Hi

Does anyone know if there is a way to upgrade / restore the firmware on a Axpert VM 2kw, or a way to calibrate the battery voltage?

The battery voltage on mine is suddenly totally wrong.

I saw that it was showing the battery voltage as 22v and decided to check the voltage at the batteries and I got 30v.

I've also checked the voltage at the connection points on the inverter and it is actually 30v, so it is trying to charge the batteries, but they are full.

2 minutes ago, Adri76 said:

is a way to upgrade / restore the firmware on a Axpert VM 2kw

You have to have a firmware update file, and these are rare, esecially for the smaller (< 5 kVA) models. But your problem does't sound like firmware will fix it.

3 minutes ago, Adri76 said:

or a way to calibrate the battery voltage?

You could try the method that works for the 5 kVA models. It involves sending commands; not everyone is set up to do that. See here.

5 minutes ago, Adri76 said:

it was showing the battery voltage as 22v and decided to check the voltage at the batteries and I got 30v.

That's a substantial error. I suspect a hardware fault. If you are handy with electronics, perhaps inspect the PCB for dust / carbon tracking / failed voltage sense resistors. Voltage sense resistors can be recognised by the long string of identical valued SMD resistors all connected in series.

I will have a look at the electronics.

Are replacement boards etc available for these inverters?

I bought the inverter used, so there might be some problems.

There are almost no software / hardware info online, so it's quite difficult to troubleshoot as most info available is for the 3-5kva units.

Thanks

These are the current software versions.

I had a look inside and the cntl board and everything else looks brand new except for the cap and resistors in the pic. Could this affect anything?

9 hours ago, Adri76 said:

I had a look inside and the cntl board and everything else looks brand new except for the cap and resistors in the pic. Could this affect anything?

Certainly that sort of muck can upset the voltage divider resistors. That might be coming from the tall electrolytic capacitor; I very much doubt it's coming from C57. Those resistors are marked 20R0 (20.0 Ω), and they appear to be in parallel. That's way too low in resistance for a voltage divider resistor, so they won't directly be the problem. They may affect the measurement of some current, however. Certainly find where it's coming from, and at least replace that part, then carefully clean off all that gunk. Re-solder pads if necessary. It could be worse on the other side; there are many more parts under the board than on top (the "component side", at least for through-hole parts).

Yes, there is way less support for the lower power models. I suspect they're less common, and/or are considered too cheap to repair.

18 hours ago, Adri76 said:

These are the current software versions.

Sorry, I have nothing like that in my collection.

Most of the inside is very clean and looks brand new. I am going to clean everything with contact cleaner and check if there are any dry joints or anything that needs to be re-soldered.

I don't think there are any bad components or anything that is blown, or none that I can see, unless it is one of the IC's.

 

Ok, so I've cleaned everything with contact cleaner and found that there was also some dirt on the CNTL board tabs which I also cleaned.

I also initially had a problem where the inverter would not use the solar panels to charge the battery.

Now everything seems to be working.

I've added an extra volt meter to my battery bank and the voltage is now within 0.1v of the voltage on the external volt meter and the solar is charging the batteries. (photo taken 6am so not much sun yet)

Thanks for the help and advice.

 

Edited January 30, 20205 yr by Adri76

Everything was working 100% for about a week now, but last night I noticed that the battery voltage on the inverter has started creeping down again.

At the moment the inverter is showing +- 1v less than my external voltmeter.

I verified the battery voltage with a multimeter at the battery connection on the inverter and it was exactly the same as the external volt meter, so the inverter is definitely showing 1v less than the actual battery voltage.

It would not be a massive problem if it stayed at 1v difference, but the difference keep increasing daily.

I've disconnected the batteries and all power to the inverter for 30 seconds and restarted it, but the voltage is still wrong.

 

I'm wondering if the muck on the PCB is due to insects, and perhaps they're moved back in.

I got the inverter from someone else and I think the dirt on the pcb came with the inverter, but I probably should open it again and have a look.

I wonder if humidity could have something to do with it as we've had a lot of rain the last week or so.

It's really weird that everytime I clean the pcb's it works fine for a week or so and then it starts acting up. (happened 2 times already)

It is inside my garage mounted to a board on the wall so not in a very hostile environment.

Do you perhaps know if there is someting that I can maybe spray on the pcb's to protect them?

2 minutes ago, Adri76 said:

I wonder if humidity could have something to do with it as we've had a lot of rain the last week or so.

It's really weird that everytime I clean the pcb's it works fine for a week or so and then it starts acting up.

Certainly humidity can be a factor. Are you using a PCB lacquer to protect the board from moisture?

No, I'm not sure what to use or where to get it.

Edited February 6, 20205 yr by Adri76

I was thinking of using some kind of corrosion protection spray, but most of them are oily and would attract dust and probably make things worse.

Something like this (the one I use is Servisol NA-1002 in a red can):

https://www.ebay.com/itm/Servisol-Plastic-Seal-60-PCB-Lacquer-6100009000/132069020240?hash=item1ebfed5250:g:4K0AAOSwkNRdytq0

I could not find a supplier in a hurry that is based in South Africa, but I'm sure there are plenty. There are plenty of other brands, too.

Key words: lacquer, conformal coating.

Ok, thanks, I will have a look.

It looks like this should work, one of my suppliers stock this.

http://www.q20.com.my/products/q30/

13 minutes ago, Adri76 said:

It looks like this should work

Yes, looks good. Seems like it's the stretchy kind; nothing wrong with that.

Ok, so I've cleaned everything again and sprayed the pcb's front and back with q30.

I let everything dry, and connected it to the battery and now it's showing 1v too high. (Actual voltage 25.4v and inverter is showing 26.8v)

I'm out of ideas.

I suppose it would be impossible to get a circuit diagram for the motherboard.

First, see if the ratio of reported / actual is consistent over time and temperature. If not, you still have a hardware problem.

If the ratio is consistent now, you can either find the resistors measuring the battery voltage, and replace one with a different value, or attempt the software calibration, using commands from the link in my first reply.

Schematic diagrams are very hard to come by. Edit: however, try following the tracks from pins 5 and 6 of the control (daughter) board. These might be the measurements for BAT.V- and BAT.V+ respectively. It might be worth following those tracks on the control board as well as on the mother board.

Edited February 14, 20205 yr by Coulomb

I checked this morning and it was reading 2v too high, this afternoon it is reading 2v too low.

The control board has 2 connectors, where would I find pin 5 and 6?

 

Here's a pic of the control board.

18 hours ago, Adri76 said:

I checked this morning and it was reading 2v too high, this afternoon it is reading 2v too low.

Not good, then.

18 hours ago, Adri76 said:

The control board has 2 connectors, where would I find pin 5 and 6?

Ah. Your control board is very different to the one I know. It also has a different processor (64 pins versus 100 pins, 4.9152 Mhz versus 20.000 MHz). But it still has 48 pins of connector (split into 2 x 24, ours is 3 x 16 all inline). I would guess pin 1 is at the top left of your photo, near the op-amps. Note the "2" and "24" near that connector at top left; the pins on that side would be numbered 2, 4, 6, ... 22, 24. On the other side will be 1, 3, 5, ... 21, 23. On our boards, it's obvious from a top view that pins 5 and 6 of the connector go via 20 kΩ resistors (marked 30C) to pins 9 and 10 of an op-amp (TL074).