PIP-5048GE/GK/MG/MGX inverters (with unsafe SCC)

[ccrisan]

No. The firmwares for the various models are usually quite different and not comparable. The first two digits (the major version number) indicate the model it's designed for; major version numbers can't be compared to judge age. Generally, minor version numbers (the last two digits) increase sequentially, by 1 for a minor update, and to the next multiple of ten for a major update. But even these you can't compare with minor version numbers from other models to compare age.

That makes sense. It was just strange that I wasn't able to find similar versions reported by users and I'm assuming that's because GKs may not be as popular as other models.

Generally, we try to list all the models that a firmware is compatible with, both MPPSolar model name and the Axpert model name. There is the "zoo" post to try to make sense of the various model names from the various resellers. Since the GK is not listed as being compatible, you should assume that it is not compatible. The GK/VM III has different hardware compared to the MG/MKS II; for example, the display is completely different (with its own processor in the case of the GK, but not for the MG), the processor is different and runs at a different speed, flash size is different, the MG/VM II is parallelable while the GK/VM III is not, and so on.

Oh that's a really useful post; it should be pinned or somehow made more visible/accessible. It certainly reduces the confusion created by so many models and vendors.

The only GK firmware I have access to (version 20.59), and every firmware I've examined so far, supports the BTA (BaTtery voltage Adjust) command. Certainly, "BAT+01" will not work, because you have transposed the first two letters. If that was just a typo from this post, check that you are sending the checksum correctly, or use Weber's clever commands that don't require CRCs with the sign bit set (as set out in this post).

Apologies for the typo. I tried in fact with the correct BTA commands (even with the simplified clever ones) but they don't seem to be supported by my GK device (firmware). It does appear to accept though the BTA0/BTA1 and BTA2 commands, but as far as I understand, they require two reference voltages for both upper and lower bounds of the battery range, which makes things more complicated.

In the meantime I contacted MPP Solar and, to my surprise, they were quick to answer asking for further details. I am now waiting for an "official answer" on how to adjust/calibrate battery voltage measurement on a PIP-GK unit.

[rthorntn]

Did we figure out if these inverters needed a Bender isoPV425 or a Type-B RCD...

They can't use the Bender unit for insulation monitoring, because the solar charge controllers aren't isolated. I can't comment on the RCD.

I assume you've read Webber's assessment of these models, available from the index in the first post of this topic.

Thanks, I have read Webber's assessment quite a few times.

[coulomb]

There is an issue with fault code 08 (error 08: bus voltage is too high), with certain types of solar panel, especially if the panels are grounded (as they should be). This has been discussed extensively on the South African Power Forum, and member APV finally found a solution for it, given in this post. See earlier posts in that topic for more details and discussion.

In summary, what seems to happen is that when it rains, leakage from the PV inputs to ground injects current into the DC bus due to the unisolated solar charge controller, and the fact that bus negative has a negative half mains cycle on it with respect to earth. The inverter firmware doesn't seem to be expecting this, and doesn't seem to take steps to mitigate against it, at least when in bypass mode. So the bus voltage increases slowly or more quickly depending on many factors such as where the average of the leak is, how wet the panels get, and so on. Eventually the bus voltage exceeds 500 V, and the inverter throws fault code 08 to protect itself, but dropping the loads in the process.

[dRdoS7]

There is an issue with fault code 08 (error 08: bus voltage is too high), with certain types of solar panel, especially if the panels are grounded (as they should be). This has been discussed extensively on the South African Power Forum, and member APV finally found a solution for it, given in this post. See earlier posts in that topic for more details and discussion.

In summary, what seems to happen is that when it rains, leakage from the PV inputs to ground injects current into the DC bus due to the unisolated solar charge controller, and the fact that bus negative has a negative half mains cycle on it with respect to earth. The inverter firmware doesn't seem to be expecting this, and doesn't seem to take steps to mitigate against it, at least when in bypass mode. So the bus voltage increases slowly or more quickly depending on many factors such as where the average of the leak is, how wet the panels get, and so on. Eventually the bus voltage exceeds 500 V, and the inverter throws fault code 08 to protect itself, but dropping the loads in the process.

Thanks for posting that.

So far, since I turned Utility charging on, first @ 2A, then 10A, I haven't had an Error 08. My system has spent a lot of time after sunset in Bypass mode since then. Both because more of the house runs of the system since I had a Transfer Switch installed, and there is less daylight, but also my faulty NiFes still haven't been replaced.

Also, not at home, but up near the NSW-QLD border, waiting for 11th July (we hope).

If I do have another Error 08, I'll give this a go. Or, maybe MPP solar will come out with a FW update. Or not.

Thanks,

dRdoS7.

[coulomb]

since I turned Utility charging on, first @ 2A, then 10A, I haven't had an Error 08.

I presume that you turned on utility charging specifically to overcome the bus voltage errors, correct? Did 2 A not suffice?

It might be a workaround, since utility charging pulls power away from the bus towards the battery. But it also may force the firmware to pay attention to the bus voltage.

I suspect that the only remedy for a too-high bus voltage when in line mode is to blow off some capacitor charge by pushing a little energy into the battery, running the inverter in reverse. In other words, a spot of utility charging. Setting the utility charger to charge at 2 A should force this. Unless your leakage was generating more than 2 A into the bus, which seems very unlikely.

[NMM]

[ Moderator note: This post was moved from the PIP-4048MS and PIP-5048MS inverters topic, since it pertains to a PIP-5048MK or clone or workalike (presumed). ]

RE-EDITED [ by NMM ]:

Hello Coulomb
Need some help.
After reading your post below [ edited Coulomb: the FAQ post, now removed ], it seems the closest situation I am facing.
Please be kind enough to assist with the following scenario:
I am using a PHOCOS 5KW-230V Inverter with 2 x US3000 Pylontechs and 14 x 410W Panels (2 strings of 7)
The PHOCOS software seems very similar to AXPERT, so the settings menus look almost identical and hence the below will be familiar to you.
I am having a problem with SBU Mode.
Menu 01 = SBU
Menu 02 = 70a (changeable, but pylon overrides this value according to circumstance....ive seen it at 10a, 20a or 70a)
Menu 05 = PYL (pylontech)
Menu 06 = Automatic Restart if AC output overload occurs - Enabled
Menu 07 = Automatic restart when over-temp occurs - Enabled
Menu 12 = 48Vdv (not changeable, as pylon overrides any changes)
Menu 13 = 52Vdc
Menu 28 = SGL
Menu 29 = 27Vdc (not changeable, as pylon overrides any changes)

Scenario:
Pv Input = 1.8Kw
Ac Output = 1.6Kw
The LCD will initially show PV & Battery Powering the load, which is correct as per SBU status.
Then randomly, it goes to GRID MODE. Load. On some occasions the MPPT points to Battery / or sometimes no PV arrows show at all.

Firstly, why does the status of powering the load change frequently from Off Grid Mode to Grid Mode, despite all conditions favouring Solar and Battery status.
And after it goes to Grid Mode....it stays on Grid mode either for 10 minutes at a time or sometimes indefinitely. Please explain the conditions of switching over to Grid Mode....and returning back to OffGrid. Assuming battery voltage has never gone below the Menu 12 value of 48V.
My SOC is always above 70% and mostly remains on 80%. Hence no chance of low voltage.
I am using a BMS Cable specifically for PHOCOS-PYLON, we know that they are communicating.
Following the above scenario, another anomaly: the LCD screen shows
Utility powering the load only.
PV is available 1.8kw
Battery is available 53.1v
Yet neither sources are shown to be powering the load (despite being in SBU mode)
Battery charging on LCD shows 0 Amps
Battery Status Indicator is flashing which suggests battery should be charging. So if it charging there should be some amps charging the battery and not zero.
Looking at the Phocoslink app simultaneously for the above scenario...it shows PV input, Ac input, and Battery being charged.
So all in all, the above situation is baffling and I'm failing to figure out what is causing this
Looking forward your feedback
Regards

[ Edited Coulomb: Removed quote of the entire FAQ post; no need for it here. ]

[dRdoS7]

since I turned Utility charging on, first @ 2A, then 10A, I haven't had an Error 08.

I presume that you turned on utility charging specifically to overcome the bus voltage errors, correct? Did 2 A not suffice?

2A worked OK. AFAIK. Hard to know for sure with an intermittent fault. I went to 10A as that is roughly equal to the house load over-night. So when the bank reaches Back to Grid setting (46V), they get a bit of charge, and the inverter goes back to battery mode (@ 56V) for a while longer. Not an ideal situation, but it will have to do until the faulty cells are replaced. I had no problems for 3 years.

According to NSW Fair Trading advice, after I contacted them last month, and I can request a refund for a major failure. Really, they only told me what I already knew from the ACL web site.

dRdoS7

[coulomb]

I am using a PHOCOS 5KW-230V Inverter with 2 x US3000 Pylontechs and 14 x 410W Panels (2 strings of 7)

This seems to be a PIP-5048GK (Axpert VM III) clone or workalike, or it might even be a relabelled genuine Voltronic Power machine.

Menu 29 = 27Vdc (not changeable, as pylon overrides any changes)

That seems a very strange value, but hopefully it's not causing problems.

Then randomly, it goes to GRID MODE. Load. On some occasions the MPPT points to Battery / or sometimes no PV arrows show at all.

That sounds like there is a high inrush load starting, that you don't notice watching the LC Display or the monitoring software (which sample the load power only every second or several). I assume you're using the power cables supplied with the Pylontech battery to connect to the inverter input, which hopefully won't have excessive voltage drop.

Firstly, why does the status of powering the load change frequently from Off Grid Mode to Grid Mode, despite all conditions favouring Solar and Battery status.

As above, it may be short term overload, or load causing a battery voltage dip below 48 V (the back to grid voltage). That's all I can think of at present. I have no experience with the direct to Pylontech BMS connection.

And after it goes to Grid Mode....it stays on Grid mode either for 10 minutes at a time or sometimes indefinitely.

Factory firmware will cause it to stay in grid mode, once it goes there, for a minimum of 10 seconds, even if conditions allow a return to battery mode.

Please explain the conditions of switching over to Grid Mode....and returning back to OffGrid.

As long as the low DC cutoff setting (setting 29) is at least 2.0 V less than setting 12 (back to grid voltage), it's quite simple: as soon as the instantaneous battery voltage goes below the back to grid voltage setting for a short time (some 100 ms from memory), it will switch to grid, if available. Once there, it stays for a minimum of 10 minutes. After that 10 minutes, if the conditions are suitable (i.e. the battery voltage is above the back to battery voltage setting), it should switch back to battery mode.

Assuming battery voltage has never gone below the Menu 12 value of 48V.

The problem is that you don't see the instantaneous voltage, and neither does any monitoring software. The battery voltage is unfortunately not compensated for battery internal resistance. So m guess is that as a fridge or other load with a high startup current comes on, the battery voltage must be drooping to below 48 V for a fraction of a second.

My SOC is always above 70% and mostly remains on 80%. Hence no chance of low voltage.

I'm saying that this may not be true at the fractional second time scale.

I am using a BMS Cable specifically for PHOCOS-PYLON, we know that they are communicating.

That's generally good, though I have no experience with it, to know if there are any wrinkles not ironed out.

Following the above scenario, another anomaly: the LCD screen shows
Utility powering the load only.
PV is available 1.8kw
Battery is available 53.1v
Yet neither sources are shown to be powering the load (despite being in SBU mode)

Your model (as with most PIP/Axpert models other than the PIP-5048MK/Axpert King) can't power the loads from the grid and battery/PV at the same time; it has to be one or the other. Hopefully, it stays mostly in battery mode, but of course this is your present problem.

Battery charging on LCD shows 0 Amps

That makes no sense, as the 1.8 kW of PV power can only be going into the battery. Are you sure that you're not reading battery discharge current? In unpatched firmware, charge current and discharge current are separate. Different models seem to treat charge current differently too; sometimes it's panel current (before the SCC), and sometimes it's battery current (after the SCC).

Battery Status Indicator is flashing which suggests battery should be charging.

That's another indication that the charge current can't be zero.

So if it charging there should be some amps charging the battery and not zero.

Exactly.

I don't know how to proceed from here. Perhaps start with double checking the battery charge current; perhaps show a photo of the LC Display.

Also, can you nail down a scenario where you're sure it should be going back to battery, it's been more than 10 minutes since it changed to grid mode, and isn't switching back to battery?

[ Edit: "Axpert MKS III" → "Axpert VM III" ]

[NMM]

attachments from previous post

[coulomb]

Wow. While the above is obviously heavily inspired by the PIP-5048MG (Axpert VM III), it seems to have some enhancements. The LC Display is different, and the area to the right of the display seems to suggest that you can change output source priority with a button, and see which priority is in effect at a glance. I'm also not familiar with the button near the bottom left; does that initiate a battery charge?

So the firmware would have to be different to support all this. Does your model have the equivalent of the U1 and U2 displays, reporting the main firmware versions? If so, what do they report?

This may yet be made by Voltronic Power, either with custom enhancements like the Inverex models, or just a new model (so others will end up with these features too). Or it may be a work-alike made by Phocos (like the Growatts). The firmware version numbers may give a clue as to which.

[NMM]

Hello Coulomb
Seems my earlier post prior to the post with attachments, does not appear.
I will have to repost.

With regards to your comments:
Yes, the Right Button is an AC OUTPUT Override Timer, so we are able to set 1 priority timer for specific parts of the day, in any mode we wish: SUB, SBU, USB.
This way, it overrides the Priority in Menu 01 for that period. (very handy)
The Left Button is the BATTERY CHARGING Override Timer, like the above, we can set a different priority for a chosen period in the day.

The Firmware does have U1 and U2 as follows:
U1: 30/03
U2: 60/04

PS. I attached an Excel Datalog in previous post. Kindly peruse and possibly it may give you a better indication as to why it keeps switching to Grid Mode?

Many thanks

[coulomb]

Yes, the Right Button is an AC OUTPUT Override Timer, so we are able to set 1 priority timer for specific parts of the day, in any mode we wish: SUB, SBU, USB.
This way, it overrides the Priority in Menu 01 for that period. (very handy)
The Left Button is the BATTERY CHARGING Override Timer, like the above, we can set a different priority for a chosen period in the day.

Interesting. I saw some hints about this in the Voltronic firmware, but it didn't seem to be fully implemented.

The Firmware does have U1 and U2 as follows:
U1: 30/03
U2: 60/04

Ah. That is so different to the Voltronic firmware numbers that I'd say Phocos have been doing their own.

PS. I attached an Excel Datalog in previous post. Kindly peruse and possibly it may give you a better indication as to why it keeps switching to Grid Mode?

I did miss that amongst the giant screendumps.

Firstly, your data logging seems to have some weird artifact whereby almost every second entry has the exact same data, apart from the time stamp: always exactly 2300 VA apparent load power, battery mode, 50.0 Hz. Surely it doesn't switch from battery to line mode every 15 seconds?

So when I ignore that, I see that nearly every time it switches to line mode, with one exception that I noticed, the output load is over 3 kW. If that load has a high startup surge power, that could definitely be tipping the inverter to line mode. Perhaps try to figure out what that load is, and consider putting it directly on the grid, not the AC output of the inverter.

I note that Pylontech US3000s are happiest at 37 A max (charge or discharge), so that's 2 x 37 x ~50 = ~3700 W from the battery or about 3.4 kW of load. A third battery module might lower the paralleled internal impedance to the point where your largish load doesn't trigger the switch to line mode.

[NMM]

Firstly, your data logging seems to have some weird artifact whereby almost every second entry has the exact same data, apart from the time stamp: always exactly 2300 VA apparent load power, battery mode, 50.0 Hz. Surely it doesn't switch from battery to line mode every 15 seconds?

Yes, the Data-logging did look weird to me, as well. However, what it probably means is that when it was in Grid-Mode, both Utility and Battery were supplying the load simultaneously (i.e both the arrows were pointing to load on the LCD) hence the simultaneous Line Mode/Battery Mode logging.
We can consider it being Grid-Mode anyway as on the actual Watchpower App, it logged it as LINE MODE only (only the export to Excel shows it this way)

So when I ignore that, I see that nearly every time it switches to line mode, with one exception that I noticed, the output load is over 3 kW. If that load has a high startup surge power, that could definitely be tipping the inverter to line mode. Perhaps try to figure out what that load is, and consider putting it directly on the grid, not the AC output of the inverter.

Amongst other loads, yes, I have a water booster pump (750w / 5A) which has an in-rush current of 16A and drops to 5A and
a Borehole pump which has an in-rush current of 17A and drops to 7A
Generally the borehole pump would be continuously on as we would be watering the lawn....so this in itself does not cause the Switchover...
Booster Pump comes on only when the house indoor taps are turned on, in turn activating the booster pump until taps turn off...
So combining the 2 at any given moment, it could peak 16A+7A = 23A....for a millisecond.

And continuing on that.....by deduction that the the voltage drop in the above scenario is probably dipping below Menu 12 Setting of 48Vdc, Perhaps we could lower Menu 12 = 47Vdc or 46Vdc for the purpose of preventing switching to Grid-Mode? What would be the absolute minimum we could set Menu 12 at?
But I suppose any lower value below 48Vdc would offer no protection to battery in the real event that the battery reaches a discharge voltage of 48Vdc which in that scenario we WOULD want it to switch to Grid-Mode to maintain the DOD.......unless we rely on the Battery BMS to do its thing?
(Menu 29 is fixed at 47Vdc)

Evidently when AC Input is disconnected, it works well (no AC to switchover to), and would be the correct operational scenario as instantaneous voltage drop should not be a reason to switchover....

I note that Pylontech US3000s are happiest at 37 A max (charge or discharge), so that's 2 x 37 x ~50 = ~3700 W from the battery or about 3.4 kW of load. A third battery module might lower the paralleled internal impedance to the point where your largish load doesn't trigger the switch to line mode.

This could work....but was hoping there was another "cheaper/easier" workaround?

[ Edited Coulomb: fixed quoting ]

[coulomb]

Yes, the Data-logging did look weird to me, as well. However, what it probably means is that when it was in Grid-Mode, both Utility and Battery were supplying the load simultaneously (i.e both the arrows were pointing to load on the LCD) hence the simultaneous Line Mode/Battery Mode logging.

No, that same data turns up nearly every second entry when in battery mode as well. It seems like a glaring bug.

Amongst other loads, yes, I have a water booster pump (750w / 5A) which has an in-rush current of 16A and drops to 5A

That would definitely do it. I'm surprised that the peak current isn't even higher; generally I find a 7:1 ratio of inrush:run current.

But I suppose any lower value below 48Vdc would offer no protection to battery in the real event that the battery reaches a discharge voltage of 48Vdc which in that scenario we WOULD want it to switch to Grid-Mode to maintain the DOD....

Exactly.

...unless we rely on the Battery BMS to do its thing?

The problem is that when the battery BMS protects the battery, it disconnects from the inverter, which may damage the inverter, and also you lose your loads. When the inverter protects the battery, you get a clean transition to grid power, and you don't lose your loads.

instantaneous voltage drop should not be a reason to switchover....

Very short term voltage drop should not be a reason, or at least a voltage drop that is caused by a short term heavy load. In our patched firmware, Weber has added load compensation, so that it adjusts the measured battery voltage to compensate for the present load (or charging, a sort of negative load). But there is no patched firmware for these models, and patching every variant is just crazy. We're unlikely to do another major patching exercise. It would be great if we could submit feature requests to the manufacturer(s), but they don't seem to be interested.

A third battery module might lower the paralleled internal impedance to the point where your largish load doesn't trigger the switch to line mode.

This could work....but was hoping there was another "cheaper/easier" workaround?

There must be economical motor soft starters out there; it's never bothered me enough (I run patched firmware) to find one and try it. Perhaps you could find one that you could just insert between your water booster pump and the mains; perhaps no re-wiring would be needed, just plugging in.

If one can't be found, perhaps a small single phase VFD (Variable Frequency Drive) could be configured to simply apply a speed ramp (say 0 -100% in 1 or 2 seconds) to the pump at turn-on. But it would have to somehow sense that the motor has turned off, unless you could insert it between the pressure switch and the motor.

It might be important to choose a soft starter or VFD that has a decent power factor; simply full wave rectifying the mains to create the bus voltage for the motor inverter could upset the main inverter. So look for one with a PFC (Power Factor Correction) stage. Whether that will turn out cheaper than another battery module, I don't know.

[Roman khan]

Hi every one
I have an axpert vmii 3kw (PIP-5024GE) inverter. I repair its DC-DC side,Bus side and inverter proper side.with all drivers.for the first time when i turn ON it with only batteries it blew up the two transistors 1020 and 2655 near IC SG3525. I repair it again and turn it ON,now during startup whenever out put voltage comes to appear on screen, it shutdowns and trips battery breaker.any body please explain how to solve this problem.

[coulomb]

now during startup whenever out put voltage comes to appear on screen, it shutdowns and trips battery breaker.

Since it only trips when the inverter comes on (at least that's what I assume from your description), it may be a MOV device that has had too many surges. They seem to degrade in a way that lowers their turn-on voltage, so they could effectively be shorting the inverter output.

See this post and another a few posts later with speculation about the MOVs and suggested replacement part numbers.

[Roman khan]

Hi
I have axpert Vmii 5kw(PIP 5048 GE,max500 voc) inverter , i want to connect lithium ion battery ( Vision 100Ah) with this inverter now i have some questions.
1) Is communication cable must in this case
2) if yes then ,which communication cable I should use
3) What should be my inverter settings for max charging(02), battery type(05 no),max utility charging(11 no),12 no, 13no 16no.
4)how i can calculate its charging time and backup on 1000watt load?

[coulomb]

I have axpert Vmii 5kw(PIP 5048 GE,max500 voc) inverter , i want to connect lithium ion battery ( Vision 100Ah) with this inverter

One 100 Ah battery is a bit light for a 5 kW inverter.

1) Is communication cable must in this case

My understanding is that PIP/Axperts still only talk directly to Pylontech brand batteries, and after a quick search I didn't find evidence that the Vision LFP batteries are Pylontech protocol compatible. Also, a VM II doesn't have an RS-485 port, and therefore no support for direct connection to a battery.

3) What should be my inverter settings for max charging(02), battery type(05 no),max utility charging(11 no),12 no, 13no 16no.

Battery type will have to be USE. The rest depend on whether Visions are 15S or 16S. Please post a link to a datasheet.

4)how i can calculate its charging time and backup on 1000watt load?

I'll assume that its loaded voltage is around 50 V. So 50 V x 100 Ah = 5000 Wh. Using a maximum discharge depth of 80% that comes to 4000 Wh usable when new. A 1000 W load at 93% efficiency comes to 1000/9.93 ≅ 1075 W. 4000 Wh / 1075 W = 3.7 h. So around 3 hours after various effects such as inverter self consumption, temperature, age, cycle efficiency are taken into account.

[ojeysky]

Please I have an emergency, I am getting a error 32 communication failure!!!! I am off-grid and really need this to come back on!

So the only solution to this was to replace the entire main boards, my inverter just came back online. The only thing that I did on that faithful day was to activate equalisation since the GK was going to premature float, I wonder if that was what caused the board to get burnt. Unfortunately it seem am stuck with the float bug on this thing hence am using 28.2/27.6 for my bulk/float settings.... higher but I can't see another option for now.... Any suggestions?

[ojeysky]

Well, the icons from the user manuals do seem to indicate

• "grid-interactive" for GK
  double-conversion for MK

But how much can we trust these to be accurate depictions of what happens under the hood?

I can confirm that this is what happens with my 3024GK

[ojeysky]

OK. Well that is some evidence that the GK does undeclared grid interactivity. Enough not to take the risk, but not enough to find them guilty beyond reasonable doubt.

It seem to me that it's well written on the manual as referenced in previous post. Can you clarify which one is undeclared?

Is there someone out there with a PIP-GK or an Axpert VM III who can tell us whether the output voltage is 230 V, or is the same as the grid voltage, when in that mode, as indicated in the manual, above left, where the load is supplied from both solar and utility at the same time?

I can confirm that when in bypass mode the output voltage is that of the grid, is there another mode you need me to check?

[ojeysky]

Hi guys, just thought I'd confirm some of your suspicions about the GK/VMiii. I've had a VMiii 3kw/24V for about 3 months now, hooked up to 6x340W panels (in one string) and 200ah of LiFePO4 cells with an Ant-bms.

The VMiii allows blending of solar and grid with or without the battery. It locks to the grid voltage as far as I can see. It will also charge the battery with solar and use excess solar to assist the grid. As stated, when in SBU it will separate from the grid and the voltage is pretty consistent at the set voltage (230V in my case). Without the battery hooked up, I noted my logging recorded some "export" power, usually a few Wh per day. But my prepaid meter has never tripped or given any obvious faults... With the battery hooked up this seems better.

I really can't fault this inverter so far - behaves as expected (once you've figured out some of the tricks in the settings). I've been logging inverter and BMS through a Raspberry Pi and doing my control with Home Assistant. My one complaint is the current measurement with low solar inputs. I can see the load on the battery reducing as solar starts up in the morning, but the inverter only registers any solar when it's at around 80W. A small thing, I know, but it messes with my logging

We have similar set-up, mine is the 3024GK, and I also just got a ANT BMS, can you share how you are logging the BMS data? Do you have the float bug on yours as well? which causes the inverter not to get past the float voltage and what's your float/bulk settings?

[NMM]

"Your model (as with most PIP/Axpert models other than the PIP-5048MK/Axpert King) can't power the loads from the grid and battery/PV at the same time; it has to be one or the other. Hopefully, it stays mostly in battery mode, but of course this is your present problem."

Hi Coulomb, not sure if I replied to this...
Though phocos does allow mixing of Sources to power the load. If in SBU mode.... if solar not sufficient to power the load then Battery tops it up....
If in SUB mode, likewise....if solar not sufficient then Utility tops it up.

Just yesterday on another phocos inverter, strangely the same scenario whereby PV was available, when changed to SUB mode with SNU charging mode..... the PV input drops completely to Zero (as if its not available).
When turning it back to SBU mode....PV returns.
I'm confused why this happens?

Any idea?

Thanks

[coulomb]

If in SBU mode.... if solar not sufficient to power the load then Battery tops it up....

All models do that, although the exact mechanism by which it is achieved varies in detail.

If in SUB mode, likewise....if solar not sufficient then Utility tops it up.

Yes, this is a mixing that I keep forgetting about, sorry.

Just yesterday on another phocos inverter, strangely the same scenario whereby PV was available, when changed to SUB mode with SNU charging mode..... the PV input drops completely to Zero (as if its not available).
When turning it back to SBU mode....PV returns.

The last part is easy: in SBU mode, Utility is last priority, so it doesn't get used and doesn't "steal the load" from the solar.

But in SUB mode with SNU charging, consider the following diagram I've posted before; ignore the red and blue parts, they relate to other models:



In order to allow for the possibility of utility charging (e.g. cloud suddenly drops PV power, utility charging should take up the difference), you have to be in line mode, i.e. the two switches have to be on; that's the only way for utility power to get to the battery. But if the battery doesn't need much power (say it's full or nearly so), then the inverter will be pumping very little power into the battery. In fact, as long as there is a little PV power, it need not pump any power from AC-in and AC-out (connected together by the top switch) at all.

So why doesn't it switch to forward power flow and pump power towards the loads? It would be pushing power into the AC-in port, even though more than that is already going to the loads. In fact it could push just enough power to make the utility power zero, and that's what you as an electricity paying consumer would want. But this is exactly what a grid interactive inverter does, but without all the safety checks and certifications. What happens if the load drops off suddenly? Power would be pushed back into the AC-in wire, potentially back to the grid. This is usually no big deal in Australia, perhaps unless you are on a SWER line. However, in some countries like South Africa, they have meters that literally cut the power if they detect more than a few hundred watts of feed in for a second or two.

My guess is that they want to avoid that, and they don't want to have one firmware for one country and different firmware for another, so they don't allow the inverter to work in forward power flow mode in this operating mode. Perhaps change to solar only charging and see if that fixes the problem. The inverter-charger could then be in battery mode, which has the top switch off, and there is no possibility of pushing power into the AC-in port (or using any of that power either). So there is no problem using PV power to its full extent

[weber]

Well, the icons from the user manuals do seem to indicate

• "grid-interactive" for GK
  double-conversion for MK

But how much can we trust these to be accurate depictions of what happens under the hood?

I can confirm that this is what happens with my 3024GK

Thanks for confirming that.

OK. Well that is some evidence that the GK does undeclared grid interactivity. Enough not to take the risk, but not enough to find them guilty beyond reasonable doubt.

It seem to me that it's well written on the manual as referenced in previous post. Can you clarify which one is undeclared?

By "undeclared" I was referring to the fact that the manufacturer doesn't admit, anywhere in the manual, or in any sales blurbs, that the PIP-GK can export power to the grid, out of its AC IN port, as you and others have confirmed it can. This is something that utilities care about very much. Inverters that are capable of doing that, have more strict safety and harmonic-distortion requirements placed on them (e.g. AS/NZS 4777) for very good reasons.

Is there someone out there with a PIP-GK or an Axpert VM III who can tell us whether the output voltage is 230 V, or is the same as the grid voltage, when in that mode, as indicated in the manual, above left, where the load is supplied from both solar and utility at the same time?

I can confirm that when in bypass mode the output voltage is that of the grid, is there another mode you need me to check?

There isn't any mode called "bypass mode" in the PIP-GK manual, but I assume you mean the sub-mode of "Line Mode", shown on the left with the red underlining, in the following image.



If that's the mode you're talking about, then no, no need to check any other mode. Thank you for this information.