Coulomb

But that photo you posted is the main DB already, correct? Well, there's your problem. Firstly, your main breaker is an RCBO (Residual Current circuit Breaker with Overload), which is a Residual Current Device combined with a circuit breaker. 50 A seems a little low to me, but I guess this is a business with no stove/oven or other high powered devices. The other thing is that it's a 20 mA device; usually you see 30 mA in Australian homes, and 50 mA or higher in industry where there are many inverters for motor drives. So this is a sensitive RCD before the inverter. This will often cause tripping. I'm surprised it hasn't been tripping before now, on the old settings. I don't know anything about South African regulations, so I don't know what will be possible. But I would hope that you could replace the RCBO main breaker with a 50 A ordinary circuit breaker and a 50 A RCBO after it. In fact, you could use the same RCBO, just put a 50 A ordinary breaker in front of it. Then the inverter breaker (the 30 A one, it should be adequate if you don't do a lot of utility charging while also running high loads) should connect after the main breaker but before the RCBO. It looks like your electrician will have enough room to shuffle the other breakers to the right to achieve this. But then for safety, you need an RCBO at the output of the inverter. This could perhaps go into one of the small breaker boxes near the inverter; hopefully, there is one already there. Yes. Yes. I was just looking at a Synerji work-alike (not quite a complete clone), and was not impressed with what I saw.

IF (and it's a big IF) the Synerji inverters are using the same fault codes as the Voltronic Power Axperts, then fault code 56 means "battery connection is open". Is it possible that you have a loose connection or blackened fuse contact between the battery and the inverter? The 08 error may follow from the 03 error, since there is a fixed ratio between the battery voltage and the inverter bus voltage, in your case of possibly 8.0:1, so bus voltage will be 8x the battery voltage. When the battery voltage exceeds 62.5 V, the bus voltage will then exceed 500 V, and this will trigger a fault code 08. There is a buck stage between the 8x battery voltage and the bus voltage, but I think it's only in effect when utility charging the battery, otherwise it's effectively "pass through".

Perhaps we could start with this. So these are brand new batteries, correct? Four 12 V modules in series, two such strings in parallel? Is 19.15 supposed to be a voltage? As measured by the inverter, or a multimeter? If so, it's way too high for a single module, and way too low for a string of four in series. Please tell us about how these battery modules are configured, and the thickness of the cables. With no load, what is the voltage across each 12 V module (or at least a few of them)? As an interim measure, please increase setting 27 (float charge voltage) to 54.0 V (13.5 V per 12 V module). That might at least get some charge into the battery, until we can figure out what else is wrong.

Also, it's a work-alike (not quite a complete clone) of a Voltronic Power Axpert; see this topic. @Chris Hobson had one for a short time, and he has a video review in that topic. He is familiar with how this type of inverter is supposed to work, and was not impressed. So: it may just be poor design and build quality, unfortunately.

The brochure I downloaded didn't have a recommendation. So I suggest using the same absorb and float settings as the Pylontechs. Sorry, you'll have to search for that yourself. [ Edit: actually, the settings are different, see Virge's post below. ] Two other settings come up, however. The minimum battery voltage should be 40.5 V, but I'd leave it at the default of 42 V, or even increase it anywhere up to the maximum of 48 V (for maximum battery safety, at the cost of slightly less run-time in the case of a long blackout). [ Edit: limit this to 46 V unless using LFP patched firmware; see this post below. ] The other thing is that when paralleled, the maximum charge current should be 20 A per unit. So setting 02 (maximum total charge current) should be 40 A. I don't think so. I doubt it. Oh wait; the inverter part of the inverter-charger will be running when utility charging (the inverter is "run in reverse" for battery charging from utility). If my guess about the RCD before the inverter is correct, then less sunshine would mean more utility charging which would mean the inverter proper is running more of the time, increasing the chances of a nuisance trip. The other slight possibility is problems due to a non-genuine (clone) inverter-charger; some of the clones look very similar from a distance (e.g. I can't tell from your photo). You could run through the genuineness section of this post (just the dot point above the first photo), or post a photo of the sticker(s) on the side of your machine. Include the nearest lid-to-case screw in the photo if possible. Cloned machines may have less suppression of electromagnetic interference, or other limitations.

Can you take a photo of inside the Distribution Board, and indicate which breaker is tripping? So your Output Source Priority (setting 1) was Uti, that means the unit would be in bypass mode all the time, not running the inverter. So it looks like something causes the tripping when the inverter is running. It's still possible that you have a Residual Current Device (with or without Overload protection) before the inverter, and that this only trips when the inverter is running. Is it one of the breakers near the inverter that is tripping, or one in the main Distribution Board further away? To stop excessive switching, you want a larger difference between settings 12 and 13 (when utility charging comes on, and when it comes off, respectively). With setting 12 too high, you'll be charging your battery a lot, which will waste utility power, and will reduce the life of the battery a little. So I'd increase setting 13 instead, or even reduce setting 12.

Was it a Residual Current circuit Breaker with Overload protection (RCBO)? (Like a two position breaker, but with a "test" button). If so, it might be wired before the inverter, and needs to be wired after it. But I would have thought that you would have had that problem before now. Or is it an ordinary circuit breaker that connects to your inverter's AC input? Otherwise, I have no idea why it would be breaking now when not before. Perhaps put the settings back to what they were, to convince yourself that it really is related to the inverter (it might be a coincidence).

Some of the defaults are pretty annoying. I'd turn off the beeping (setting 18 bOf), turn off auto return to main screen (setting 19 kEP), turn on overload bypass (setting 23 byE), and you should check that the float and absorb voltage settings (27 and 26) agree with the manufacturer's recommended values. Check out Chris Hobson's Axpert Settings 1.1: https://powerforum.co.za/files/file/4-axpert-settings/ .

Well, this way you'll be using your batteries (which does reduce life, compared to not using them at all ), but keeping them fairly full. Just keep away from 54 V (3.6 VPC) for too long, as this will reduce life. That's why I suggest 52 V for setting 13, and only 53 V if you need to prevent too many switchings between battery and grid. The battery voltage isn't compensated for load or charging, so 52 V at the inverter may be less than 52.0 V at the battery (depending on your wiring, mainly). Hence 53 V for setting 13, which will be somewhat less than 53 V at the battery itself, might be OK. So no, no damage, just using them per your stated need (highest priority is to be ready for a blackout).

Yes, I think you've got it. The SOL Output Source Priority (setting 1) will run off battery (using solar where available) during the day, and switch the loads to grid at night (if present, obviously). You want a combination of solar and grid charging, so you can charge during the day from grid on rainy days, in case you get a blackout on such a rainy day. So the SNU (Solar aNd Utility) charge source priority (16) is what you want. You want to run off the battery during the day when you can, since that's the only way to use your solar panels effectively. But you will want to go to utility charging (and therefore utility supplying loads) pretty quickly, since your loads are about equal to your best solar power most of the time, so you can easily fall behind with a few clouds. So as soon as the battery voltage falls to about 3.33 VPC [ edit: that's 85-90% SOC ] (50.0 V, assuming I'm right that your Leochs are 15S), you want to get back to utility charging. So I'd set your Back to Utility voltage (setting 12) to 50, and back to Battery voltage (setting 13) to 52, maybe 53 V. You may need to play with these settings a bit to prevent too many switchings between battery and utility. I have no trouble with these, I never have a computer drop off for example, but you might get annoyed by lights flickering.

Are you using any monitoring software? That could be overriding your inverter settings, so then you have to find your monitoring software settings.

Dang those inverter part numbers. Unfortunately, the service manuals don't seem to be any help with troubleshooting the SCCs. I think all you can do is to check the MOSFETs and their drivers. Hopefully there will be enough drivers with similar components that you can compare one to the other. It will probably an interleaved configuration, like this: That's about all I can find in the service manuals, sorry.

If this is equivalent to the model you have: http://voltronicpower.com/en-US/Product/Detail/Axpert-VM-II-3KVA-5KVA then you need at least 120 VDC to start the MPPT. It might need even higher voltage to get it started. Your video showed less than 60 V, presumably across the PV inputs.

No, that should do it. As long as the LCD went dark for at least a second or two (and not due to timeout, obviously). Unfortunately, yours is a textbook example of likely blown SCC/MPPT. If you had recently taken it apart, one of the classic gotchas is swapping the two white 4-pin JST cables (one goes to the SCC, the other to the LCD). But it sounds like yours failed while running. Yeah, mine does that too, despite a capacitor-and-MOSFET upgrade and fan rectification. I think something in there gets really hot, the transformer or a large inductor perhaps, and hot resin might trigger the faint "burned semiconductor" reaction. Mine has been running trouble free for almost three years now, and the smell might be from other electronics nearby. I have a 48-12 V DC-DC converter module nearby that runs close to maximum capacity, and runs very hot, which might be the source of the smell. So: it might mean nothing, but if you've never noticed it before, it's not a good sign.

The lifetime of lithium batteries, particularly LFP, is much higher than that of most lead acid, with a few exceptions. I've heard of lead acid cells with 10 year warranty, that still failed well before the expected lifetime. Trojan lead acid seems to be long lived, but there is the hassle of maintenance. Super capacitors just don't have the energy density of batteries. Some that claim to be super capacitors with million cycle lifespans are clearly not what they claim to be, for example the Arvio.

You won't be wanting supercaps, but do the research and convince yourself. Actually, there is, but till recently the received wisdom on this was wrong. The SOL output source priority (setting 1) does exactly what you describe: Battery mode daytime (when possible) and Utility mode when the PV input stops. See the correction in this post.

That does not sound good for the new machine. Leaving the parallel board in should make absolutely no difference. NE means "NEw". Every machine starts of being a "new" before it decides if it's a "master" or "slave". That sounds like setting 28 is still set to PAR though, or it would go straight to "master". Did you remember to turn the power off when changing setting 28 from PAR back to SNG? It sure sounds like the new machine isn't happy. I assume it must be the same or very similar firmware version, and not a clone... is that the case? Otherwise, you'd presumably not see HS and SL.

This was fine for 3 strings, but while you are running 2 strings, I'd reduce that to 40 A. Yes, that's what the recommended voltages translate to. I'd stick to the low side of these numbers, since Egypt has a hot climate. Those figures of X mV/°C/cell should be minus only; lead-acid battery voltages should reduce as temperature increases (this is called temperature compensation; ideally you should have a temperature probe and the inverter should apply the compensation automatically. In practice, adjusting the settings a little for Winter versus Summer around equinox time is OK). The absorb ("cycle use") figure is a bit on the high side for gel cells, so I'd maybe even reduce them a little more (say 57.5 V). 52.8 V sounds good for the float voltage, although it seems low compared to other gel cells. I guess you just have to trust the manufacturer. Yes, it could make a difference for the performance and/or life of your batteries.

That one should be fine. For whatever reason, the communications board has evolved far more than the other parts of the inverter-chargers; see for example this post. His is a 2015 model (looks exactly like the customer's); yours is a 2017 model (we don't have a photo of one of those as yet). They should inter-work just fine. Edit: I pinched a photo so we have a 2017 board photo; I hope you don't mind. I'm happy to blank the serial number if you thing that's important (I don't). You can see that there is little to no real difference between the 2015 and 2017 boards.

Yes, definitely. It's toast. I'm no expert on lead acid, but I think yes, it might. The other strings will discharge trying to overcharge the short string. All the other cells in the short string will go closer to or even past the gassing voltage, so they will likely have lost most of their water, so the acid will be dilute. Your two remaining strings might recover somewhat with a good charge, now that they don't have the bad string weighing them down. If you have absorb time settings, try a long absorb time for a week or two.

I'm pretty sure there was one on here some time ago. I've just uploaded my copy: https://powerforum.co.za/files/file/30-axpert-mks-1-3kva-service-manual-20130611pdf/ You didn't say if yours was a 24 V model; this manual is for 24 V models only. Not the best, but a lot better than nothing.

This is quite helpful, thanks. The basic problem seems to be that that day at least, you never reached absorb voltage from PV. You need to be at some 56.4 V (14.1 V per 12 V module) for an hour or so to have the battery soak up charge. That peak of battery voltage when you got to shore power (I assume that was soon after 17:00) should have happened at about 11:30 when the load dropped off. Maybe you just didn't have enough solar power to cover your loads and charge the battery properly that day. In fact, your loads are quite high. Your solar panels add to 4400 W nominal, of which you could count on perhaps 3 kW in the best hours, times perhaps 5 hours in summer. So that's only 15 kWh, and you say you use 10-15 kWh during the day. So you can barely cover the highest usage days, even with good panel orientation. In a boat, I imagine that your orientation will be highly varied when at sea. I note you barely reached 2.3 kW on Dec 10th. So I think that you could use more panels or less load. But there seem to be days like Dec 9th when the battery had a small discharge, and no charge at all. So I think your settings must be off. Please post your settings.

It probably would be, if we could see the images. I think you need to allow access to other than yourself somehow. Digging a little, I get EC-4008, which seems to be associated with "authorization error".

I don't know the Infinis at all well, but I know their cousins, the Axperts, moderately well. On the Axperts, the "bus" refers to the DC bus that feeds the main inverter. It operates at some 400-500 VDC. There are large electrolytic capacitors on the bus, and these require pre-charge or "soft" charging. On the Axperts, this is achieved by a small power supply whose output is rectified and the power is applied directly to the bus. The power supply is turned on and off my the DSP (it might be different in an Infini, since they have multiple processors). When the bus voltage reaches a certain level, this soft-start power supply is turned off, and the usual ways of powering the bus (e.g. from the solar charge controller and/or battery DC-DC) are enabled. The soft start should take of the order of a second or so. If it takes too long to charge the bus capacitors, the soft start power supply is turned off, and the soft start time out error is generated. Usually, the reason that the soft start fails is due to shorted IGBTs in the main inverter, or one of the power units that feeds it (e.g. the battery DC-DC). The soft-start power supply itself is usually a single oscillator/driver IC like a UC3845, driving a single MOSFET, powering a multi-turn inductor in a flyback configuration. So there will be a single rectifier at the output, and there will be a current sense resistor in the source lead of the MOSFET. I have a partial schematic trace of the Axpert's soft start power supply here. Usually, there are pads near the large bus capacitors marked with BUS+ and BUS- or similar. After checking for residual voltage, your next test would be resistance across the bus (all power off, of course). You should see a low resistance rapidly rising to the mega-ohm region, or at least many tens of kilo-ohms. If it's something like 4 ohms and steady, then you have shorted IGBTs and/or MOSFETs. Good luck.