Coulomb

The Axpert is perfectly capable of pushing the current into the battery when there is enough solar power available (e.g. when the Outback charge controller would do so), it's just a logic error that makes it think that charging is over and it's time to float the batteries, not charge them. I run 2S of 72-cell panels with no issues that I'm aware of. Sometimes the MPPT voltage will be down to around 63 V, but it still seems to charge just fine. In summer I get nominal or even over nominal power for short periods of time. My lithium iron phosphate (LiFePO₄) battery charges at a few volts lower than lead acid (my CV voltage is set at 55.2 V), but I don't think that would affect solar charging significantly. The charge bug issue is described in the "discussion post" about the latest patched firmware. The essence of the problem is that the Axpert judges the charge to be finished when the charge current drops below a certain value. That could be because the battery is full, or because there is a cloud. It can even happen with utility charging, with perfectly steady charge current, if you charge at a rate less than the threshold. For most cases, that threshold is a fifth of the maximum charge current setting (parameter 2) (the threshold is different for the LiFePO₄ patched firmware and/or with paralleled inverters). Prove it for yourself with a test utility charge: set parameter 2 (maximum charge current) to 120 A and parameter 11 (maximum utility charge current) to 20 A. The threshold will be 60 A, and the charge current will be 20 A, so after 10 minutes (after only 30 seconds with pre-72.70 official firmware) the charge will change from bulk charging at 20 A to float charging at a handful of amps (depending on the exact state of your battery). Details here. You might not notice the charge bug, especially with official firmware version 72.70, because when you watch it, you might not get 10 minutes of uninterrupted cloud. It's harder to miss with the earlier versions, because they only needed 30 seconds of cloud to terminate the charge, but you could still miss it, especially if your weather has been like it has been in Brisbane Australia lately (weeks of nothing but blue sky; my rainwater tanks are empty). If you remove *all* solar input, the SCC will turn off, but you can probably simulate 11 minutes of cloud by turning off the isolators / breakers for most (not all) of your solar strings. Watch the battery voltage on your monitoring software or on the LCD display, and watch for the change from bulk/absorb to float charge, on your monitoring software or watching the charge LED on the front panel go from solid (bulk/absorb) to flashing (float). It's really scandalous that this charge bug hasn't been fixed after all these years. Charging batteries is a major job for an inverter/charger, and a battery worth more (sometimes many times more) than the cost of the inverter may be ruined as a result of the bug, which is probably a single line of code to fix.

Pretty much, yes. There are some relays inside to do the bypass for you. No, the power only flows into the battery when in bypass mode, never from battery to load and AC in. [ Edit : only hybrid inverters can blend power from the battery and inverter with an AC source. The Axpert does however combine any solar charging with AC charging.] I'm not aware of a three stage switch ; that sounds like something added to your installation. You can't force the inverter into bypass mode with isolators, but I believe that you can do it with monitoring software that sends the appropriate commands (e.g. output source priority). The inverter will switch to bypass mode automatically and without fuss when the load exceeds 5 kVA or 4 kW for more than a few seconds, or if the battery voltage droops too low. When battery and load are suitable for 10 minutes, it will switch back to battery mode (if priority is set that way) automatically. The inverter output phase is synchronised with AC input to make the transfer smooth.

Well, per this post about the Pip-5048, arguably the latest version of the Axpert: http://forums.aeva.asn.au/viewtopic.php?p=65520#p65520 "And they have finally put the fans in the other way round, now they draw the cool air in through the bottom and the heat comes out the top, As it should." I note that the heatsink has the ability to sink heat to the atmosphere, more so than the bottom plate. Also, newer models don't have the heatsink at the top, and the SCC is in the middle between the two main heatsinks.

Actually, in the 4 kW Axperts, there are really only 2 modules, unless you count the processor daughter board as a third. The AC mains charger is actually the inverter run in reverse, so the power flow is from the AC input (connected to the AC output when AC charging) to the battery. In lower power models, I believe that there is a separate AC charger. In some Axperts, there is a power transformer that is part of a voltage regulator, which can look like an AC power supply. On the heatsink that is not on the edge, they are all IGBTs or diodes. Four are for the full bridge at the higher voltage end of the DC-DC converter, one (two on some models) are for the buck converter, and four are for the 230 V inverter full bridge. [quote[ 5 of the 9 blew up during the gunshot blast. [/quote] We have a tentative theory that it's possible for the capacitors on the battery side of the DC-DC converter to go high impedance with age and heat, and this can cause transients on the DC bus (the 400-500 V bus), which can blow up the IGBTs. But it's far from certain. Even so, having opened up the inverter, it would seem wise to replace those capacitors. As a point of interest, which ones blew up? Hopefully that includes inspection of the battery-side MOSFETs. The problem is that usually the gates sort to collector and/or emitter, so you often blow up parts in the gate drivers. See below. It may be better to leave the replacements out until you test the gate drivers when powered up. But that's pretty advanced, and we haven't figured out all the details, unfortunately. If you want to to attempt this, see this post. You should as far as possible test all the gate driver components: for the higher voltage full bridge of the DC-DC converter (looks like we haven't traced those yet, but from poor memory they are pretty simple) for the buck converter (similar to the below, I believe) and/or for the 230 V inverter but of course only the gate drivers for the sections where the IGBTs failed shorted to gate. Be aware that it's possible to think you have fixed the gate drivers, put it all back together, and have it fail again. That seems to be more of a hazard for the battery side MOSFETs, whose drivers are more complex, but it's still something to be aware of. It could well be ~60% of the cost of a new inverter.

I believe that the delay is 10 minutes, though there are signs that it once was 1 minute. I'm trialling a patch to do this, [ edit: i.e. make the delay 1 minute again ] but my first attempt resulted in 8 seconds. That could cause excessive relay operation with a load like an induction cooker that switches on and off every half minute or so. At a friend's suggestion, I doubled the constant and ended up with a 1 minute delay. Mission accomplished? I'm concerned that I can't figure out why the changes aren't behaving as expected, possibly leading to unexpected consequences, so I'll think about it before releasing that patch.

There have been some blow-ups in Australia, which we've attributed to the relatively low life ratings of the capacitors in the battery-side DC-DC converter (not the MPPT charger). Some of us, myself included, have pre-emptively replaced all the MOSFETs with 100 V models (up from 75 V and lately 80 V), and replacing the 2000 hour capacitors with 80 V 10,000 hour models (some were only rated at 63 V). But we've so far ignored the MPTT charger. http://forums.aeva.asn.au/viewtopic.php?p=64358#p64358 Are you sure all the devices (MOSFETs and IGBTs) on the main board heatsinks are ok? Usually if they blow up, there are cracks or stains from the smoke coming out. You will need to either take out the main board (a lot of work and not recommended), or use a light and mirror to see them all. It's possible that after a gunshot type failure (usually caused by electrical shoot-through), fragments from the main board could end up in the SCC (Solar Charge Controller). The SCC board isn't too hard to remove (the only trick is that there is no need to remove 4 screws securing the heatsinks, but no harm is done if you do), and it should be obvious whether the MPPT MOSFETs have failed. If the SCC has failed, it can be left out and the rest of the inverter should power up ok (I'd check the battery terminals for short circuit first, though the reverse battery protection circuit might mask a short circuit of the battery-side MOSFETs). That should give you confidence that the repair should not be too expensive. [ Edit: the 4 screws securing the heatsinks are near the end of the long SCC board near the PV input terminal. ] [ Edit: I'm assuming you have the more modern type with a long SCC board on top of the main board. Older ones have all the electronics at the top, and they have a heatsink at the top of the Axpert visible from the outside. ] As for repairs, I don't know the situation in South Africa. It may well be better to buy a replacement SCC board if possible, and either install it yourself or get someone to install it for you. From Australia, I'd order a replacement from Maximum_solar on Ebay, but as noted in recent posts, this may not be straightforward. As a point of interest, would the Axpert have gotten more than average heat? Perhaps in the sun for some hours of the day? Heat seems to be the enemy of the capacitors involved. I hope that your Axpert will be running again soon.

Perhaps try Maximum_solar on Ebay: http://m.ebay.com/itm/Solar-power-inverter-5000w-48v-230vac-MPPT-solar-charger-80A-battery-charger-/162509114133 Edit: oops! The shipping mentions every land continent except Africa. But it may still be worth an enquiry.

Be aware that the Axperts are picky about speed sensing. A friend had some quiet fans in one, and there would be occasional locked fan errors. Enough to be a real nuisance, and the fan error could mask a more serious error. It may be possible to modify the sense circuit, but it was not worth the trouble, and a firmware upgrade basically fixed the noisy fans anyway. So they were swapped back to the original Adda fans. See http://forums.aeva.asn.au/viewtopic.php?title=pip4048ms-inverter&p=62186&t=4332#p62186 for details.

It seems to me that as long as you don't allow the battery to drop below 36 V, which is seriously low and very bad for the battery, and you have utility power or solar power available to keep it that way, then you can live without the TX6 power supply. But if you want to replace it, Plonkster's idea of using a DC power supply via D62 should work fine, as long as you can supply fifty odd volts at low current (100 mA or 5W should be plenty at a guess).

Yes, I believe that you have the power flow correct. I don't understand your comment "I believe this SMPS with TX6 must work in my case". Power for the control electronics can come from three sources: the battery, the utility, and the SCC. You have TX6 badly burned, so it can't be getting power through that. But that just means you need power from the battery to run the control electronics. (You can't use power from the SCC for other reasons.) I note that when utility charging at say 30 A, that doesn't come via the TX6 circuit; that's just a few watts to run the processor, LCD, and other control electronics. The 30 A comes through running the inverter in reverse (or effectively so). TX6 isn't needed for that; power can continue to come from TX9 where is usually comes from. When utility power is available, you can charge your battery from that source, as long as the inverter is in bypass mode. I note that the primary power supply (with Q36 and TX9) can be turned off by the processor. I believe it does that when it senses the inverter switch is off, shuts down in an orderly manner, then activates opto U8 (which I recently added the the primary power supply traced schematic on AEVA), which will kill power to the processor itself. It seems to me that you have two independent faults (likely both caused by the same event, but maybe not). One has burned the TX6 circuit, but you can survive without that (later models don't even have it), so that may have happened months ago and you did not notice it. The other is that the SCC fails to make its presence known to the main DSP, so it's as if the sun never shines. Somehow, the SCC is not talking to the main DSP, It could be because it is completely blown up and the SCC's processor doesn't have power, or it could be something on the inverter main board to do with comms to the SCC. There is a six or so pin connector between the processor daughter board and the SCC. Check that cable and any parts near it. There is a serial port on the main DSP that is dedicated to talking to the SCC's processor. They have to exchange commands for the SCC to be detected and charge the battery. Charge current is determined by the main DSP, not the SCC.

It's probably on your monitoring software's screen, or you can just use the up or down button to cycle through the display modes till you get to one starting with "U1".

Alas, no. I do my tracing on a main board kindly donated by someone who bought a replacement main board, but we repaired his original board. This one has blown a second time MOSFETs, but we still might repair it. Anyway, it's not running. Mine and a friend's are later models that don't seem to have the TX6 circuit at all. Ah, that might answer something I've been wondering for a long time. I always thought that you could run an Axpert without a battery, at least to change settings or update firmware, if you powered it from either mains or solar. But I can't get that to work on my own machine. Maybe only the old models (circa 2013) could run off the mains / utility, and later models simply can't because the TX6 circuit has been left out. So @maxo, maybe that's yet another alternative : leave out TX6 altogether, and it might still work! Just like later models, you'll always need a battery connected to use the LCD screen or to do flash updates.

Thanks for the details, @maxo. Indeed, the TX2 circuit is completely different to that of TX6; it seems that TX2 is part of the bus soft start circuit (pre-charge for the large 470μF 500 V bus capacitors C40, C41). I've edited my earlier post to correct the error. I'll publish a partial schematic trace on AEVA soon and put a link here.

I note that in later models, TX6 doesn't even exist; [ edit: thanks to my colleague Weber for pointing that out] they presumably use the circuit around TX2 instead. Perhaps it's worth tracing the circuit around TX2 to find out what it does. Please post whatever you find. [ Edit: No No! TX2 is the bus soft start circuit; it pre-charges the large 470μF capacitors to some 400 V before the DC-DC dual full bridges do their job. So TX2's secondary will be a much higher number of turns. Despite appearances, TX2 and TX6 are rather different. (Though maybe they could still be similar if primary and secondary are swapped. But best not to assume any similarity. ] Please also feel free to post corrections to my trace of the TX6 circuit, if your notes disagree with mine.

The partial schematic in this post may help a little: http://forums.aeva.asn.au/forums/forum_posts.asp?TID=4332&PID=65232&title=pip4048ms-inverter#65232 Also, TX2 (near TX6) may have similar characteristics, so you may be able to get characteristics from TX2 that you can't get from burned up TX6. Attempting to wind your own transformer deserves respect! But if you fail, it might be possible to get spare parts from a source like MPP Solar. They seem to be the main Ebay seller for Voltronic Power, and seem to have good access to the developers. Finally, if all else fails, @plonkster's clever idea of replacing the whole power supply has merit. You can see that its purpose is to supply square wave power to AC_PS_OUT, which I believe is mentioned in one of the service manuals. I'm not at my computer now, but from memory, that supplies isolated 15 V or so to some parts of the inverter. [ Edit: actually, it appears in this partial schematic here: http://forums.aeva.asn.au/forums/forum_posts.asp?TID=4332&PID=59548&title=pip4048ms-inverter#59548 So it seems to need a little more than battery voltage, so that diode D59 will be reverse biased, and the battery won't be used to supply power to the inverter control electronics when mains power is available. So something that supplied 55 V or so would be ideal. Alas, that's a bi unusual. Maybe a 48 V supply with an adjustable output turned all the way up would work. ] If you supply that, perhaps via a diode, [edit: you can use D62 for this, as long as TX6 is removed] using the same mains input that the circuit in the linked post, that should work. Suppliers like Mouser will have small AC to DC power supply modules at reasonable prices that could be made to fit. Taking out the 4 diodes D17-21 would prevent the existing power supply from drawing needless power.

There are some loads that inverters don't like. For my Axpert, it's a hot air gun. When it's running, the lights flicker markedly. I can imagine that with another inverter, it could cause buzzing instead of or in addition to flickering. My suspicion is that it has to do with the voltage regulation. A sharp load change, either higher or lower, necessitates a change of inverter output voltage to compensate. If that load changes back at just the wrong rate, it can set up a kind of instability. There isn't much to be done about that. For me, the flickering is just a nuisance, a bit more for my daughter who has had an epileptic event. She says it doesn't bother her, so I accept that and keep an eye on it. You could try some sort of mains filter to try and correct the current spikes. I have not looked into that as yet; it's not been necessary so far.

@Mohamed Arafa, is it always a steady 150 Vac between two phases? Might it slowly drift with time? This would tell if the comms cables are working but the firmware gets the phases wrong, or if the inverters are not talking to each other. I figure 38° between two phases, and 161° between the other two pairs. I can't think of any reason for this, or anything special about those phase angles. It might also be interesting to know what voltages you measure soon after going to battery mode when the mains is present. 220 V line to neutral implies line (bypass) mode. I think you should see close to 398 V line to line (sqrt(3) x 230 V) because the inverters should stay synchronised to the mains in battery mode when mains is present. Then keep measuring phase to phase while you open the AC input breaker (remove utility mains). Does it spring to 150/450 or slowly drift towards these figures or stay at 398 or something else?

Possibly, given enough time (I assume you are referring to the fan locked rotor problem here). There's just not enough incentive to justify the time required.

Actually, that is a concern with fan reversal. One of the three thermistors is on the PCB to one side of the main transformer: From http://forums.aeva.asn.au/forum_posts.asp?TID=4332&PID=60223&title=pip4048ms-inverter#60223 . That one could be affected by air flow, though the air flow is pretty much blocked by the transformer, so there is no "flow through" effect. But the fans are approximately where the camera is in the above photo, so by reversing the fans and blowing up (towards the HF transformer), you would be cooling the thermistor more, and therefore you may get less cooling than was needed or intended. But the temperature used is the maximum of the three sensors, not the average/mean, so having one drop out through air flow doesn't make much difference. There doesn't seem to be an issue with the HF transformer getting hotter after a fan reversal. Note that it's no longer recommended that the fans be replaced with ultra quiet versions, because there can be an issue with nuisance tripping due to fan "locked rotor" warnings and errors.

See here: http://forums.aeva.asn.au/pip4048ms-inverter_topic4332_post60086.html#60086 But note the shortcuts contained at the end of that post! It will void the warranty, since merely opening the case requires breaking a sticker that will void warranty. To me, there are so many reasons that the warranty has to be voided. But others may well see things differently. I like to be very self-reliant in this respect.

And Axperts are known for their short term overshoots, at least in battery charging. But with any grid feeding inverter, if a 2 kW load drops off suddenly, it would be very difficult to prevent a slight power flow to the grid for a cycle or two. Those meters should allow a little more more slack, it seems to me.

I would not know about that. It sounds pretty serious to me. Not every failure is accompanied by obvious blackening or blown leads visible on the outside. The repair manuals do say that if one MOSFET is blown, replace them all. But blown MOSFETs often also take out driver transistors, resistors, and even the driver chips. Also a good idea. They are suspected of causing MOSFET failures after as little as 3 months of operation. Yikes. You need to check any semiconductors connected to those capacitors. Wow. I was thinking the other day how tough the AC side seems to be; I've rarely heard of the IGBTs failing. Of course, lightning is a rather special case. Sorry, I've never tried. I just leave the heatsink there and replace the components bolted to it as needed. I believe that there are two slotted studs that poke out of the heatsink that are soldered to the board. So it would take a couple of large ("Big Bertha") soldering irons to remove those. Is it necessary to remove the heatsink to get access to some of the components? I've had to remove some of the large film capacitors temporarily for some repairs.

As a point of interest, what version of firmware did the older models with the heatsink on top come with? Can you remember when they were bought? Do they perhaps have a manufactured date tag? (Perhaps between the DC input terminals inside the cable cover?

There is a user in Australia with a very similar problem, who has contacted his supplier over this issue: http://forums.aeva.asn.au/forum_posts.asp?TID=4332&PID=65020&title=pip4048ms-inverter#65020 Would that be you? If not, it might be worth keeping an eye on developments there. As you can see, they haven't progreed past the "must be idiot user" stage yet. But sometimes things do get fixed via that route.

You can find the patch and discussion here: http://forums.aeva.asn.au/pip4048ms-inverter_topic4332_post64095.html#64095