The story begins…
The day after April Fools’ Day 2020 was not a funny day for me 🙁
That day my sleep was interrupted by a constant beep from the garage several minutes after 5 in the morning. I woke up from bed and go through complete blackout to the inverter.
There was only an annoying beep in my ears and a failure red diode with error code 09 on the front of the inverter. I also smelled a fried semicondutor in the air… First thing which comes to my mind was that maybe there was some overload so I stupidly tried to start the inverter up for several times, I also discovered that main overcurrent protection for the AC-input was triggered down.
Before I shut all things down I only did this photo:
Then I disconnected the inverter and created a bypass to have the power back on all devices…
Visual inspection
I removed the inverter from the wall and take out the cover. And this is what I saw:
From the visual inspection I could see that the MOSFETs was broken and also some traces + resistors was also broken on the bottom of the main board.
I decided to start googling if someone has similar failure and one of the first link was the AEVA Forums which was known to me.
There is a inverter repair topic which I read from the first to last post:
PIP inverter repairs and hardware modifications
My main source of information was the above thread and also a service manual PDF for similar inverter. There was some differences but mainly it was very similar to mine.
I am dropping this manual here for the reference:
Copy_of_PIP-HS_MS_4-5KVA_new_Service_manual_201506A.pdf
Diagnosis
My inverter all the time was running at about 25-30% load, no PV attached ever. It blow out almost 5 years from production date. The environment is hard: during summertime it is hot in the garage and parking a hot car doesn’t help either. Besides my electronic hardware, my fridge, freezer and water pumps (during winter) was also connected to the inverter output.
Nevertheless after some discussion it seems that the main problems was the drying out capacitors which could not filter out some spikes at some point and this can lead to broke the MOSFETs.
The plan and repair
Firstly my plan was to change the MOSFETs – the service manual advices to replace all MOSFETs even if only some was broken. There was also two resistors which I need to replace.
After additional measurements it turns out that also two of the IGBT was broken (QB2 and QD2):
To sum it all, here is the complete list what I’ve replaced in my case:
- 4x 4200uF/35V capacitors, replacement: EKZN500ELL332MM40S
- 16x IRFB3206 MOSFETs, replacement: IRFB3206PBF
- 1x SMD 200 Ohm resistor (in the MOSFET driver), replacement
- 1x SMD 10 Ohm resistor (in the MOSFET driver), replacement
- 2x GP4063D IGBTs, replacement: chinese clone from ebay
Thanks God, so far it is working fine…
Big thanks to weber and coulomb from the AEVA Forums!
I’m glad to see you found 10,000 h rated capacitors, even though they were over £3 each.
For other readers, I note that he replaced 60 V MOSFETs with the same. This is fine for a 24 V model, but owners of 48 V models should replace the 75 V or 80 V originals with 100 V parts.