14 minutes ago, Abra said:

At the time of measurement, it was about 0.2-0.3 volts. I will remeasure now.

 

15 minutes ago, Abra said:

Now I will record a video from the inverter screen starting from the moment the voltage is applied from the PV with the power button immediately turned on

 

23 minutes ago, Coulomb said:

16 - 4.915
15 - 4.913
14 - 4.913
13 - 11.991
12 - 11.993
7 - 11.987
6 - 11.988
5 - -12.059
4 - -12.056

15 minutes ago, Abra said:

Now I will record a video from the inverter screen starting from the moment the voltage is applied from the PV with the power button immediately turned on

I see no output voltage. Is this the same problem as half way down page 1, where the output voltage increases?

I'll assume not. So this is powered with PV only, no battery, with a small load connected? And the problem is no output voltage, and very low PV power? And there is enough PV power available, as proven with another inverter in similar circumstances?

Finally, the warning code 15 (PV energy is low, they must mean PV power) comes on at the end. Has it always done that? I assume that the good inverter in similar circumstances does not show that fault.

With the first video, the gate drive is well under 0.1 V for most of the video, and goes to solid zero at the end. What coincides with that gate drive going to zero? Is it the appearance of warning code 15?

10 minutes ago, Coulomb said:

With the first video, the gate drive is well under 0.1 V for most of the video, and goes to solid zero at the end. What coincides with that gate drive going to zero? Is it the appearance of warning code 15?

No, this happens every time I unhook the tester probe and reattach it. It feels like some kind of capacitor accumulates something insignificant, and when I connect the multimeter, it gives it back to it.

13 minutes ago, Coulomb said:

I see no output voltage. Is this the same problem as half way down page 1, where the output voltage increases?

It does not run from PV. If a working inverter, when connected only to PV, after 5 seconds after switching on, clicks the relay and starts supplying power to the output, just at this moment the bus switches from 240 + v to 360 + v, then the broken one does not start the transistors, does not start the relay and does not supply voltage to the output. I wrote a few posts earlier about 3 stages of launch where a broken one can only go through 2 of them.

On 12/23/2022 at 9:59 AM, Abra said:

So:
1 - I turn on the power from the solar panels / 241 volts
2 - Inverter turned on, warning bP! alternates with HS / 241 volts
3 - There is a click of a relay and a voltage of 364 volts at the same time, the system is up and running.

On broken:
1 - I turn on the power from the solar panels / 241 volts
2 - Inverter turned on, warning bP! alternates with HS / 241 volts
3 - There is no relay clicking, the voltage is still 241, a warning 15 appears!

This

19 minutes ago, Coulomb said:

I'll assume not. So this is powered with PV only, no battery, with a small load connected? And the problem is no output voltage, and very low PV power?

Nothing is connected to the outputs of the inverter. Only 2 wires are connected to the inverter: - and + from PV. I carried out a similar test on a working inverter. It turns on and starts the circuit.

21 minutes ago, Coulomb said:

Finally, the warning code 15 (PV energy is low, they must mean PV power) comes on at the end. Has it always done that? I assume that the good inverter in similar circumstances does not show that fault.

Yes, a good inverter just starts working with only PV, it has enough power, because his transistors start up and 360+V starts to flow to the bus.

We've been talking about the SCC board, the control board and the main board, but there is also the IND (INDctor) board, with only one major component on it: the boost inductor. This is crucial for sending any power through the Solar Charge Controller on high voltage models.

Can you swap the IND boards between working and faulty inverters? Or at least check that you get a low resistance (well under one ohm) across P2 and P3? If this inductor is open circuit, short circuit, or there are shorted turns, or the core is cracked leaving even a millimetre of air gap, this could be the problem. Top left of this photo, with the blue and brown wires leading to it:

25 minutes ago, Coulomb said:

We've been talking about the SCC board, the control board and the main board, but there is also the IND (INDctor) board, with only one major component on it: the boost inductor. This is crucial for sending any power through the Solar Charge Controller on high voltage models.

Can you swap the IND boards between working and faulty inverters? Or at least check that you get a low resistance (well under one ohm) across P2 and P3? If this inductor is open circuit, short circuit, or there are shorted turns, or the core is cracked leaving even a millimetre of air gap, this could be the problem. Top left of this photo, with the blue and brown wires leading to it:

Yep, I tried to replace it too Now installed this one:

 

I will probably try to measure and compare between a working and a broken inverter the readings of all sensors that go to CN9 in the state when power is supplied from the solar panels, but the button on the inverter is not turned on and in the state when the button is already on. As I understand it, the sensors go in pairs to the CN9 pin? 1,2 - 3,4 - 5,6 etc. Except CN9.7, CN9.8 and CN9.15, CN9.16

23 hours ago, Coulomb said:

If you like, but I don't think it's productive. The control board isn't telling the transistors to conduct. And the problem has to be on the main board.

According to my logic, the whole system works like this:

1 - I am powering the PV.
2 - In all the necessary circuits for the life of the inverter, voltage appears only from PV, without the transistors turned on, which must raise the voltage to the system required for full operation, let's say 360v.
3 - It also goes through the blue cube on the SCC board, which measures the power consumption. At this point, it measures the amount of current consumed by the inverter components. I'm not sure about its accuracy at low current consumption, so my opinion is that those 15-80 watts that it shows are not entirely correct, but at the same time they tell us that the blue cube is working.
4 - Next, the system checks for some time whether everything is in order, receiving information from different sensors and most likely, having rechecked all the necessary information several times, it turns on without errors (only with a warning, BP!, because it could not detect sufficient voltage on the BAT+, BAT- contacts)
5 - Next, after I press the power switch (as it happens on a normal inverter), the system starts doing some more checks for 5 seconds. After the system makes sure that everything is in order - the control board sends some voltage to some pins of the SCC board to start the driver for the transistor, which in turn will start the transistors, which in turn will start to help convert the 240v voltage to the required 360v.
6 - Immediately after the start of the transistors, when the control board receives data from the sensor that there are 360v on the bus, the control board starts the remaining components, which in turn direct the voltage from the bus to the output.

Why am I all this. As I understand it, there is a problem here "the control board sends some voltage to some contacts of the SCC board", which is why the transistors do not turn on. Accordingly, I want to understand where exactly the voltage should go to the driver board in order to check whether it is there in principle. And if it is not there, trace the contacts on the control board from where it gets to the SCC board, if there is no necessary voltage there, trace its path on the control board, etc.

3 hours ago, Abra said:

Except CN9.7, CN9.8

These might be a sort of pair, with the negative end actually being GND.

3 hours ago, Abra said:

and CN9.15, CN9.16

Yes. CN9.15 is an output, and I don't know what if anything is on CN9.16.

The others are in pairs as you say, with + on the odd numbered pins.

On 2022/12/31 at 11:29 AM, Coulomb said:

These might be a sort of pair, with the negative end actually being GND.

Yes. CN9.15 is an output, and I don't know what if anything is on CN9.16.

The others are in pairs as you say, with + on the odd numbered pins.

Hi all. Unfortunately I was very busy and had to take a break. Another inverter broke today... With the same problem as we discussed. Opened it for the first time. And I started juggling SCC and Contol boards. And I noticed such a thing, for some reason I had not paid attention to it before. The left air duct of the inverter, or rather of all inverters, with its lower part, falls on part of the control board. And there accumulates the maximum layer of dust and dirt. Also, apparently due to high humidity, let me remind you, I live in Thailand, along with this dust, moisture gets there and begins to literally corrode the control board. The coating of the board (paint) begins to swell, as it were. After a total cleaning with acetone, this piece of the board looks like this. One of the boards was also cleaned with WD-40, but I had literally 1 drop left, so I cleaned it poorly.

All in all. The point is. The last inverter after cleaning the board started working (all functionality) but not for long. In principle, all the functionality works for it, but sometimes charging or power from PV stops (the inverter shows 0W, 0A, 240-250V). Several times I just popped the control board out after the charging stopped going, clean it with acetone and a toothbrush again, put it back and everything started working for a while.

This problem now exists on all 4 inverter control boards that I have. 2 boards do not charge in any way. 1 board either charges or not. 1 board is still charging, but it can be seen that the incoming current began to sometimes jump and be unstable. I think soon the board will also fail.

Can you tell me how and what should I check and how can I fix it?

First board:

Second board:

Wire voltage comparison between Control Board and SCC Board (8pin) (Left - Working, Right - Broken)
1 - ground.
2 - 12.5 => 12.4
3 - 13.0 => 13.2
4 - 17.0 => 16.9
5 - 12.4 => 12.0
6 - 18.5 => 23.9
7 - 23.9 => 23.9
8 - 12.0 => 12.0

Edited January 13, 20233 yr by Abra

2 hours ago, Abra said:

The coating of the board (paint) begins to swell, as it were. After a total cleaning with acetone, this piece of the board looks like this.

Wow. I've never seen anything like it. I'm thinking that acetone was not the right thing to clean it with; too strong of a solvent. It's possible that you've ruined these boards, but it might be possible to recover them. I would clean them again, carefully, in something way less strong, like alcohol (here it's called metho, for methylated spirits, basically ethyl alcohol with poison in it to stop you drinking it). Let it dry thoroughly (should not take long), then spray liberally both sides with a conformal coating, highly preferably one that is designed for protecting printed circuit boards. This will hopefully prevent stray currents on the board from upsetting the measurements, and possibly the operation of the board. Other boards may also need this treatment.

Edit: But was the solder resist coating bubbling and coming off before the application of the acetone?

Edited January 13, 20233 yr by Coulomb

3 minutes ago, Coulomb said:

Edit: But was the solder resist coating bubbling and coming off before the application of the acetone?

Yes, the coating was bubbling. I removed most of the paint with just a toothbrush without acetone. Then I additionally rubbed with acetone stronger to wipe off the dirt and remnants of peeling paint.

6 minutes ago, Coulomb said:

spray liberally both sides with a conformal coating, highly preferably one that is designed for protecting printed circuit boards

And you can throw off an example of coverage, I'll try to find something similar here. I have not come across this yet Therefore, I don’t know what exactly is needed for coverage

 

Do you think PV may not be charging due to the fact that there is simply no coverage?

I ordered this one, I hope I did everything right

https://www.lazada.co.th/products/relife-rl-uvh900-rl-uvh900b-rl-uvh900bl-curing-solder-mask-ink-10cc-i1979182474-s6335332797.html

UPD:

1 column: Working one (Shows around 2.5kW on input PV when measure)

2 columb: Broken one (Shows 0 watt on input PV when measure)

3 columb: Broken one (Shows 40-50 watt on input PV when measure)

4 columb: Broken one (Shows 0 watt on input PV when measure)

 

1 - ground.
2 - 12.5 => 12.4 => 12.4 => 12.6
3 - 13.0 => 13.2 => 13.2 => 13.3
4 - 17.0 => 16.9 => 16.9 => 17.0
5 - 12.4 => 12.0 => 12.0 => 12.0
6 - 18.5 => 23.9 => 23.9 => 23.8 !!! this pin have a big difference between working one and all brokens.
7 - 23.9 => 23.9 => 23.9 => 23.9
8 - 12.0 => 12.0 => 12.0 => 12.0

UPD:

1 column: Control board original from Working one

2 column: Control board original from semi working one

3 column: Control board from 3rd broken invertor

4 column: Control board from 4th broken invertor

 

So what's the point of all this.

I went up to the inverters now, to see what works, what doesn't.

And I see this. Both inverters are charging, although the control board from the non-working inverter is installed on the half-working inverter. Payment from column 4.

I'm starting to measure the pins. And what is happening. When measuring pin 6, charging immediately stops, the pin shows 23.8v. After 2-3 minutes, charging continues again as if nothing had happened. I'm trying to measure this pin again. Charging stops again, then resumes again, and so on ad infinitum.

On the worker, when measuring, charging does not stop.

Here is an interesting behavior associated with pin 6 I noticed. Not only does it show different voltages with a working inverter, but also when trying to measure, if charging suddenly starts, charging stops immediately. I understand the somewhere problem is here.

1 hour ago, Abra said:

you can throw off an example of coverage, I'll try to find something similar here.

This is what I use:

No doubt there are many bands and variations. It is touch dry in about 5 minutes, usable in perhaps 30 minutes.

1 hour ago, Abra said:

Do you think PV may not be charging due to the fact that there is simply no coverage?

It seems possible, and you might well be chasing shadows until the corrosion / leakage issues are all fixed.