Power-One solar Inverter driving me NUTS!!!

[mitsu2k]

something strange happening in my last reply .. second link again here .its just an example and not saying buy from there
https://allegro.pl/oferta/zrodlo-odn...69-12780645910

No worries Pete, I got it on my shopping list. Thank you again!!

[mitsu2k]

the chinese translates to . You must report any pirated pictures. Any problems in pirated stores.

L O L, I just snapped that picture off the web because it was exactly like my component and a better picture of it

[petehall347]

something strange happening in my last reply .. second link again here .its just an example and not saying buy from there
https://allegro.pl/oferta/zrodlo-odn...69-12780645910

[petehall347]

REF3033
https://www.s-manuals.com/smd/r3



https://allegro.pl/oferta/zrodlo-odniesienia-3-3v-ref3033-r30d-sot-23-4069-12780645910

[petehall347]

the chinese translates to . You must report any pirated pictures. Any problems in pirated stores.

[mitsu2k]

Anyone know what this is?

[mitsu2k]

Awesome! You found more damage! Maybe circle the area in a picture where you found them. I was thinking about the output circuitry of the CT. After all it goes into a fluxgate, a driver and a differential amp, plus a reference voltage, That does explain at least a part of the GFI instant error, but you would still have a GFI fault problem with these voltages you measured against earth ground. Keep in mind the location of these little components.

I will get that photo to you shortly. Please forgive the less than optimal quality of the photo. I sent my phone out for Repair so I'm using my spare old phone right now. But just to let you know, I found several diode‘s, Zener diode‘s, and those 351A MOSFETs that were shorted in that general area.I think I have identified them all and hopefully I'll be placing an order with Mouser today after I check other components

[CapLeaker]

Awesome! You found more damage! Maybe circle the area in a picture where you found them. I was thinking about the output circuitry of the CT. After all it goes into a fluxgate, a driver and a differential amp, plus a reference voltage, That does explain at least a part of the GFI instant error, but you would still have a GFI fault problem with these voltages you measured against earth ground. Keep in mind the location of these little components.

[mitsu2k]

Anyone know what this is? It is a SOT-23 package I believe.

[mitsu2k]

Now this might be real big issue with why you are getting this fault light if this is the part of the circuit that does this function
But this is what you need to be looking for any type of visible damage like that this could help you a lot

I would agree that if you are getting a GIF error light there has to be some voltage finding it way to ground or the circuit that is suppose to be measuring the voltage leakage is not functioning correctly now the question is what part of circuit is not functioning correctly and where on the circuit board is fault is actually taking place and like what was mentioned earlier not having a circuit diagram for this inverter it is going to be difficult at a minimum but not impossible if you have the background in these types of circuits

google tear down of this inverter on YouTube and see if there any that would be worth looking into first and you might get lucky and someone else has had this issue before or find another inverter just like yours on eBay and look for “not working for parts”
hopefully it will not be the same exact issue that you are having with your inverter beyond that I do not know what else to suggest to you

I have done this in the past when I have had a device that I really like and it malfunctions and I want to repair it I look for the exact same model and I make sure that the boards look the same or as close as possible it makes it easier if can find the same exact one like the one you have

Sam, I agree with you 100%. My mistake. I should have visually inspected all boards under the microscope. The burned components were actually on the logic board where I focused almost no attention. Lesson learned and I will definitely pay more attention next time. Thank you all for helping me learn. I'm still not done but I am cleaning up all the goop and trying to find some kind of signs to lead me to the part numbers on those components. There are at least 3 zener diodes and that 351A part. Making good progress. I will post more photos after cleanup. Thank you again. You and CapLeaker and Peter have been great!

[mitsu2k]

I think I may have found a twin. What do you guys think?

[sam_sam_sam]

Now this might be real big issue with why you are getting this fault light if this is the part of the circuit that does this function
But this is what you need to be looking for any type of visible damage like that this could help you a lot

I would agree that if you are getting a GIF error light there has to be some voltage finding it way to ground or the circuit that is suppose to be measuring the voltage leakage is not functioning correctly now the question is what part of circuit is not functioning correctly and where on the circuit board is fault is actually taking place and like what was mentioned earlier not having a circuit diagram for this inverter it is going to be difficult at a minimum but not impossible if you have the background in these types of circuits

google tear down of this inverter on YouTube and see if there any that would be worth looking into first and you might get lucky and someone else has had this issue before or find another inverter just like yours on eBay and look for “not working for parts”
hopefully it will not be the same exact issue that you are having with your inverter beyond that I do not know what else to suggest to you

I have done this in the past when I have had a device that I really like and it malfunctions and I want to repair it I look for the exact same model and I make sure that the boards look the same or as close as possible it makes it easier if can find the same exact one like the one you have

[mitsu2k]

Exactly what I expected to happen. Made no difference. A Megger properly used is hard to beat. If I Megger test let's say a submerged pump that runs on 240v, I am pretty damn sure that thing won't run at 1MOhm. I think your better off to test this Inverter on the bench, with an isolation transformer for the PSU's and see where the voltage come from by strategically taking components in and out in certain places and go from there. You don't have much of a choice left anymore.

OMG! I think I found something on the logic board. I was trying to trace the output of the CT to the logic board and I think I may have found the issue. I don't think it was a true ground fault at all. Look at this photo if these two burned components. One diode and one transistor. Omg how am I going to identify these parts? They are burned. I can read the first two characters looks like a “35” maybe. Now look g to see if I can find another one on the board like it that I can read.

[CapLeaker]

Exactly what I expected to happen. Made no difference. A Megger properly used is hard to beat. If I Megger test let's say a submerged pump that runs on 240v, I am pretty damn sure that thing won't run at 1MOhm. I think your better off to test this Inverter on the bench, with an isolation transformer for the PSU's and see where the voltage come from by strategically taking components in and out in certain places and go from there. You don't have much of a choice left anymore.

[mitsu2k]

Now if I can only figure out which one is the embedded controller. I will try and clean off the goop and see if I can get some decent photos

[mitsu2k]

Ok I just did the test again differently. This time I jumpered to a ground pin of another IC (not earth ground). This time I got a reasonable voltage reading from the CT devices.

on the DC side I got as follows:
CT #1 = 2.354 vdc
CT #2 = 2.352 vdc

on the AC side as follows:
both read 2.464 vdc

does that indicate the CT devices are working properly?

there was no AC connected during this test. I don't think having the AC/grid connected would make any difference anyway due to there not being any current on that side seeing that it is shut down due to the GFI error.

CapLeaker does this information help with comparing the CT devices?

by the way, I did jumper these pins to ground to no avail

[CapLeaker]

That's why I said before that the Megger test should have shown it. If you want, you can take that CT out or swap the 2 CT's on the DC input side where you measured the lower MOhm. And I say nothing is going to change when you redo the Megger test. You could compare them 2 CT's to each other and see if they measure any different.

To me it doesn't make much sense on what the other guy is saying, because the inverter goes into an instant GFI fault before or at the same time the thing is booting up. Voltage is there yes, but no current is drawn. The only idea that comes about is what if the driver or the differential amplifier after the CT output is shot?

[mitsu2k]

I grounded the output pins of the CT devices to "earth ground" and still no change. I decided to test the voltage of the output pins. Red probe to pin and black probe to earth ground and got -78V. Not sure what is going on here as the data sheet for these devices state output voltage is 4.65V max.

https://cdn.badcaps-static.com/pdfs/...5921ef33dd.pdf

[mitsu2k]

Why not marking the 4 CT's, take them off the board and measure them against each other?

Do you think the CT themselves could be the problem?

[CapLeaker]

Why not marking the 4 CT's, take them off the board and measure them against each other?