but this seems to be a bit tricky to diagnose, even though the engine is pretty simple.

 

[Fiat1271300]

Hello, I have the 16v version and my brother has the same version 8v as you. I have the same symptoms and am wondering if fuel related rather than electrical problems. ( apart from ECU which I can’t check). I am going to check inertia fuel cut off switch, which is apparently is under the dash somewhere.

 

[jonik]

I have found out from multiple cases that if there's no error codes crankshaft position sensor is usually the fault. If ecu cannot be found from obd2 port, it usually helps to undo the ecu connectors and wait a bit, then connect them back

 

[varesecrazy]

Hi

Worth checking the main Battery to Chassis to Motor Earth connection..
The factory cable can fail internally

 

[00Terror00]

Did you measure alternator voltage? of course when engine working

 

[GrandePunto PL]

Classic thread, where you should do real Diagnostics, with capital "D". Not parts swapping like a monkey.
That means oscilloscope (multi channel, at least 4). Multimeter will not help much.
Best example on YouTube: https://www.youtube.com/c/AutomotiveTestSolutions/videos
THIS is how you diagnose cars.

 

[g8rpi]

There are other automotive oscilloscopes. My personal preference is Pico Technology.

https://www.youtube.com/@PicoScopeAutomotive

However despite PicoAutomotive's automation an oscilloscope just shows a waveform it takes knowledge,skill and experience to understand the waveforms.

On the OP's fault. it is most likely something that was disturbed during the work. My bet is with Varescrazy: The battery/engine/chassis cable. Try a jumplead from battery negative to engine block and another from block to chassis.

 

[judderbar]

>>If I wait for a while, or if I disconnect and reconnect the battery, the ECU connection seems to work again and the engine starts fine and runs again for few minutes max. This happens over and over again.

How variable is the amount of time the engine runs once it restarts? Is the time a random amount or is it running consistantly quite a long time?

 

[Fiat1271300]

Does anybody know if a stillo accelerator pedal will work on a punto mk2?

 

[varesecrazy]

Does anybody know if a stillo accelerator pedal will work on a punto mk2?

Unlikely, as not closely related

What happened to your pedal?

 

[Fiat1271300]

I don’t have a spare to keep and I can get one from the scrap yard. I have seen them on eBay so no shortage. AUTODOC have them as well new

 

[Ksport472]

Apparently I only got a notification for the first response in this thread, so I forgot to check this.

I have found out from multiple cases that if there's no error codes crankshaft position sensor is usually the fault. If ecu cannot be found from obd2 port, it usually helps to undo the ecu connectors and wait a bit, then connect them back

I would also suspect that if would only die, but not sure if that should/could cause the ECU to completely "disappear". Unplugging the ECU connectors seem to indeed help also (and the situation when the issue happens is a bit similar to the case if you try to start the car with ECU unplugged). The right side ECU connector alone (when looking from the front of the car) seems to also "reset" the issue.

Hi

Worth checking the main Battery to Chassis to Motor Earth connection..
The factory cable can fail internally

I tried to connect extra ground cables from the battery to the ECU chassis and to the engine body (and I measured and cleaned the ground cable on the throttle body / ecu). Those didn't have any effect. Also I tried to wiggle all the cables I could see, but that doesn't have any effect either. I also checked the main positive cables and cleaned the terminals on the battery and fuse box.

Did you measure alternator voltage? of course when engine working

Yes, although only using a multimeter, which has pretty low sample rate and it might do some internal averaging. Anyways it was showing approximately 14.2 V so that seems to be fine.

Classic thread, where you should do real Diagnostics, with capital "D". Not parts swapping like a monkey.
That means oscilloscope (multi channel, at least 4). Multimeter will not help much.
Best example on YouTube: https://www.youtube.com/c/AutomotiveTestSolutions/videos
THIS is how you diagnose cars.

Well at this point I have swapped zero parts because of this issue, except for few relays. Unfortunately I didn't have an oscilloscope at hand (I have a 2-channel oscilloscope somewhere), so I had to do at least some initial diagnostics using just a multimeter.

There are other automotive oscilloscopes. My personal preference is Pico Technology.

https://www.youtube.com/@PicoScopeAutomotive

However despite PicoAutomotive's automation an oscilloscope just shows a waveform it takes knowledge,skill and experience to understand the waveforms.

On the OP's fault. it is most likely something that was disturbed during the work. My bet is with Varescrazy: The battery/engine/chassis cable. Try a jumplead from battery negative to engine block and another from block to chassis.

I have lots of experience with oscilloscopes (and electronics in general), but sadly I don't have one with me right now. I'll try to find one portable oscilloscope that I have somewhere, but I won't buy a new one solely for this purpose. It would feel bad to buy expensive tools for a 500 € car
There are indeed multiple things that could be measured, multimeter (especially a cheap basic one) is too slow and useless for any proper time domain analysis. Mostly I wanted to first ask for all the common failure points and symptoms to see if this issue could be solved easily. Sadly those extra ground cables didn't help, as said above.

>>If I wait for a while, or if I disconnect and reconnect the battery, the ECU connection seems to work again and the engine starts fine and runs again for few minutes max. This happens over and over again.

How variable is the amount of time the engine runs once it restarts? Is the time a random amount or is it running consistantly quite a long time?

There is a slight variation. Maybe the shortest time has been something like 1 minute and the longest one was 7 minutes (I almost started to celebrate too early when that happened), but usually it is pretty constantly 2-3 minutes. After the issue happens, it takes a while to recover from that unless the battery terminal or ECU connector is removed and attached again.

 

[Ksport472]

From another thread I found a wiring diagram for the ECU: https://www.fiatforum.com/threads/m...diagrams-wanted-urgently.283445/#post-2875469

..but that doesn't seem to match exactly with mine. Some of the wire colors etc. are different, but that is for the model year 1999, so maybe there are some differences in this facelift model. The exact ECU model is IAW5AF (P3) in my case, and that is definitely for an older ECU model.

I did some additional measurements, and when the issue happens, there is no power to the fuel pump or probably anything behind the highlighted relay K46 (no voltage seen at fuse F22 or on the oil pressure sensor plug). However, it sounds like the relay opens AFTER the engine dies (= probably not causing the engine to die), and won't close again until the ECU starts to respond. That relay is controlled via the ignition switch, but probably ECU has also control for that, since it seems to be grounded through the ECU based on that diagram. ECU might be able to latch the pin. The ignition switch continues to supply power regardless of the issue. Although that diagram might not fully apply, because if I remember correctly, the ignition coils are for example constantly grounded, but the diagram indicates that they have a constant +12 V when the ignition is switched on, and the ground side goes through the ECU, which is not the case on my car.

I also happened to have new ignition coils and spark plug wires for the car. I didn't expect them to help, as they didn't, but it was a good time to replace them anyways.

(I wrote a bit longer reply above for the previous comments, but the message is waiting for a moderator approval, so it'll probably be visible soon.)

 

[judderbar]

From another thread I found a wiring diagram for the ECU: https://www.fiatforum.com/threads/m...diagrams-wanted-urgently.283445/#post-2875469

..but that doesn't seem to match exactly with mine. Some of the wire colors etc. are different, but that is for the model year 1999, so maybe there are some differences in this facelift model. The exact ECU model is IAW5AF (P3) in my case, and that is definitely for an older ECU model.

I did some additional measurements, and when the issue happens, there is no power to the fuel pump or probably anything behind the highlighted relay K46 (no voltage seen at fuse F22 or on the oil pressure sensor plug). However, it sounds like the relay opens AFTER the engine dies (= probably not causing the engine to die), and won't close again until the ECU starts to respond. That relay is controlled via the ignition switch, but probably ECU has also control for that, since it seems to be grounded through the ECU based on that diagram. ECU might be able to latch the pin. The ignition switch continues to supply power regardless of the issue. Although that diagram might not fully apply, because if I remember correctly, the ignition coils are for example constantly grounded, but the diagram indicates that they have a constant +12 V when the ignition is switched on, and the ground side goes through the ECU, which is not the case on my car.

I also happened to have new ignition coils and spark plug wires for the car. I didn't expect them to help, as they didn't, but it was a good time to replace them anyways.

(I wrote a bit longer reply above for the previous comments, but the message is waiting for a moderator approval, so it'll probably be visible soon.)

maybe this pinout is nearer? https://www.fiatforum.com/downloads/punto-iaw-59f-m3-ecu-pinout-pdf.405/

But your facelift is sort of halfway towards being a grandepunto? Punto evo?

 

[mcf1lmnfs]

if you open the hood, at the right side you have this fuse box. Take off the black cover, undo the nut on the incoming 12v connection bolt. Now come closer and you should see that this white plastic inside is like a door on a hinge, it can be opened - rotated upwards, there its clipped in place, so to undo the clips you use flat screwdriver, and then it should be possible to rotate it to see connection cables under this. Undo them all, visually check all the pins and you can spray them with WD40 - its perfect for cleaning electrical pins without any chance for harming the electronics or plastic.

From what you told, i dont think its engine immobilizer. But to remove this thought for sure, try getting the car on and go driving immeditly. Does it also die if your car is moving? If yes - it is not immobilizer.

Can you provide more info about how does it die? when it dies does it sound exactly like you would turn it of with a key?
When you get it to reset the issue, Did you try to turn it off with a key and back on, before it dies by itself? Try and report back please, with all you find hear and see.
When your get your issue to reset again, can you turn your key to MAR position (dont start the engine yet) and listen please do you hear your fuel pump make a little noise? It is correct that on every turn of a key from off to MAR you hear eeeeeeeeeeeee sound inside the cabin - this your fuel pump creating fuel pressure. So try multiple times: MAR, off, MAR, off, MAR off, does your fuel pump make the sound every time or only the 1st time?
Then turn the car to on, and immediatly off, and try MAR, off, MAR, off.. does your fuel pump make noise or not? noise it means its still working correctly.
Try running the car until it dies, dont reset the issue and try MAR, off, MAR, off, MAR off, does your fuel pump make noise ? Is it dead now?

Come back please with as much as you can gather on the behaviour of the fuel pump according to above, and hopefully we will get closer to the where the issue is coming from.

No need to check large cables like car battery, alternator, ground on gearbox - its obvious they are perfect if your starting motor is turning - even if the engine doesnt start.

Can you also time with timewatch how long does the car stay in idle. I suggest following 3 experiments:
1. Start timing when you reset the issue, go inside the car, start the car, mark the time, when it dies mark the time again, Do 5 repeats of this, and please report back with all 10 times. Also try to have in this 1st experiment try to have as similar between 0 and 1st mark of time between reapats as possible.
2. In this experiment try to have different times from 0 to 1st mark for all of 5 repeats. Lets say you try 10sec, 1min, 3min, 5min, 10min. So now you know when to start the car, now only mesure how long does it stay on.
During 1st and 2nd experiment do not touch the gas padal or clutch or brake or gearbox - all of these could change the fuel consumption a little and the experiments would have no purpuse - keep it same in all of the repeats.
3. Again as in 1st exp. try to have same times between 0 and 1st mark, and then when the car is started go ASAP to 5000rpm, no cloutch, no touching gearbox at all (out of gear obviously). and mesure 5 repeats here too.

So here we have 20 numbers, please take notes, and report them back here.
With this experiment we will see how repetable the timing of the issue is and if it may depend on fuel consumption or rpm, which could lead to easier identification of the problem.

If you solve the issue please report back the fix and what it was.

 

[Ksport472]

maybe this pinout is nearer? https://www.fiatforum.com/downloads/punto-iaw-59f-m3-ecu-pinout-pdf.405/

But your facelift is sort of halfway towards being a grandepunto? Punto evo?

Hmm at least the ignition coil wire colors are not the same. Although I have only a picture of one ignition coil at hand, because I'm not by the car currently. The coil has two wires and one is black/yellow (I think it was the ground wire) and the other one is red/orange (or maybe red/pink, hard to tell from the picture). I could still check if the pinout is matching.

if you open the hood, at the right side you have this fuse box. Take off the black cover, undo the nut on the incoming 12v connection bolt. Now come closer and you should see that this white plastic inside is like a door on a hinge, it can be opened - rotated upwards, there its clipped in place, so to undo the clips you use flat screwdriver, and then it should be possible to rotate it to see connection cables under this. Undo them all, visually check all the pins and you can spray them with WD40 - its perfect for cleaning electrical pins without any chance for harming the electronics or plastic.

I already had the +12 V bolt nut open on the fuse box when I cleaned the battery cable connectors from both ends, and also I tried to re-attach the connectors and wiggle the connectors when the engine was running. It didn't make any difference. I didn't spray them yet with anything, because they looked pretty clean.

From what you told, i dont think its engine immobilizer. But to remove this thought for sure, try getting the car on and go driving immeditly. Does it also die if your car is moving? If yes - it is not immobilizer.

The car has been sitting on a lift for the whole time and I haven't tried to actually drive it after the issue. Now the outside temperature is between -20 and -30 C so I wouldn't like to take a risk by bringing it outside of the garage if it happens to die completely in the snow. Maybe I could try to move it back and forth at least a little.

Can you provide more info about how does it die? when it dies does it sound exactly like you would turn it of with a key?
When you get it to reset the issue, Did you try to turn it off with a key and back on, before it dies by itself? Try and report back please, with all you find hear and see.
When your get your issue to reset again, can you turn your key to MAR position (dont start the engine yet) and listen please do you hear your fuel pump make a little noise? It is correct that on every turn of a key from off to MAR you hear eeeeeeeeeeeee sound inside the cabin - this your fuel pump creating fuel pressure. So try multiple times: MAR, off, MAR, off, MAR off, does your fuel pump make the sound every time or only the 1st time?
Then turn the car to on, and immediatly off, and try MAR, off, MAR, off.. does your fuel pump make noise or not? noise it means its still working correctly.
Try running the car until it dies, dont reset the issue and try MAR, off, MAR, off, MAR off, does your fuel pump make noise ? Is it dead now?

It just dies instantly and I think the sound is more or less similar to the case if I turn it off from the key. Although logically when the key is used, the fuel pump relay (other things are also behind this relay) is released immediately, but in this case I hear a relay click (not 100% sure if it's that specific relay) maybe after 1-2 seconds after the engine dies (it is probably caused by the ECU, since the ignition switch is still supplying power to the relay, but the other terminal of the relay is probably controlled by the ECU, intentionally or unintentionally). The engine is not running badly before that or anything, so most likely the spark just instantly stops, because it probably would run a bit differently if only the fuel supply is cut. If I hold the RPM higher, the tachometer might "freeze" when the engine dies and stay up for a couple of seconds. Most likely the instrument cluster loses the connection also to the ECU. Usually the ABS and brake lights start to very slowly flash on the instrument cluster, probably indicating a malfunction due to the lost connection. If I turn the key off and back to the MAR position when the issue is active, the yellow immobilizer key icon might show up, but that most likely indicates a connection loss to the ECU, because the symptoms are similar if I disconnect the ECU connectors and turn the key to the MAR position (or even try to start the engine without ECU, which obviously fails).

When the issue is active, I cannot hear the fuel pump running or the relay clicking (but I measured that the ignition switch still provides +12 V to the relay box in MAR position). If I turn they key off and to the MAR position, I cannot hear the fuel pump or relay clicking either. When the issue is not active, I can hear the the relay closing and the fuel pump starting (I think it happens every time when I turn the key to the MAR position -> off -> MAR -> off -> MAR, but I'll doublecheck that). If I start the engine and turn it off using the key before the issue triggers, I can start it again and I hear the fuel pump running. The car seems to work normally (I can switch engine on and off etc.) until the issue kills the car for a while.

I can try to investigate and log those other things more carefully when I have time.

 

[mcf1lmnfs]

About the fuse box, rotate it to get to many more connectors under it, as described in previous my reply. This 1st.

I assmue the communication can fail either because of failiure on body computer or the ecu... Ecu is usually good protected from water intrusion so give 1st look to how this fuse box looks from underneath, and wiggle all of those connectors while car in idle. 2ndly go check your body computer and all its connections - this the module holding the fuses inside your cabin.

If nothing is found, it might be needed to replace body computer, but it cant be just switched with another, some data should be copied from this to the other. And same goes for ecu if new one will be needed.
I cant say much here since such difficult problem never came to me.

I think its pointless to look at other components than ecu and body computer, if there would be a problem with any other these two should record and report the error. So because it doesnt i assume only these two should be looked at. And maybe communication lines, can high, can low.

 

[Ksport472]

I took few time measurements of the issue. The car has been sitting again for a while with the battery disconnected. After I connected the battery and started the car, it ran for 9 minutes until it died. After that I took few measurements from the point of "resetting" the issue by reattaching the negative battery terminal

Time from the battery cable

Battery attachment -> engine start	Battery attachment -> Engine death	Total engine runtime
23 s	1 min 42 s	1 min 19s
22 s	1 min 13 s	51 s
24 s	11 min 52 s	11 min 28 s

The third attempt was probably the longest time that the engine has run with this issue. After that I played around with some connectors and tried again getting a 5 min 22 s runtime. So there seem to be some variance, maybe a bit more than earlier, but it still consistently happens. There were no fault codes after the tests. It started to get so cold inside the garage because I had to keep the door open, so I didn't do more test rounds.

Come back please with as much as you can gather on the behaviour of the fuel pump according to above, and hopefully we will get closer to the where the issue is coming from.

If you solve the issue please report back the fix and what it was.

I verified this and when the issue is not active, the fuel pump starts every time when I turn the key from off -> MAR (also every time if I turn the key repeatedly off -> MAR -> off -> MAR, and so on). When the issue is active, the fuel pump (and most likely some other components also) never starts when turning the key to MAR.

About the fuse box, rotate it to get to many more connectors under it, as described in previous my reply. This 1st.

I already did that earlier. Now I tried to again disconnect every connector under the fuse box and cleaned them a bit, but nothing happened. I have tried also to wiggle all of the fuse box connectors when the engine is is running.

I assmue the communication can fail either because of failiure on body computer or the ecu... Ecu is usually good protected from water intrusion so give 1st look to how this fuse box looks from underneath, and wiggle all of those connectors while car in idle. 2ndly go check your body computer and all its connections - this the module holding the fuses inside your cabin.

If nothing is found, it might be needed to replace body computer, but it cant be just switched with another, some data should be copied from this to the other. And same goes for ecu if new one will be needed.
I cant say much here since such difficult problem never came to me.

I think its pointless to look at other components than ecu and body computer, if there would be a problem with any other these two should record and report the error. So because it doesnt i assume only these two should be looked at. And maybe communication lines, can high, can low.

Yeah, probably the issue is with the ECU or the body computer. I have experience with ECU reading, flashing and custom remapping, but sadly the tools I have don't support this ecu so well that I could virginize the immobilizer data from the ECU. Used ECUs are pretty cheap for these (like 50 €), but it costs a bit more to buy a plug-and-play virginized one. Not terribly expensive, but something between 120-150 € including shipping. I haven't investigated the BCM coding thing yet though, if that happens to be the culprit instead of the ECU.

Oh and one thing came into my mind. Earlier I said that I can hear a relay click only after the engine dies, but at least now I heard a click right before the engine died and then another click right after that. Not sure which relay the first click was. It is hard to say if I heard correctly earlier.

Sadly I still couldn't find my oscilloscope to be able to do better measurements. Let's see if I can find or buy some better diagnostic tools to dig more info about the issue. Or maybe I should just buy a virginized replacement ECU and try with that straight away.

 

[mcf1lmnfs]

These time mesurments are very nice. As i see them, they tell me all your mechanical parts are okay, so no need to look at fuel pump or any other mechanical component. I am also pretty sure that your ecu and body computer are ok since the car works at least some time.
I think you might have one loose connector and vibrations make ecu and body computer to loose comunication lines eventually. There should be good can high and can low connection between them. I am attaching pictures of your body computer - you need to unscrew 3 bolts to get it out, and check the condition of all connectors..
The body computer can get faulty if one of the high current connectors is getting loose, like the attached picture shows, then some stuff can get burnt and this should be obvious if you get it out..
if this is not the case, then your body computer is ok, and probably only some low current communication lines are weak, which should be possible to repair with some cleaning an maybe a little adjustment on female pins.
With oscilloscope would be quite easy to catch the problem, also vcc and gnd are to be monitored.
Btw, they dont tell you any error because this loosy connectors were never meant to happen and if they do, the circuit gets confused and program running on it freezes,.. newer cars has better circuits and better programs which are able to tell you CAN communication errors.. but still if this happens you need to find loosy wires on your own.

Funny thing about loosy connectors is that they are usally temperature dependant, like winter/sunmer, they usually work in one better than in the other.

Very small chance is that connectors are all ok but there might be one fault on the circuit of body computer, here i suspect one soldering blob might be broken,.. this can be repaird if you take body computer circuit out od the casing (this green plate) and then you resolder all components by hot Air soldering gun (look up for this tool, and at least one video about using it), in this case i would go over all components, heating all parts until the solder melts.. vibrations in cars are known to be strong enough to be able to break soldered connections, this is why on wires the crimping is always used, and never soldering,. but on pcbs this issue is very rare, close to non-existant.
Anyway it pays to check, maybe there is some other thing that happend, maybe the car was flooded and there is mud iside body computer - half of year ago, we had some cases like this.

 

[Ksport472]

These time mesurments are very nice. As i see them, they tell me all your mechanical parts are okay, so no need to look at fuel pump or any other mechanical component. I am also pretty sure that your ecu and body computer are ok since the car works at least some time.
I think you might have one loose connector and vibrations make ecu and body computer to loose comunication lines eventually. There should be good can high and can low connection between them. I am attaching pictures of your body computer - you need to unscrew 3 bolts to get it out, and check the condition of all connectors..
The body computer can get faulty if one of the high current connectors is getting loose, like the attached picture shows, then some stuff can get burnt and this should be obvious if you get it out..
if this is not the case, then your body computer is ok, and probably only some low current communication lines are weak, which should be possible to repair with some cleaning an maybe a little adjustment on female pins.
With oscilloscope would be quite easy to catch the problem, also vcc and gnd are to be monitored.
Btw, they dont tell you any error because this loosy connectors were never meant to happen and if they do, the circuit gets confused and program running on it freezes,.. newer cars has better circuits and better programs which are able to tell you CAN communication errors.. but still if this happens you need to find loosy wires on your own.

Funny thing about loosy connectors is that they are usally temperature dependant, like winter/sunmer, they usually work in one better than in the other.

Very small chance is that connectors are all ok but there might be one fault on the circuit of body computer, here i suspect one soldering blob might be broken,.. this can be repaird if you take body computer circuit out od the casing (this green plate) and then you resolder all components by hot Air soldering gun (look up for this tool, and at least one video about using it), in this case i would go over all components, heating all parts until the solder melts.. vibrations in cars are known to be strong enough to be able to break soldered connections, this is why on wires the crimping is always used, and never soldering,. but on pcbs this issue is very rare, close to non-existant.
Anyway it pays to check, maybe there is some other thing that happend, maybe the car was flooded and there is mud iside body computer - half of year ago, we had some cases like this.

Today I took out every fuse and relay from the engine bay fuse box (I took a picture beforehand so I could easily put everything back) and opened it to see if there is anything strange inside, but it looked fine. No corrosion/oxidation could be seen. Then I did the same for the fuse box that is connected to the BCM, but nothing there either. I also opened the BCM and there were no signs of moisture, cracked solder joints or anything unusual. There was a small amount of residue around the electrolytic capacitor, but most likely it is just flux residue. I assembled everything back to their places and now when I turn the key the rear wiper starts to immediately wipe, the fuel pump starts running and some relay is continuously clicking on the BCM fuse fox. Nothing else happens anymore

Maybe there were some solder joints that were almost broken already on the BCM or then some of the wires/connectors are damaged (cannot see anything though). It is worth to mention that maybe a week ago I noticed an aftermarket remote control box for the central locking attached on the backside of the BCM. I've had symptoms that doors might lock when I'm braking. I removed the box since I don't have any remotes for that anyways, but it didn't have any effect on this issue. Today I removed the wires and the splice connectors (I taped the sliced insulations, but otherwise the wires seemed to be intact enough) from the original wiring harness when I worked with the BCM. Moving any of the wires and connectors didn't do anything either when this new issue is active.

I can try to check the solder joints on the PCB. I have access to a hot air soldering gun, so I could try with that and maybe go through those more beefy connectors with a sodering iron to avoid heating other components too much. If nothing helps, maybe I'll need to replace the BCM. I think I need to read the content from the EEPROM chip from the old board and write it to the new board. I can also transfer the old chip to the new board if needed, which could be even easier.