Good day,

I am new here. My system consists of 1 x Axpert (type) 5kVA 4KW inverter, 6 x 285W panels and 1 x Pylontech 3.5kW Lithium iron phosphate battery

The problems:

- Sometimes, very early in the morning, even before the sun is out,  the inverter detects the solar panels. Sometimes, even on sunny days, it takes until 11:00 in the morning to detect. When I test the voltage over the panels in total, it tests between 98V and 103V, but it still doesn't detect. 

- After I changed from Lead acid batteries to Li ion, it still struggles to detect the panels early in the morning and when it finally detects the panels, it disconnects from them frequently, still showing the panel sign on the LCD screen, as in the attached picture. (in the picture, it shows 76V, but that was when I experimented with configuration). Later during the day, though it gets better and the battery eventually starts charging.

During my investigation, I was testing the panel voltage with my multi meter and it showed no drop or rise in voltage. It was consistently around 100V, varying by only 1 or 2 V. The wattage was also over 1kW all the time.

Also, from the picture, is it only my imagination, or did they sell me a clone inverter. They advertised and invoiced it as an "Axpert type" inverter. Can I make things warm for them?

Thanks for your help

Reinhard 

7 minutes ago, Reinhard said:

Sometimes, very early in the morning, even before the sun is out,  the inverter detects the solar panels.

OK, that one is a bit weird.

8 minutes ago, Reinhard said:

Sometimes, even on sunny days, it takes until 11:00 in the morning to detect.

Ah. I suspect that you have one of the clones that comes with patched firmware version 72.70c. Can you confirm by using the up and down buttons till you see the U1 or LC1 screen?

If so, what you're seeing (e.g. in your photo of the LC Display) is the PV being detected, but the power is less than 500 W. The number of rows of "solar cells" in the display varies from zero to 3, depending on the PV power, rounded to the nearest kilowatt. So for example 600 W rounds up to 1 row, 2.2 kW rounds down to one row. It might take till 11am some days to get to 500 W. Zero rows, yet the empty PV frame, actually means PV detected, just low PV power. When the PV is not detected, the panel frame goes away altogether.

21 minutes ago, Reinhard said:

During my investigation, I was testing the panel voltage with my multi meter and it showed no drop or rise in voltage. It was consistently around 100V, varying by only 1 or 2 V. The wattage was also over 1kW all the time.

So then it should look like the below, the panel frame with one row of cells, and the dotted line emanating from the panels towards the battery. Does it?

23 minutes ago, Reinhard said:

Also, from the picture, is it only my imagination, or did they sell me a clone inverter. They advertised and invoiced it as an "Axpert type" inverter.

Yes, definitely a clone. If it comes from the supplier with patched firmware, then it's definitely a clone. Also, the serial number under the barcode has too many digits, and the screw is silver.

Thanks Coulomb

I definitely have the clone with firmware ver 72.70c on it.

Is the problem with the firmware? Can I tell the supplier that?

Could there be anything else that I could look for?

Will I be able to load another firmware version on the inverter? Will it work on a clone?

1 hour ago, Reinhard said:

I definitely have the clone with firmware ver 72.70c on it.

72.70c is genuine Voltronic Power firmware with a bunch of patches applied by forum members Weber and yours truly. It's only "clone" in the sense that having that firmware out of the box is a sign that it's a clone.

1 hour ago, Reinhard said:

Is the problem with the firmware?

I think that a part of it is that you're misunderstanding the displayed data. If you are seeing 1kW of panel power yet no charging of the battery, that doesn't mean that the panels aren't being used. That kilowatt is supporting the load, so that the battery doesn't have to provide that kilowatt. The charge and discharge currents are combined into one data item with the patched firmware. See the AussieView™ and KettleKomp™ manual for details.

However, ignoring the PV (panel not connected by dotted lines to the battery) is bad (assuming that the battery isn't full, and you have at least some load). All I can think of is that there must be some hardware incompatibility, perhaps corruption of the data between the solar charge controller and the main DSP processor.

If you were led to believe that you were buying a genuine Axpert, you may be able to convince your supplier that you need it replaced by a genuine model. That has happened before.

1 hour ago, Reinhard said:

Will I be able to load another firmware version on the inverter? Will it work on a clone?

Alas, the clones that come with 72.70c tend to be ones that can't update firmware at all. If you are lucky and it will allow updates, your only real option is to update to patched firmware version 73.00e.

If you can't update firmware, at least you have the second latest firmware for your model, and you have all the bug fixes and features of fully patched firmware. 

Finally I found out what the problem was. One day, after becoming crazy deparate (my wife struggling without electricity and me working 700km away from home), I jumped into the car and drove home. When I got home, I went to the inverter, opened it and say a small white plug, that wasn't plugged in properly, like the one in the picture:

I plugged it in, the solar detected and that was the end of the problem.

The inverter ran perfectly, never switching off and the battery never went flat for 2 months. Then one day I was in the garage and I heard a loud pop and the power was down. Now the battery (Pylontech UB3000) is detecting a short and the red (ALM) light goes on and refuses to start. There is a short over the 2 poles where it connects to the inverter. The inverter is dead. What could have happened and what could the damage be?

I have ruled out problems with the PV, as I am borrowing and inverter and it shows no problems being connected in place of the broken one.

The wind was not blowing, there was no thunder and its a completely off grid system, so there was definitely no power surge.

The inverter had a 1 year warrantee, that expired a month ago.

14 minutes ago, Reinhard said:

The inverter had a 1 year warrantee, that expired a month ago.

Well, there's your problem lady (quoting Eric O. from South Main Auto, love his Youtube channel).

Probably blew the FETs on the H-bridge causing a dead short. Can usually be repaired by a competent shop. Or buy a better inverter.

Thanks,

 

I took it to a guy that said he can fix it and he could not find the problem. The pics that I saw of blown FETs, that would make the noise that I heard, doesn't seem to correspond with what I see inside the inverter. I took it apart to see if I could find the blown component, but I cannot see anything that blew. I also tested the FETs on the DC side and (except if I did it wrong) they seem to be okay. I was however wondering about the four caps between the poles where the battery connects. 

There is a short over the battery connectors. My pylontech goes into self protection mode when I switch it on.

I cannot see anything on any of the PC boards that is visibly damaged, or that could have overheated.

Any recommendations on a technician that I could take it to in the Freestate, Northwest or Eastern Cape (South Africa)?