Coulomb

That looks to me like it's shorting pins 11 and 12, which should be connected anyway, according to my traced schematic. But well worth checking what's underneath. I don't use Solarpower or even Watchpower, so I have no idea what that's about. Is it possible that you accidentally entered a password? If you go back to V1.16 does it ask for the password? Or more likely the saved settings have a different layout V1.16 versus V1.20. So try a clean re-install. Maybe check for any leftover files between the uninstall and the reinstall.

👍 Yes, they dropped the leading 1 when they figured out that one of the firmwares works fine for both display chipsets. Maybe. Certainly worth a try.

Huh. So it might be for ease of paralleling. But then the lead lengths would not be the same. Do you have the settings for controlling the Twin AC output port? That would require a suitable display firmware, of course.

Hmmm. Sounds like you have a display/comms board with slow opto couplers. They operate at a very lazy 2400 bps, except when reflashing, in which case they switch to 4x the speed (9600 bps). If you're handy with electronics, you could replace them yourself, or you can try and get the inverter repaired (they will replace the whole display board). Or it may be that your USB to RS-232 adapter isn't one that works well with these inverters. I would try copying the 60.03 hex file to a blank USB stick, renaming it to dsp.hex, and attempt updating via the USB stick. That will at least bypass the opto couplers between the display and the RS-232 port, and also your USB to RS-232 adapter. Firmware update instructions for models with a round display.

It might be the same supercapacitor that is found in some display boards, like the gold/yellow one below from an Axpert Max II:

There are many shorts between pins on the paralleling boards. Assuming that the same cards are used in Infini/iGrid models as in Axpert/iSolar models, I have traced the schematic of the paralleling boards here: https://forums.aeva.asn.au/viewtopic.php?f=64&t=6007&p=90479#p90479

Alas, I have no 47.xx in my collection of about 200 firmwares. Does it always read exactly 120 V? I'm wondering if the measurement is faulty. You need a DC clamp meter to measure PV current. An ordinary multimeter can measure the PV voltage, though at this voltage level it would be safer to have a decent one (not a cheap one from a car shop). It would be good to measure the PV voltage at the PV input terminals when it is reading 120 V. At a wild guess, your inverter's solar charge controller (sometimes misnamed the "MPPT") isn't working. Either it's not reading the voltage correctly, or it's doing something strange to clamp the voltage like that. Or something is preventing the MPPT from drawing significant power, e.g. high resistance in the PV cabling. Your inverter seems fairly new (manufactured July 2024), and it seems genuine to me. So you should be able to get it repaired or replaced under warranty, though I realise that this may not be convenient when it was purchased from afar.

Tell me if I can reflash it to the MAX II 11000 TWIN model. Hard to say, as yours is a clone. It seems to be using Voltronic firmware, so newer Voltronic firmware may work. 122.12 is fairly old. The latest in the x22.xx series is 22.24, available from the link below. Firmware upload instructions (for Axperts; your clone will hopefully update the same) for models with a round display. That's due to the main firmware. The latest patched 64.xx firmware for Axperts is x64.62,available below. Use the same firmware upload instructions.

Duh! If course, it's a Max II, so it will have 39.xx not 38.xx display firmware. The latest I know of is 39.27, much more recent than 39.06, and available from the link below. Max II displays use the same firmware as VM IVs.

You should be able to see it by pressing the up or down buttons on the display until you see something like U1 12 34 which in this example means main firmware version 12.34 .

I assume you mean bP, i.e. Battery not Present warning. I'll guess that it's a 16S LFP battery, 5.12kWh made up of 16 100 Ah cells at nominally 3.2 VPC. Check for a short circuit at the battery terminals. But my guess is that the battery has been discharged too far, and one of the cells is a bit weak, so it can't supply the surge power needed to start the inverter. * Does the battery voltage go to 0 immediately that you connect to the inverter, or only about 10 seconds after starting the inverter? My suggestion: Switch off the inverter's load switch. This means you will not get power to your loads, unless you have a separate transfer switch. The inverter should stay on if there is a battery present and a charge source (AC-in or solar). It should then be possible to charge the battery, without the BMS shutting down. After half an hour or so (depending on many factors), the battery should be strong enough to power the inverter. I note that a 5.12 kWh (100 Ah) battery is a little bit light for a 6 kW inverter, but it should be OK once you change the charge settings. Setting 12 at 46 V is way too low in my opinion. I would set it to 51 V, or maybe 50 V. The CV and float voltage settings should be reduced to about 56.0 and 55.2 V respectively. I use 55.2 and 54.0 V myself, on my 16S LFP battery. Manufacturers suggest ridiculously high voltages, which are not good for longevity, but make the batteries look slightly better because they run slightly longer. These values are 16/15 of the values that many users settled on for their 15S Pylontechs. It's possible that at the high values you were using, the BMS was cutting off, which might have caused the cells not to equalise. If so, that would have exacerbated your problem. To save utility cost, you could change setting 13 to 53 V, especially if you have plenty solar.

Then it might be firmware related. Alas, I have no 76.xx for Axpert Max II 8 kW. But it might be the display/comms firmware, since that firmware talks to the BMS. What is your present U2 (MCU, display/comms) firmware version? Edit: I first assumed it was a non-twin, then realised that the sticker says Twin. If it's 38.xx, then the latest I have is 38.11, which is well over a year old now.

Hmmm. I'd say that the connector marked with a red 3 (CN6) is the one you want. In a 5 kVA inverter like mine, that connects to a board like this to convert it to isolated RS-232. The comms board on models with a removable or round display is quite different, but the pinout could well be the same. From this AEVA post: So this isn't even "TTL", but you can probably adapt it with a simple circuit. I believe that you can determine the polarity by noting that idle = dark. What is marked TX above (transmitting from DSP to outside world) likely has 3.3 V behind it, something like the right side of this: I've never seen that populated. It's possibly for a supercapacitor, but if so there would have to be additional components for limiting the supercapacitor's voltage, and these may not be present. Sorry, you are in uncharted waters here (as far as I know at least).

You mentioned master and slave. I assumed that this means that you have paralleled inverters, or a 3-phase configuration of three or more inverters, each with paralleling boards, and interconnected with the appropriate cables. But now I'm thinking that by "master" you mean the one that is connected to the battery BMS, and by slave you mean the other inverter(s), not paralleled or 3-phase connected. I think that if you have more than one inverter connected to a battery with BMS, and that BMS talks to one of the inverters via an RS-485 or CAN-bus cable, then the others need to be properly paralleled. Do you have these paralleling boards installed? Ignore the highlighting, it was the first paralleling board image I found. Most of this board would be hidden inside the inverter, but you should recognise the connectors that poke out, especially the pattern of two green ones right next to 2 high density D sockets:

Yes, but for technical reasons you have to tell us what present main firmware you have. That's the U1 (DSP) firmware version. You can find this from monitoring software, or from the front panel via the up or down buttons. That's an unusual pattern on the front of the inverter; I've not seen it before. Hopefully it is a genuine Voltronic, or at least one that can be updated with Voltronic firmware.

I assume you mean BMS cable. That could be normal, if the battery was nearing full, or the cells were unbalanced. I'm trying to decide whether your inverter is a clone. The QR code points to ags dubai dot com, and not https. Usually the QR code is "busier" (more dots because more characters). But that might just be a customisation that resellers are allowed to ask for.

With Axperts, when in parallel or 3-phase output types, the master takes over charging control and tells the slave what to do over the CAN bus that is inside the grey paralleling cables. The master doesn't have to be the one connected to the BMS. In theory, the one that ends up as master should be random. In practice, 99% of the time the same inverter ends up as master. Have you actually tried setting the inverters that are not connected to the BMS to USE battery type? I agree that it's a silly requirement; the ones not connected to the BMS should just ignore the battery type setting. As before, the above is how Axperts work, and I would expect Infinis to work similarly, but I'm not certain.

I'll just post my recommended Pylontech battery settings, for those not having the BMS control the charge via the BMS to inverter BMS port cable. It's in dire need of redoing, but the important numbers are there: 52.5 and 51.8V. These figures are a little higher than I use myself when adjusted for 16S (Pylontech is 15S), but were a consesnsus amongst actual Pylontech users. Edit: The maximum current setting is for the original 2.4kWh models. Adjust accordingly for recent models' total capacity.

@Tomeek, thanks for the detailed testing. It seems that 0.50 A (as per factory firmware) is too high a current threshold, and 0.05 A (as per Georg594's original 90.06 based patches) may be too small. 0.25 A seems to be best for you, but perhaps some lower value yet larger than 0.05 is better still. In the pursuit of the ideal current threshold, I've made MPPT1 0.25 A as requested, and MPPT2 half that at 0.12 A. That's still 2.4x large than 0.05 A. No other MPPT patches are made. The only other changes are that the firmware reports version 372.66 to monitoring software, and the premature float bug is fixed. Tomeek or others, if you could compare the performance of MPPT1 versus MPPT2 and report the results here, that would be great. I'm happy to make another patched version with whichever current threshold seems to work best, for both MPPTs. It's comparatively little extra work now that I have 372.66 written. This is patched firmware version 372.66 for the Axpert Max E 8 kW '28066 non-Twin. Do not use with any other model or chipset. It is the latest of the available 72.xx firmware versions that has the "presumed good" MPPT logic, has the QDOP command, and does not have the bug with reading the serial number. The current threshold for MPPT1 is reduced from 0.50 A to 0.25 A, and for MPPT2 from 0.50 A to 0.12 A. Use at you own risk. Firmware upload instructions for models with a removable display. dsp 372.66 patched 25 12 Max E 8kW 28066 non-Twin.zip

Phew! I guess 72.72 is deprecated then. 72.74 is 6 months later, and seems to have fixed that serial number problem, but is missing 6 commands compared with 72.72, and one of those is QDOP. Hence 72.74 does not have the ability to use SoC% for settings 12, 13, and 29. The latest 72.xx firmware with QDOP and the "presumed good" MPPT logic is 72.66, now about 21 months old. I'll see if I can patch 72.66 with the 0.25 A change, a la 172.62. This might be the best we can do for any firmware: the "presumed good" logic plus halving the current threshold. Maybe slashing the current threshold to a tenth as in 90.08 etc is too much.

Ah. So that's a patched 90.06 firmware. I was confused by this thread being about Axpert MKS IV models. 90.06 is now fairly old, although at least Georg594 seems to think it's pretty good, and I haven't heard about problems with it. The latest in that line is 46.82 from about 13 months ago as I type, and it has the advantage of having the presumed good MPPT logic (i.e. the stuck at 90 V issue is almost not present). Available here: https://forums.aeva.asn.au/viewtopic.php?p=102981#p102981

Mangling of the serial number to the "null" number of 55355535553555 is quite serious. It could lead to fault code 90 down the line. Does the serial number appear correct again if you go to one of the other firmwares, or is the loss of serial number permanent? If not permanent, it would mean that the serial number is not stored at the same address in the EEPROM in versions 72.72. I'm surprised that 72.72 behaves differently to 72.74; at first glance, they seem to be using the same addresses. I don't understand this sentence; perhaps a bad translation? Do you mean you want it to show the SoC from the BMS instead of its usual very bad voltage-based guess? If so, you need the BMS cable connected, although Solar Assistant and possibly other monitoring software might be able to fix this if they connect to the BMS. If you mean you want to be able to use SoC% in settings 12 (back to utility), 13 (back to battery), and 29 (low battery cut-off voltage), then you need main and display firmware that processes the QDOP command. I may make a list of these. Only firmwares released recently (I forget the cut-off, perhaps late 2023?) will have this. Did you mean to post an image here? Or do you mean that you are somehow bypassing the removable display? I don't recommend doing that, as some commands are subtly different between the ones you see via the display, and the ones you get connecting directly to the DSP.

I'm sure that there would be; it's hard to see unless you get the exact right angle. But it doesn't matter; 46.05 on the white sticker indicates that it's a '2809, so the 890.06 patched firmware that I PMd you should fix the problem. Get a suitable USB to RS-232 cable in conjunction with the cable that came with your inverter, or make your own (see this post from the upload instructions). For what it's worth, that control board does look genuine. I wonder why so many fault code 90s are coming up in error?

I have never seen any firmware update files for the Bluetooth / WiFi processor, and I don't think that there is actually a mechanism for updating it. Showing 00.00 as the U3 firmware version seems to be common, whether Bluetooth/WiFi is working or not, so sadly it doesn't seem to tell us anything useful. Sorry, I don't use these, so I have no idea how to fix it. PS, please post in English.

Zero rings sounds like it's open circuit (or you tested the wrong pairs of pins). If the transformer is open circuit, I can't see how that would shut down the power supply when the 3525 is enabled. [ Edit: Though maybe your tester is testing for shorted turns, and rings in the case of no shorts. ] Did your tests include the small MOSFETs (UT304G) across the outputs of the NPN/PNP push-pull pairs? If these are shorted or not turning off, that might cause the power supply collapse.