PIP-4048MS and PIP-5048MS inverters

[coulomb]

It's definitely Lithium Ion (not Lithium Iron Phosphate) so 3.7v per cell.

Ah. Actually, lithium ion is the umbrella term; LFP is a form of lithium ion battery, with an LiFePO₄ cathode. Other lithium ion batteries have cobalt or NMC or other cathode materials. Hence my confusion over what you are using.

So does the LC flavour give a more realistic battery % indicator for Lithium Ion batteries on the front display?

[ Edit: As Weber points out in the post after next, I had quite forgotten an improvement that I suspect he suggested and largely implemented in the fully patched firmwares. So the following paragraph and short answer are quite wrong. ]

Marginally, in that the wild guess is a load compensated wild guess. The guessed SOC in percent is still basically the number of tenths of a volt above the DC low cutoff setting (setting 29) minus 2%. You might be able to find a value for setting 29 that provides an acceptable indication. Non-LFP lithium ion batteries tend to range in cell voltage from about 3.6 to nearly 4.2 VPC, so that's about 0.6 VPC. For a typical 14S battery, that's about an 8.4 V range. The firmware assumes about 9.5 V (for 0% to 95%), so that's about a 10% scale error, or just over.

Short answer: not really.

Great. I may never connect solar panels to it as I already have 3 other grid connected inverters for that.

Heh. You'd think that would be enough...

So would I set the utility charging value to max (30A) and then control the actual utility charge level with the MNCHGC command?

Yes, though I think you'll find that the maximum is 60 A, and the default is 30 A. But yes, that is the idea.

How risky is it to update the firmware? I understand you can completely brick the inverter if it goes wrong?

99% of the risk is choosing the wrong firmware to update to. The inverter firmware design eliminates some bad combinations (by helpfully refusing to update, and not display anything to indicate this), but there are combinations that are not checked for, that can in fact end up with a bricked machine.

About 10% of the time (really wild guess), some machines (I have no idea what proportion) will appear to be bricked when they are not, due to a communications failure during the updating. The problem with a bad update is, you are only running the bootstrap loader, which doesn't bother talking to the LC Display or even the LEDs. Maybe they could have turned all the LEDs on or something simple, but they haven't bothered. That can be scary, because your machine appear to be so very dead. However, calmly starting another upload will fix the problem, at least all the times it has happened to myself or people I know. It seems that some of the communications boards are a touch marginal under certain conditions.

In the very worst case, you have to buy another control board from your supplier, and that could take some weeks to arrive, as well as cost some 10% of the inverter's purchase price. Then you try it again

Short answer: the risk is slight.

[ Edit: "plus 2%" → "minus 2%". Sigh. Also adjusted the maths slightly to suit. ]

[derBastler]

Hey coulomb,

I work for a company in the field of Li-Ion battery technology. We have tons of Li-Ion-Batteries (NMC and NCA, so 3,7V per cell).

we have VW E-Golf Batteries, VW Passat GTE, VW E-Up, BMW i3 and all other manufacturer batteries. A very typical batterie nominal voltage is 44,4V. Upper limit 50,2V, Lower Limit 38,4V

It would be great to change the PIP-5048MS to this values to bring a hugh mass of batteries into second use applications.

Do you think this is possible?

[weber]

So does the LC flavour [of the patched firmware] give a more realistic battery % indicator for Lithium Ion batteries on the front display?

Marginally, in that the wild guess is a load compensated wild guess. The guessed SOC in percent is still basically the number of tenths of a volt above the DC low cutoff setting (setting 29) minus 2%. You might be able to find a value for setting 29 that provides an acceptable indication. Non-LFP lithium ion batteries tend to range in cell voltage from about 3.6 to nearly 4.2 VPC, so that's about 0.6 VPC. For a typical 14S battery, that's about an 8.4 V range. The firmware assumes about 9.5 V (for 0% to 95%), so that's about a 10% scale error, or just over.

Short answer: not really.

Coulomb is quite mistaken here. Our patched firmware gives a much more realistic battery SoC indicator on the front display, for cobalt-blend lithium ion batteries. Our SoC figure is not calculated in the way that Coulomb claims. Instead it treats the Float voltage setting as 100% and the Low Voltage Cutoff setting as 0% and scales the SoC proportionally in between, based on the load-compensated battery voltage.

Estimated_SoC = 100% × (Vbat_comp - Vcutoff) / (Vfloat - Vcutoff)

Of course it's nowhere near as good as a coulomb counter. It's maybe ±10%, assuming you have set the float and cutoff voltages correctly, and you have set the strength of load compensation ("kettle" compensation) correctly, as described in the AussieView™ and KettleKomp™ manual.

How risky is it to update the firmware? I understand you can completely brick the inverter if it goes wrong?

See Coulomb's post: If you think your inverter is bricked.

[coulomb]

A very typical batterie nominal voltage is 44,4V. Upper limit 50,2V, Lower Limit 38,4V

It would be great to change the PIP-5048MS to this values to bring a hugh mass of batteries into second use applications.

Do you think this is possible?

It would be possible, but I think you would not be able to reach full specifications. For example, maximum power output might be around 4500 W, reflecting the ~10% lower battery voltage.

It would be a huge job. At a wild guess, it would only be possible if some company was to sponsor the development cost. It might make more sense to approach the manufacturers about a custom model. I have no idea what this would cost, but I'd say it would be way more than could be justified for hobbyist use.

If you had a commercial source of second life batteries and needed a cost effective inverter-charger to complete the business plan, and were expecting considerable volumes, it might be viable.

Short answer: not really, unfortunately.

[weber]

I work for a company in the field of Li-Ion battery technology. We have tons of Li-Ion-Batteries (NMC and NCA, so 3,7V per cell).

we have VW E-Golf Batteries, VW Passat GTE, VW E-Up, BMW i3 and all other manufacturer batteries. A very typical batterie nominal voltage is 44,4V. Upper limit 50,2V, Lower Limit 38,4V

It would be great to change the PIP-5048MS to this values to bring a hugh mass of batteries into second use applications.

Do you think this is possible?

So this is 12 cells in series (12S) where normally folks would use 13S or 14S (of NMC or NCA) with the PIP inverters.

I think that our load compensation (KettleKomp™) in our patched firmware, if correctly set up, would make it possible to just barely operate with 12S of NMC or NCA cells. You would set the cutoff voltage to its lowest value of 40 V and you would set the float voltage to its lowest value of 48 V. Absorb voltage could be 50 V.

There's no harm in trying it.

It might benefit from also setting the AC output voltage down to 220 V (instead of the standard 230 V) to reduce clipping of the output sine wave when the battery voltage is low, which can create radio interference and is bad for motors. Or it may be that the PIP handles that automatically.

[coulomb]

... would make it possible to just barely operate with 12S of NMC or NCA cells. You would set the cutoff voltage to its lowest value of 40 V...

I think the biggest hassle would be that the firmware will treat the battery as "weak" (with several implications, including a low battery warning and switching to bypass mode if possible) when the battery voltage drops to below 42.0 V (2.0 V above the low DC cutoff voltage, setting 29). It won't become non-weak until 4.0 V above that setting, so 44.0 V. That's nearly its nominal voltage (@derBastler mentioned 44.4 V). We do know how to change those values (the 2.0 and 4.0 V thresholds), but that would mean yet another patched firmware version or flavour to maintain.

But yes, worth a try.

[weber]

I think the biggest hassle would be that the firmware will treat the battery as "weak" (with several implications, including a low battery warning and switching to bypass mode if possible) when the battery voltage drops to below 42.0 V (2.0 V above the low DC cutoff voltage, setting 29). It won't become non-weak until 4.0 V above that setting, so 44.0 V. That's nearly its nominal voltage (@derBastler mentioned 44.4 V).

But in our patched firmware, all those comparisons use the estimated open-circuit voltage of the battery, not the actual battery voltage. Figure 1 here shows that for 12S open-circuit, 40 V (3.33 Vpc) is about 1% SoC, 42 V (3.5 Vpc) is about 7% SoC and 44 V (3.67 Vpc) is about 40% SoC. Those are not unreasonable SoC thresholds for the behaviour you describe.

I agree we don't want to maintain yet another version.

[derBastler]

I think the biggest hassle would be that the firmware will treat the battery as "weak" (with several implications, including a low battery warning and switching to bypass mode if possible) when the battery voltage drops to below 42.0 V (2.0 V above the low DC cutoff voltage, setting 29). It won't become non-weak until 4.0 V above that setting, so 44.0 V. That's nearly its nominal voltage (@derBastler mentioned 44.4 V).

But in our patched firmware, all those comparisons use the estimated open-circuit voltage of the battery, not the actual battery voltage. Figure 1 here shows that for 12S open-circuit, 40 V (3.33 Vpc) is about 1% SoC, 42 V (3.5 Vpc) is about 7% SoC and 44 V (3.67 Vpc) is about 40% SoC. Those are not unreasonable SoC thresholds for the behaviour you describe.

I agree we don't want to maintain yet another version.

Okay, your open-circuit compensation for the batterie voltage is definitely a big first step.

Why is it so complicated to set the limits other values? I am an electrical engineer with proper skills in microcontroller and communication stuff. I would be willing to learn to deal with this topic and maintain a NMC/NCA firmware for all these applications. Yesterday i disassembled one of the firmware files with IDA and checked what's up inside, but without your knowledge this is a lot of work.

12S NMC/NCA batteries from cars will became common case in future. In europe you can buy them for private use if you want to (exmpl https://ecaraccu.nl/) and I know that there will be more dealers for this batteries in future. 2xTesla Model S Modules(22,2V) in serial are 12S, too. That would be a good deed for the community.

What would be a reason for you to set up a firmware like this? I could send you 34kWh (one BMW i3-94) Li-Ion-Batteries of this type for motivation

[coulomb]

Why is it so complicated to set the limits other values?

One of the reasons is that the settings are packed into CAN packets and sent from one machine to another, when they are paralleled and/or in three-phase. You have to keep track of the new bit pattern meanings, and make sure that all the uses agree with each other. Once you know where to look, it's not all that bad.

I am an electrical engineer with proper skills in microcontroller and communication stuff. I would be willing to learn to deal with this topic and maintain a NMC/NCA firmware for all these applications. Yesterday i disassembled one of the firmware files with IDA and checked what's up inside, but without your knowledge this is a lot of work.

Check for a PM; we may be able to work together on this.

What would be a reason for you to set up a firmware like this? I could send you 34kWh (one BMW i3-94) Li-Ion-Batteries of this type for motivation

That's certainly motivational. Let's see how things go.

[Wulfee]

Once you have updated the firmware in your PIP, is it possible to revert back to the original (does the update create a backup of your existing firmware first) or is this a one way trip?

Will the modified firmware remove any features from the PIP? What I'm trying to say is that my PIP is the 64V version and if I change the firmware will I lose that 64V ability? Any other similar values that could be reduced (say Max charge current or similar)?

Thanks for all the advice .

[weber]

Once you have updated the firmware in your PIP, is it possible to revert back to the original (does the update create a backup of your existing firmware first) or is this a one way trip?

No. Unfortunately, the update does not create a backup of your existing firmware. The existing firmware is copy-protected.

The zip file containing our patched 64 V firmware (72.20e) does however also contain the unpatched firmware of the same version (72.20), so you can at least revert to that if it turns out there is some bug in one of our patches that somehow got past testing. But of course this is not the same version that is in your inverter now (72.30). It can be a tough decision when you need to downgrade if you want to benefit from our patches, because we haven't yet obtained a copy of the latest version that is in your machine.

Will the modified firmware remove any features from the PIP? What I'm trying to say is that my PIP is the 64V version and if I change the firmware will I lose that 64V ability? Any other similar values that could be reduced (say Max charge current or similar)?

Since your PIP has the 64 V option, you can't load any non-64 V firmware. You should only load LC1_72.20e if you load anything at all (LC1 since you have lithium cobalt-blend cells, and 72.20e because that's our latest patched firmware for the 64 V PIP-5048MS). This will certainly not remove any features like 64 V ability or max charge current. In fact it only adds features and fixes bugs, relative to a PIP-5048MS(64V) with the factory 72.20 firmware. The only unknown here is what did they change between 72.20 and 72.30?

[Wulfee]

The only unknown here is what did they change between 72.20 and 72.30?

OK, I think I'll take the hit and do it. If they have added anything (unless it gives the dynamic charge control that your firmware offers) then it's nothing that I'll miss since I mainly want to tell the unit to charge and at what rate dynamically. I need to get my control software written up and tested first so that I'm ready to utilise it.

Thanks again.

[Wulfee]

Is it possible to change the AC voltage from the default of 230v to 240v. Some posts on here seem to suggest it is by using POPV2400 command but when I tried this it did nothing?

[weber]

Is it possible to change the AC voltage from the default of 230v to 240v. Some posts on here seem to suggest it is by using POPV2400 command

Yes, that's correct.

but when I tried this it did nothing?

Did you at least get a "(NAKss" in response? If not, how are you communicating with the inverter? What inverter port? What cable? What terminal software?

If so, then did you send the correct CRC characters as described here?
http://forums.aeva.asn.au/viewtopic.php ... 646#p66646

[Wulfee]

I'm using a python module and it works fine for all other commands I am sending. Not sure it handles NAKs too well, (seems to crash) I'll look into that but when I do any command that gets ACKed I get the ACK printed to the screen. I'm assuming that cos I don't see an ACK that I got a NAK. I've tried turning the battery switch at the bottom of the unit off before doing the command but it made no difference.

[paulvk]

Been working with the Nextion display.
Not have it working with inverters yet as I have to write code to go into an atmel AVR
to sit between the display and inverter but getting the graphics together
Note the yellow dots are animated just like watch power but I had to build this screen myself even though it looks the same just looked for icons on web probably as they did!
There is another display by 4D systems similar but might be able to stand alone without a micro in between
may look into them later.
Note this is a touch screen so it will be able to control the inverter not just display parameters.

[Mantas]

Hi All,

My one axpert 5kva/4000W inverter blew and was just replaced through my insurance but the contractor supplied a different branded one but it's still an axpert (Same stickers on the side). This new one however is a 5kva/5000w and the solar charger is rated higher at 80A vs 50A on the old one. The firmware on the new one shows as LC1 72.70. The old one and my parallel unit is on firmware U1 72.60. Can I upgrade the U1 to 72.70 and put them back into parallel? OR do I flash the LC1 72.70 to UI 72.60? I have 4 x 200ah flooded deep cycle batteries, not Lithium.

Currently I have the new one in Single mode and nothing but the batteries connected to the parallel unit just to give it juice.

I appreciate any feedback.

[weber]

Hi @Mantas. Welcome to the forum. Thanks for giving all the needed detail up front, to enable us to help you.

The replacement you have been supplied with, is not a rebranded Voltronic Power Axpert MKS 5K but a pirate copy or "clone". See Coulombs post:
Do I Own a Clone? (Oh, moan!).

And I'm guessing its firmware is actually LC1_72.70c. Note the "c" on the end. The pirates are unethically using firmware patched by Coulomb and I!

Even if you had been supplied a proper rebranded Axpert MKS 5K, it would be a 5 kVA/5 kW (PF1) variant, as the 5 kVA/4 kW (PF0.8) variants are obsolete. This causes problems for being able to parallel with the remaining 5 kVA/4 kW variants. See Coulomb's other posts:
Parallel compatibility of various firmware versions
Can I update my firmware, and if so, to what?
The only firmware that will allow 5kW to be paralleled with 4kW, while fixing the premature float bugs, is PB1_72.70b. You should load this into all your parallel machines.

However the clones typically do not allow their firmware to be changed at all! We suspect they do not even contain a bootstrap loader. So that's the first thing to find out. You should attempt to reflash the new one with PB1_72.70b. You may need to ask the contractor to swap it for for a real Voltronic.

[weber]

@Mantas I just realised you have another (and better) option, that doesn't require you to reflash the new clone. You can reflash the old inverter(s) with the same firmware that is in the clone, which I assume is LC1_72.70c.

[Mantas]

@Mantas I just realised you have another (and better) option, that doesn't require you to reflash the new clone. You can reflash the old inverter(s) with the same firmware that is in the clone, which I assume is LC1_72.70c.

Hi Weber,

Thank you so much for your warm welcome and your excellent responses. You are 100% correct that this is a clone How on earth did you know just by the firmware version? Very impressive!

I have spoken to the installer and he will replace it for me. In fact, I'm buying an original myself and he will reimburse me for it. Shame, the guy was so surprised and apologetic and didn't even try to justify. All he said was that he couldn't find stock and ended up buying from a company he's never dealt with before. That'll teach him

Hopefully I can be back on solar by the end of the day.

[weber]

Hi @Mantas. If you want to have any chance of paralleling the new machine with the old ones, you must not buy a 64 V machine. That means you must not buy a PIP-5048MS. You need an Axpert MKS 5K that does not have the 64 V option, or a rebadged equivalent such as the Mecer SOL-I-AX-5P (not the SOL-I-AX-5P64 or the SOL-I-AX-5KP).

[wolfandy]

Hi all

I am a new member from South Africa

I have an older 5kVA / 4kW model (with ICC software) at home as backup for loadshedding but am currently contemplating of adding PV to my system. The primary reason is to reduce grid consumption during the day and to recharge my existing battery. I have a Pylontech 3.5kWh installed and would run it down to 20% SoC during the night before switching to grid - but am not looking at adding more Pylontechs to completely go off-grid.

I have read in other places on the web that my inverter is not able to mix PV and grid sources. Is that correct?

If yes, how would the inverter behave in the following cases:

1. PV supply exists but is less than my current consumption; battery is not fully charged.
2. PV supply exists but is less than my current consumption; battery is fully charged
3. PV supply exists and is higher than my current consumption; battery is not fully charged; PV excess is below max charging current for battery. Will it then cover current consumption and charge with whatever PV excess exists (so simply slower charging)?

From what I have read, the Axpert MKS II is able to mix power sources. Is that correct?

Thanks in advance for your help

[coulomb]

I am a new member from South Africa

Welcome!

I have an older 5kVA / 4kW model ...

Ah but what model? See next question.

I have read in other places on the web that my inverter is not able to mix PV and grid sources. Is that correct?

It depends if it's a 450 V max SCC model or not (i.e. it's an early Axpert MKS II with PF 0.8; I don't know if they ever existed). You seem to talk about an Axpert MKS II as being different to your machine, so in that case, yes, your machine can't mix AC-in and solar, but it can mix solar and battery power.

If yes, how would the inverter behave in the following cases:
1. PV supply exists but is less than my current consumption; battery is not fully charged.
2. PV supply exists but is less than my current consumption; battery is fully charged

Both these cases are really the same. The battery will run down, supplying the difference between the load power and solar power (ignoring losses for simplicity). Once the battery runs down far enough, the inverter will switch to "line mode", where the AC-in supplies all the load, and PV power goes only to charging the battery. If enabled, the AC-in may charge the battery at the same time. When the battery has recovered to a certain voltage, the inverter goes back to the original case where the battery supplies the difference between the load power and solar power.

3. PV supply exists and is higher than my current consumption; battery is not fully charged; PV excess is below max charging current for battery. Will it then cover current consumption and charge with whatever PV excess exists (so simply slower charging)?

Yes. And in the case that the available PV power excess exceeds the maximum charge rate, there will be unused solar power.

From what I have read, the Axpert MKS II is able to mix power sources. Is that correct?

I don't know the Axpert MKS IIs well. From reading the manual, it looks like any time it's in line mode, it is blending AC-in and PV power. But provided that you have enough battery storage and PV power, you spend little time in line mode, so it seems to me that the difference is not all that great.

[wolfandy]

Thanks a lot for the prompt reply, Coulomb

Ah but what model? See next question.

It is listed as 145V if I am not mistaken. So I don't think that I have an MKS II. Here the sticker to confirm

It depends if it's a 450 V max SCC model or not (i.e. it's an early Axpert MKS II with PF 0.8; I don't know if they ever existed). You seem to talk about an Axpert MKS II as being different to your machine, so in that case, yes, your machine can't mix AC-in and solar, but it can mix solar and battery power.

Thanks. That's what I suspected

Both these cases are really the same. The battery will run down, supplying the difference between the load power and solar power (ignoring losses for simplicity). Once the battery runs down far enough, the inverter will switch to "line mode", where the AC-in supplies all the load, and PV power goes only to charging the battery. If enabled, the AC-in may charge the battery at the same time. When the battery has recovered to a certain voltage, the inverter goes back to the original case where the battery supplies the difference between the load power and solar power.

So that means if I have a spike in my consumption that exceeds my available PV (e.g., coffee machine or vacuum cleaner), it would then not only stop charging the battery but if required even draw on the battery to provide the required power. How streneous is this on the battery? And does this not mean that the battery is going through multiple cycles a day, which will have an impact on expected battery life? Or is the impact only theoretical and the battery will still last 10+ years?

Yes. And in the case that the available PV power excess exceeds the maximum charge rate, there will be unused solar power.

Thanks

I don't know the Axpert MKS IIs well. From reading the manual, it looks like any time it's in line mode, it is blending AC-in and PV power. But provided that you have enough battery storage and PV power, you spend little time in line mode, so it seems to me that the difference is not all that great.

I would appreciate your view on my situation and thoughts. Here a few facts about my situation:

• As mentioned, I have 3.5kWh of battery storage

• I live in Cape Town and generally have good weather and plenty of sun here

• My baseload is around 500-600W during the day and 300W during the night when everyone is asleep

• When the TV etc are running in the evening, the load goes up to 900W

• My heavy consumers (ACs, pool pump, heat pump, washing machine, tumble dryer, dishwasher, oven) are all directly connected to the mains

My idea so far was the following - and maybe I am completely off:

• I was thinking somewhere between 2.0kW and 2.5kW of PV

• In the evening I would be running the battery down to 20% SoC (empty) and then switch over to grid (I don't think the battery will last me through the night and am ok with it)

• In the morning, any PV would supplement my consumption (battery still empty)

• Once PV exceeds my consumption, it would charge the battery

• Once battery is fully charged and PV is still higher than consumption, then obviously wasted solar power (if I see that this is for extended period every day, then maybe switch pool pump on timer over to PV to utilise the excess power)

• As PV ramps down, battery provides difference and we are at the beginning again

Would that work and does it make sense? Or would you propose something different?

[coulomb]

It is listed as 145V if I am not mistaken.

Yes, that's an ordinary Axpert MKS PF0.8. Just like my PIP-4048MS.

So that means if I have a spike in my consumption that exceeds my available PV (e.g., coffee machine or vacuum cleaner), it would then not only stop charging the battery but if required even draw on the battery to provide the required power. How streneous is this on the battery?

It's a microcycle, but that's what the battery is for (to me at least). It won't affect the life drastically.

I would appreciate your view on my situation and thoughts. Here a few facts about my situation:
...
Would that work and does it make sense? Or would you propose something different?

That's a perfectly ordinary situation. And you can add more panels and more battery to move more loads to your inverter. If the inverter overloads, it switches to AC anyway.