PIP-8048MAX Adventure

non-EV Solar, Wind and other renewable power sources
non-EV batteries and other energy storage stuff
Forum rules
Important!
This forum is for discussion of Non-EV matters.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

PIP-8048MAX Adventure

Post by alexfsr »

Hello everybody,
I recently came accross this lovely forum while researching the endless pile of problems resulting from my purchase of brand new PIP-8048MAX inverters. I got 2 of them directly from MPP Solar's official AliBaba seller and they arrived in Eastern EU in about 2.5 weeks. So far so good! The 2 inverters are to work in parellel with a total of ~22kW of Suntech 445W panels to power my small crypto farm in the countryside.

Getting the inverters to work has been a giant PITA. The infamous premature float bug is indeed very much present and infuriating (i didn't know it's a "feature" of PIP inverters until i found this forum, thought i was doing something wrong). Fortunately, I'm using LFP batteries with a Chinese AliExpress BMS boards with active balancing so i think i have managed to circumvent this bug by setting up the float voltage to equal the charge voltage without damaging the batteries. Still testing that so gonna report back soon. However, the inverters suffer from very weird and unpredicatble behaviors that have no evident explanation:

The inverters would for no reason seemingly at random stop drawing current from the panels (but still reading voltage in the expected range, so connection is good and panels are working) and will not recover until the inverters are fully power cycled (everything supplying power to them is disconnected) or until on the next day, when the panels have enough power to cover the output loads (~10kW). Until that point, they don't draw a single mA from the the panels and at night the AC chargers doesn't work either. After some testing, it seems to be either a firmware bug or an undocumented protection feature of setting 41 (Max battery discharge current) that kills the SCC until it's power cycled. I have now disabled setting 41 entirely and only relying on BMS to disconnect batteries in case of current overdraw. Will report back in a day or 2 with results.

The inverters get very loud and very hot on the PV input side at high PV input power. I read in other threads here that MPP Solar are mounting the fans on these units backwards. Is this true? Are they actually this stupid? If i open them up to swap them around, do i lose any sort of warranty i may have?

Since the utility grid in the area where the inverters work is a bit of a meme from commie times, the voltage drops considerably at certain times of day under the load of my mining rigs when in line mode, down to ~170V at times (from 230V nominal). As such i have to operate the inverters in Appliance input mode, otherwise they die at certain times of day. The long switching times are obviously undesirable however, so i installed a 9kW On-Line UPS between the inverters and the rigs to hold the transitions, but this has brough a new issue. When working with the supposedly 230V "pure sine" wave of the inverters, the cables between the UPS and the electrical panel (massively oversized for the power draw, high-quality copper) get hot and the circuit breakers very quickly heat up and shut off from overheating. I tested and measured this, the Amps going thorugh the circuit breaker in the electrical panel and the built-in one on the UPS (63A) are ~50A, but the breakers toggle off from over-heating. There is no apparent reason for this to be happening. Any ideas?

In the short amount of time so far where i have gotten them to work properly, they work pretty sweet. Sadly, this has only been the case for only ~2 days of the ~10 they have been in operation. I'll put some pictures when the forum allows (new acc). Here is some more info of the inverters for those that are in the know (seems like some admins here are quite knowledgeable).

Main FW: 00046.05
Secondary FW: 00012.06
SBU Priority
CSO Priority
57V for settings 26 and 27.
54V for back to battery setting.
16S CALB 200Ah LiFePO4 battery pack with active balance BMS (2A active balance). Have one more such pack but its BMS went kaput because i accidentally bent a pin in the balancing leads connector and it shorted with the one next to it. Now when a cell is connected to that pin, the BMS refuses to boot (or dies if i connect the wire after booting it). Waiting for a new one to arrive in a few weeks.
paulvk
Senior Member
Posts: 457
Joined: Fri, 23 Oct 2015, 23:45
Real Name: Paul
Location: Sydney

Re: PIP-8048MAX Adventure

Post by paulvk »

What is the max solar panel voltage, if it gets too high what you are experiencing happens
the 500 volts is the maximum voltage the MPPT charger will work at a some what lower voltage.
For example the PIP4048 has a max solar voltage of 145 volts but only works if its below 120 volts
it does exactly the same thing as your units are doing if it goes above that 120 volts.
Now having a UPS on the output of these inverters can have problems because they
are transformer less in designe and may have problems stabilizing the output voltage.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

paulvk wrote: Wed, 28 Jul 2021, 11:49 What is the max solar panel voltage, if it gets too high what you are experiencing happens
the 500 volts is the maximum voltage the MPPT charger will work at a some what lower voltage.
The MPPT is considered "lost" on these models if outside the range of 90-450 VDC.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Wed, 28 Jul 2021, 12:16
paulvk wrote: Wed, 28 Jul 2021, 11:49 What is the max solar panel voltage, if it gets too high what you are experiencing happens
the 500 volts is the maximum voltage the MPPT charger will work at a some what lower voltage.
The MPPT is considered "lost" on these models if outside the range of 90-450 VDC.
paulvk wrote: Wed, 28 Jul 2021, 11:49 What is the max solar panel voltage, if it gets too high what you are experiencing happens
the 500 volts is the maximum voltage the MPPT charger will work at a some what lower voltage.
For example the PIP4048 has a max solar voltage of 145 volts but only works if its below 120 volts
it does exactly the same thing as your units are doing if it goes above that 120 volts.
Now having a UPS on the output of these inverters can have problems because they
are transformer less in designe and may have problems stabilizing the output voltage.
The panels are configured in strings of 7 and 6 panels, with 2 strings in parallel for each PV input of each inverter (52 panels total). One input operates around 250V, the other around 300V.
Safe to say both are well within the operational range of the MPPT. Panels have Voc of 49V and Vmp of ~40-41V.

If i'm able to obtain a firmware update file from MPP Solar, would u be able to patch the ridicolous premature float bug?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Wed, 28 Jul 2021, 18:40 If I'm able to obtain a firmware update file from MPP Solar, would u be able to patch the ridiculous premature float bug?
99% sure I could. That part of the firmware is same old same old. I could not test the patch, but it's two very small patches that have been made to many firmwares. Plus a little extra just to display a different firmware version, so you know that the firmware updated and the patches are in place.

Edit: this would be a very light patch, just those three changes.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Wed, 28 Jul 2021, 21:33
alexfsr wrote: Wed, 28 Jul 2021, 18:40 If I'm able to obtain a firmware update file from MPP Solar, would u be able to patch the ridiculous premature float bug?
99% sure I could. That part of the firmware is same old same old. I could not test the patch, but it's two very small patches that have been made to many firmwares. Plus a little extra just to display a different firmware version, so you know that the firmware updated and the patches are in place.

Edit: this would be a very light patch, just those three changes.
That's great! I would very much appreciate it if you would and i'll test it. I've started an email thread with MPP Solar support. Let's see if i can work them to give it to me (from what i've read around here today, it doesn't seem like they are particularly inclined to give these firmwares out).
User avatar
lynx
Noobie
Posts: 3
Joined: Sun, 21 Nov 2021, 17:32

Re: PIP-8048MAX Adventure

Post by lynx »

I'm in the same boat just now. How did you guys get on with patching the firmware?
w0jtas
Noobie
Posts: 2
Joined: Wed, 01 Dec 2021, 17:29

Re: PIP-8048MAX Adventure

Post by w0jtas »

Just wanted to click Buy on Alibaba official store, but now i'm not sure if it is not a better option to give twice more money for Victron RIG. Is there any chance of updating this firmware ? Any news ?

Best regards,
w.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

lynx wrote: Sun, 21 Nov 2021, 17:33 How did you guys get on with patching the firmware?
w0jtas wrote:Is there any chance of updating this firmware ?
I still have not seen any 8 kW MAX firmware to patch.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
w0jtas
Noobie
Posts: 2
Joined: Wed, 01 Dec 2021, 17:29

Re: PIP-8048MAX Adventure

Post by w0jtas »

Contacted with MPP Solar at Alibaba. Look at the pic attached... I don't know what to think. Don't they read the internet what ppl are saying ?

Image http://firenet.eu.org/w0jtas/mpp.png

w.
Revlac
Groupie
Posts: 87
Joined: Sat, 19 Aug 2017, 19:43
Real Name: Aaron
Location: Scenic Rim

Re: PIP-8048MAX Adventure

Post by Revlac »

Perhaps they expect us to adjust the settings until we get it working to some extent.

I don't have this model, had an old pip that would go to float within 2 minuets of turning the unit on to charge the batteries, it completely ignored the bulk charge time setting.....yes its most annoying and not good for the batteries, at the time I had a small FLA setup, the fix was to adjust the charge voltage and lower the charge current, so it stayed charging above the 10% thresh hold, I never used the full current the charger could do, didn't have enough panels.

Since moving to a lithium battery never had the problem again, used custom settings, also maximum charge voltage had to be set a little lower than the BMS to avoid the solar charger cutting out.
4048HS Inverter 2015.
5048HS Inverter with 200AH GFB cells
5048MS Inverter with 200AH GFB cells
Mad Inverter
87 Dream
Noobie
Posts: 3
Joined: Sun, 22 Sep 2019, 16:01
Real Name: Steve Rebelo
Location: Johannesburg & Doha

Re: PIP-8048MAX Adventure

Post by 87 Dream »

Hi guys, Unfortunately the only work around this problem is to set a bulk and float as high and as safely your lithium battery can go. If it's a lead acid, high floats can hurt the battery. However, Lithium batteries of 16s and 55V tends to work really well.

Good luck with the fix. There are some unofficial stuff floating around but the user settings above tend to help.

87
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

Hello again everybody. Sorry for not posting here in a while, it has been a wild year.

First off, I did manage to get a firmware for the 8048 from MPP solar support. I'll see if i can post it here for the people who know how to fix MPP's cr*p.

These inverters were littered with issues it's amazing they are allowed to sell them. With that out of the way, the way i resolved my issues was through the patched firmware and using the input/output priority timers. When i first setup the timers, i set them up so that the inverters switches into USB mode in the evening (when the night-time electricity pricing starts) and switch back to SBU around 10AM the following the day when i expect the sun to be up on a sunny day. This way the inverter switches into battery mode immedietly upon hitting the target time since the solar is already stong enough to handle the load. So this was my solution to the getting stuck in line mode problem. However, shortly after, a weird issue would start occuring where early in the morning (~5AM), the inverters would occasionally enter a "runaway AC charge" where they would start pumping the max amount of power they can into the battery from the grid (~100A/60V), regardless of the limit settings (programs 02/12/13). In my setup this immeditly pulls several circuit breakers and kills the entire electrical system on my property. Obviosuly a very bad problem. The firmware MPP gave me addressed this issue succesfully but the other problems, mainly the line mode stick, were not. They attributed the issue AC charging issue to be realted to poor quality AC input, which i guess is fair (grid really is a joke in my area).

The issue with the battery is unresolved, as expected. I got around that by setting up the float voltage to equal charging voltage as mentioned by others above and then using the BMS on the LiFe batteries for protection if necessary. I cycle my batteries every day so it's not much of an issue.

The issue with the UPS i resolved by removing the UPS haha. The inverters have an absolutely insane switching times when they work in parallel. Spec sheet says max 50ms in parallel, but it's more like ~5000 (literally 5-10 seconds, i have it on video from my CCTV). Fortunately for me, my loads are seperatable so i was able to reorganize my electrical panel and seperate the inverters into 2 seperate units working on seperate inner loops with seperate batteries (1 pack for each). When working in single mode the switching time is <50ms as advertised. [To be fair again, the grid in my place is really sh*t so that may be the cause for the long switching time as it mainly occured when switching back to grid (so phase locking sounds like the probable culprit)]

Overall, it was a rather painful experience to be honest, but these PIP-MAX units are really hard to beat at these prices. I have them working fairly stable now and the electricity bill savings i get are quite substantial, especially with the exponential increase in electric prices in the EU in 2021. I'm currently negotiating with their AliBaba seller for a few of their new industrial 30KW hybrid model (for a new client project) as well as one of these new 11KW PIP-MAX units to expand my own setup. Hopefully, they've improved the software at least a little bit.
Attachments
Reflash(90.10).7z
(1.01 MiB) Downloaded 133 times
Firmware Upgrade SOP via RS232.docx
(2.16 MiB) Downloaded 126 times
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Wed, 16 Feb 2022, 10:09 I'll see if i can post it here for the people who know how to fix MPP's cr*p.
I believe it's actually Voltronic Power's firmware, and MPPSolar are merely one of the largest of the resellers.

This firmware version 90.10 differs from earlier firmware 90.07 by only 7 words, all in one function. It indeed now only updates a global variable related to maximum charge current if the new value is smaller than before. That seems vaguely hackish to me, but I haven't had a chance to check the code in detail.
a weird issue would start occurring where early in the morning (~5AM), the inverters would occasionally enter a "runaway AC charge" where they would start pumping the max amount of power they can into the battery from the grid (~100A/60V), regardless of the limit settings (programs 02/12/13).
The part of the code affected is only reached when the line voltage falls to or below 220.0 V. Perhaps by 5am, load is starting to pick up in your area, the the power station workers sometimes haven't caught up with the increased demand, so the utility voltage falls and the bad code is triggered.
They attributed the issue AC charging issue to be related to poor quality AC input, which i guess is fair (grid really is a joke in my area).
I think that the grid dipping below 220 V isn't really poor quality, depending on what the nominal voltage is I guess.
The issue with the battery is unresolved, as expected.
The code appears to be identical apart from this one function, so I would not expect anything else to have been fixed.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Thu, 17 Feb 2022, 12:08
alexfsr wrote: Wed, 16 Feb 2022, 10:09 I'll see if i can post it here for the people who know how to fix MPP's cr*p.
I believe it's actually Voltronic Power's firmware, and MPPSolar are merely one of the largest of the resellers.

This firmware version 90.10 differs from earlier firmware 90.07 by only 7 words, all in one function. It indeed now only updates a global variable related to maximum charge current if the new value is smaller than before. That seems vaguely hackish to me, but I haven't had a chance to check the code in detail.
a weird issue would start occurring where early in the morning (~5AM), the inverters would occasionally enter a "runaway AC charge" where they would start pumping the max amount of power they can into the battery from the grid (~100A/60V), regardless of the limit settings (programs 02/12/13).
The part of the code affected is only reached when the line voltage falls to or below 220.0 V. Perhaps by 5am, load is starting to pick up in your area, the the power station workers sometimes haven't caught up with the increased demand, so the utility voltage falls and the bad code is triggered.
They attributed the issue AC charging issue to be related to poor quality AC input, which i guess is fair (grid really is a joke in my area).
I think that the grid dipping below 220 V isn't really poor quality, depending on what the nominal voltage is I guess.
The issue with the battery is unresolved, as expected.
The code appears to be identical apart from this one function, so I would not expect anything else to have been fixed.
Hey again.
The grid voltage here regularly drops down to ~180V, espcially during peak hours (from 230V nominal), so i guess it all makes sense now. Is it possible to see in the code how the inverter decides to switch between modes (line/battery) and potentially fix it as this is still my number 1 issue that im currently getting around with timers but these timers need to regularly be adjusted manually on site (no support for them in the watchPower software for some reason, both on PC and Android). I have friends as well as company clients that i would like to build solar systems for with this inverter but the software needs to be fixed and MPP/Voltronic don't seem too bothered about it. It took me a month of back and forth with Andy from MPP support to get this firmware and as u say, they have barely changed anything :D
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Fri, 18 Feb 2022, 01:13 The grid voltage here regularly drops down to ~180V, especially during peak hours (from 230V nominal), so i guess it all makes sense now.
Yikes. That must make switching from battery to and from line mode especially hard on the relays.
Is it possible to see in the code how the inverter decides to switch between modes (line/battery) and potentially fix it as this is still my number 1 issue ...
I would rate the mode switching code as medium complexity. I can mostly follow it in the firmwares I'm most familiar with, which are now quite old models (5 kVA PF0.8 models). I have a lot on my plate at the moment, so I likely can't get to this for some time, sorry.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
jonescg
Senior Member
Posts: 4715
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.
Contact:

Re: PIP-8048MAX Adventure

Post by jonescg »

So it sounds lke one of these 8 kW inverters isn't a good idea for our TAFE project then?
AEVA National President, WA branch director.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Fri, 18 Feb 2022, 08:42
alexfsr wrote: Fri, 18 Feb 2022, 01:13 The grid voltage here regularly drops down to ~180V, especially during peak hours (from 230V nominal), so i guess it all makes sense now.
Yikes. That must make switching from battery to and from line mode especially hard on the relays.
Is it possible to see in the code how the inverter decides to switch between modes (line/battery) and potentially fix it as this is still my number 1 issue ...
I would rate the mode switching code as medium complexity. I can mostly follow it in the firmwares I'm most familiar with, which are now quite old models (5 kVA PF0.8 models). I have a lot on my plate at the moment, so I likely can't get to this for some time, sorry.
Thank you for the info and no problem. Would you perhaps know if these inverters can share a battery without working in parallelo mode (like having them connect to the same bus-bar that has a few battery banks connected to it), but operating in single mode with seperate inner AC loops (common ground)? The mode switching time in parallel mode is multiple seconds and kills my rigs so it's not an option. In single mode they switch in a few ms, as advertised. However seperating the batteries for each inverter is a bit suboptimal for multiple reasons. Is there any reason the inverters can't share a battery without working in parallel mode?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

coulomb wrote: Thu, 17 Feb 2022, 12:08 This firmware version 90.10 differs from earlier firmware 90.07 by only 7 words, all in one function. It indeed now only updates a global variable related to maximum charge current if the new value is smaller than before. That seems vaguely hackish to me, but I haven't had a chance to check the code in detail.
I've had more of a look at it now. This code is intended to derate the maximum battery charging current for when the case when the line voltage is less than 220 VAC. The hardware limit is 6000 W of battery charge current (50 V rounded battery voltage times 120 A maximum battery-side charge current), which is about 27.3 A at 220 V (ignoring losses). The idea seems to be to keep the line current used for battery charging below that value. They use a figure of 30.5 A, 12% higher than 27.3 A, which presumably takes care of losses, and possibly is an attempt to be slightly kind to utilities whose voltage is really badly low. The calculation involves subtracting 745 .00W, and there is no check for the value becoming negative. Interpreting a negative value as an unsigned value would produce the reported symptoms, i.e. essentially unlimited utility charge current. However, the mains voltage would have to be under 24.4 VAC for the calculated maximum power to become negative, and I certainly hope that this never happens. Though I suppose it's possible that it could happen briefly at the start of a load shedding or severe brownout. All this is of course assuming that my maths and my understanding of the firmware is correct.

The "fix" as applied in Axpert Max 8 kW firmware version 90.10 is to never allow the result of the derating calculation to increase the current limit, which at that point is either the battery voltage times the utility current limit setting or 6000.00, whichever is lower. So despite the hackish feel of the patch, it seems to be valid. It won't keep the charge power at its lowest value even when the line voltage increases, for example, as I first feared. I note that this fix, whose dsp.hex is dated August 2021, is not present in firmware 90.06, whose dsp.hex file is dated 11th of February 2022. Make of that what you will. Yes, the firmware numbering does seem to be all over the place.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Fri, 18 Feb 2022, 10:20 Would you perhaps know if these inverters can share a battery without working in parallel mode (like having them connect to the same bus-bar that has a few battery banks connected to it), but operating in single mode with separate inner AC loops (common ground)?
I've never seen inside an Axpert MAX. I would assume that it has basically the same architecture as the 5 kVA models, but this isn't certain. On the 5 kVA models and as far as I know all the other smaller models, there is galvanic isolation between the battery circuit and the rest of the inverter (which is basically referenced to the inverter neutral output. So my guess is that yes, this should be safe.

To be a little more confident than that, with no power applied, use a multimeter on high ohms range and check for continuity between the battery positive terminal (don't use the battery negative terminal as some models have a reverse polarity protection circuit on the negative terminal) and the AC-out neutral. There should be at least megohms of resistance for sharing the battery to be safe. It won't be infinite because of the battery voltage sense resistors. The problem will happen when the inverter goes to line mode and AC-in connects to AC-out, but that won't happen with the power off, and you only want to poke around like this with the power firmly off. [ Edit: I thought for a few moments that this isn't valid, but it is; see next post. ]
The mode switching time in parallel mode is multiple seconds and kills my rigs so it's not an option.
That sounds extraordinary to me. Do you mean that you have multiple seconds with no AC output, or multiple seconds with unstable output, or the changeover is quick but happens several seconds after you believe that it should? Or something else?
In single mode they switch in a few ms, as advertised.
That would seem to indicate that the relays are OK, at least. I was concerned that you might have damaged your relay contacts by switching between 180 V utility and 230 V inverter output too many times.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

coulomb wrote: Fri, 18 Feb 2022, 11:06 To be a little more confident than that, with no power applied, use a multimeter on high ohms range and check for continuity between the battery positive terminal (don't use the battery negative terminal as some models have a reverse polarity protection circuit on the negative terminal) and the AC-out neutral. There should be at least megohms of resistance for sharing the battery to be safe. It won't be infinite because of the battery voltage sense resistors. The problem will happen when the inverter goes to line mode and AC-in connects to AC-out, but that won't happen with the power off, and you only want to poke around like this with the power firmly off.
I had a panic attack there. I had forgotten about the output relay, which of course would be open when the power is off. But that only disconnects the live output of the inverter, not the neutral. Only the AC-in disconnects both live and neutral. So the test as above is valid, as far as I know.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Fri, 18 Feb 2022, 10:49
coulomb wrote: Thu, 17 Feb 2022, 12:08 This firmware version 90.10 differs from earlier firmware 90.07 by only 7 words, all in one function. It indeed now only updates a global variable related to maximum charge current if the new value is smaller than before. That seems vaguely hackish to me, but I haven't had a chance to check the code in detail.
I've had more of a look at it now. This code is intended to derate the maximum battery charging current for when the case when the line voltage is less than 220 VAC. The hardware limit is 6000 W of battery charge current (50 V rounded battery voltage times 120 A maximum battery-side charge current), which is about 27.3 A at 220 V (ignoring losses). The idea seems to be to keep the line current used for battery charging below that value. They use a figure of 30.5 A, 12% higher than 27.3 A, which presumably takes care of losses, and possibly is an attempt to be slightly kind to utilities whose voltage is really badly low. The calculation involves subtracting 745 .00W, and there is no check for the value becoming negative. Interpreting a negative value as an unsigned value would produce the reported symptoms, i.e. essentially unlimited utility charge current. However, the mains voltage would have to be under 24.4 VAC for the calculated maximum power to become negative, and I certainly hope that this never happens. Though I suppose it's possible that it could happen briefly at the start of a load shedding or severe brownout. All this is of course assuming that my maths and my understanding of the firmware is correct.

The "fix" as applied in Axpert Max 8 kW firmware version 90.10 is to never allow the result of the derating calculation to increase the current limit, which at that point is either the battery voltage times the utility current limit setting or 6000.00, whichever is lower. So despite the hackish feel of the patch, it seems to be valid. It won't keep the charge power at its lowest value even when the line voltage increases, for example, as I first feared. I note that this fix, whose dsp.hex is dated August 2021, is not present in firmware 90.06, whose dsp.hex file is dated 11th of February 2022. Make of that what you will. Yes, the firmware numbering does seem to be all over the place.
It all makes sense now. The time of morning the issue occured for is also around the time substation switching happens. It's very possible the grid AC-in went down as low as 0 even for a second or 2 and triggered the bad code. The patch does indeed work since flashing the inverters with the 90.10 firmware the issue has not occured again. I pushed MPP support very hard when this was happening around July/August 2021 and they gave me this firmware towards the end of August 2021. From the versioning issues you mention, I guess this fix was done by MPP themselves instead of whoever originally makes the firmware and hardware for these inverters (as i understand, it's Voltronic?).
coulomb wrote: Fri, 18 Feb 2022, 11:06
alexfsr wrote: Fri, 18 Feb 2022, 10:20 Would you perhaps know if these inverters can share a battery without working in parallel mode (like having them connect to the same bus-bar that has a few battery banks connected to it), but operating in single mode with separate inner AC loops (common ground)?
I've never seen inside an Axpert MAX. I would assume that it has basically the same architecture as the 5 kVA models, but this isn't certain. On the 5 kVA models and as far as I know all the other smaller models, there is galvanic isolation between the battery circuit and the rest of the inverter (which is basically referenced to the inverter neutral output. So my guess is that yes, this should be safe.

To be a little more confident than that, with no power applied, use a multimeter on high ohms range and check for continuity between the battery positive terminal (don't use the battery negative terminal as some models have a reverse polarity protection circuit on the negative terminal) and the AC-out neutral. There should be at least megohms of resistance for sharing the battery to be safe. It won't be infinite because of the battery voltage sense resistors. The problem will happen when the inverter goes to line mode and AC-in connects to AC-out, but that won't happen with the power off, and you only want to poke around like this with the power firmly off. [ Edit: I thought for a few moments that this isn't valid, but it is; see next post. ]
The mode switching time in parallel mode is multiple seconds and kills my rigs so it's not an option.
That sounds extraordinary to me. Do you mean that you have multiple seconds with no AC output, or multiple seconds with unstable output, or the changeover is quick but happens several seconds after you believe that it should? Or something else?
In single mode they switch in a few ms, as advertised.
That would seem to indicate that the relays are OK, at least. I was concerned that you might have damaged your relay contacts by switching between 180 V utility and 230 V inverter output too many times.
I will try this test as soon as my 3rd unit of these arrive before i hook it up as my system has gotten quite complex and shutting it down and disconnecting everything for testing is a bit too much work to do right now. However, just to clarify, ALL neutral lines in my system connect together due to the way the electric grid is setup in post-USSR states. Here, there is no ground coming from the grid, it's just phase and neutral. The ground is tied to the neutral or you make your own ground with a literal conductive rod stuck in the ground :D . For my setup i do use my own such handmade "real" ground (compliant with local regulations, all good) so both AC input and AC output of both inverters have their ground wires going to literally the ground. In the electrical panel, ALL neutral wires from both AC inputs as well AC outputs of all my inverters and other equipment connect together onto one small bus-bar. Would this affect your proposed test in any way (and the subsequent opertaion of the inverters if test is successful)?

As for the switching time, yes i do mean multiple seconds with no AC output. When the inverters are wired to work in parallel mode and they need to switch from battery back line mode (it also takes longer than the spec sheet says when switching from line to battery, but not nearly as bad), the AC output cuts out for a few sconds before coming back up. The inverters themsevles don't power off so they are operating and there are no faults or errors recorded. Unfortuantely, ATX computer power supplies are only capable of milliseconds of hold up time, not seconds :D . This issue completely disappears when the 2 inverters work independently. However, having to seperate my batteries to seperate inverters leads to suboptimal use of both the batteries and my soalr array hence why i'm asking if i can have them share the battery (as in a big bank of several 17S battery packs running in parallel) without the inverters actually working in parallel mode as the MPP solar manual describes.

And finally, am i correct in assuming the Axpert MAX2 8KW is identical internally to the PIP-MAX 8KW? Their spec sheets are literally completely identical, just different enclosure. If the Axperts are made by the original designer and manufacturer of the hardware internals and firmware (again, as i understand it, Voltronic), shouldn't they be better than MPP Solar's stuff? Or are they plagued with the exact same issues 1:1? I found some local solar equipment distibutors who have the Axpert 8KW in stock and i'm wondering if i should switch to them for future projects, if they are better in some way (mostly firmware and support). Any thoughts?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Sat, 19 Feb 2022, 05:14 However, just to clarify, ALL neutral lines in my system connect together due to the way the electric grid is setup in post-USSR states. ... Would this affect your proposed test in any way (and the subsequent operation of the inverters if test is successful)?
No, that should be fine.
As for the switching time, yes i do mean multiple seconds with no AC output. When the inverters are wired to work in parallel mode and they need to switch from battery back line mode (it also takes longer than the spec sheet says when switching from line to battery, but not nearly as bad), the AC output cuts out for a few seconds before coming back up. The inverters themselves don't power off so they are operating and there are no faults or errors recorded.
Wow. I can't imagine what's happening there. [ Edit: Actually, on re-reading an earlier post, I agree it could be that the grid frequency is out of spec for several seconds. ] Perhaps a fault in the paralleling boards? When I had a shorted transistor in one of mine, I noticed that the inverter was slow to respond, and it was difficult to change settings as a result. Not several seconds or delay, but several tenths of a second. I suppose your system might have been worse and that could explain it. If you're interested, see my post that ended up with a partial schematic trace of the parallel boards.
And finally, am i correct in assuming the Axpert MAX2 8KW is identical internally to the PIP-MAX 8KW?

The PIP-8048MAX is identical to the Axpert MAX 8K-48. I'm pretty sure that these will be replaced by something like a U 8048MAX model, which will be identical to the Axpert MAX II 8K-48. As far as I know, all PIPs, Axperts, etc are designed by Voltronic Power in Taiwan and manufactured by Voltronic Power in mainland China. Resellers like MPPSolar merely supply artwork to Voltronic for the logo etc. So yes, PIPs and Axperts are identical inside, and literally come from the same factory. As for whether an Axpert MAX and Axpert MAX II are the same internally, they mostly are (apart from the obvious changes to the display and LEDs). It's still not clear to me whether the firmwares are compatible or not. I'm leaning towards not.
If the Axperts are made by the original designer and manufacturer of the hardware internals and firmware (again, as i understand it, Voltronic), shouldn't they be better than MPP Solar's stuff? Or are they plagued with the exact same issues 1:1?
Same issues 1:1. I highly doubt that MPPSolar are making changes to the firmware; they would have good contacts in Voltronic Power to get these things done. That's part of the reason that it's so slow; first you have to convince the reseller (MPPSolar in this case) that you have a real problem, that you're not a clone maker, and so on. Then you have to get them to understand the problem in enough detail that they can request a fix from Voltronic. The fix is sent back to the reseller, and finally back to the user. If this doesn't fix the problem, the whole process has to be repeated. This appears to be so that Voltronic never have to deal with end users.

Edit: I suspect that for all their flaws, MPPSolar are probably one of the best (least worst?) resellers to deal with.
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-8048MAX Adventure

Post by coulomb »

alexfsr wrote: Fri, 18 Feb 2022, 01:13 Is it possible to see in the code how the inverter decides to switch between modes (line/battery) and potentially fix it as this is still my number 1 issue that im currently getting around with timers but these timers need to regularly be adjusted manually on site...
I thought I'd have a quick look at this, but realised that I don't know exactly what behaviour you want fixed. Could you elaborate, please?
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
alexfsr
Noobie
Posts: 25
Joined: Wed, 28 Jul 2021, 08:11
Real Name: Alex

Re: PIP-8048MAX Adventure

Post by alexfsr »

coulomb wrote: Sat, 19 Feb 2022, 08:00
alexfsr wrote: Sat, 19 Feb 2022, 05:14 However, just to clarify, ALL neutral lines in my system connect together due to the way the electric grid is setup in post-USSR states. ... Would this affect your proposed test in any way (and the subsequent operation of the inverters if test is successful)?
No, that should be fine.
As for the switching time, yes i do mean multiple seconds with no AC output. When the inverters are wired to work in parallel mode and they need to switch from battery back line mode (it also takes longer than the spec sheet says when switching from line to battery, but not nearly as bad), the AC output cuts out for a few seconds before coming back up. The inverters themselves don't power off so they are operating and there are no faults or errors recorded.
Wow. I can't imagine what's happening there. [ Edit: Actually, on re-reading an earlier post, I agree it could be that the grid frequency is out of spec for several seconds. ] Perhaps a fault in the paralleling boards? When I had a shorted transistor in one of mine, I noticed that the inverter was slow to respond, and it was difficult to change settings as a result. Not several seconds or delay, but several tenths of a second. I suppose your system might have been worse and that could explain it. If you're interested, see my post that ended up with a partial schematic trace of the parallel boards.
And finally, am i correct in assuming the Axpert MAX2 8KW is identical internally to the PIP-MAX 8KW?

The PIP-8048MAX is identical to the Axpert MAX 8K-48. I'm pretty sure that these will be replaced by something like a U 8048MAX model, which will be identical to the Axpert MAX II 8K-48. As far as I know, all PIPs, Axperts, etc are designed by Voltronic Power in Taiwan and manufactured by Voltronic Power in mainland China. Resellers like MPPSolar merely supply artwork to Voltronic for the logo etc. So yes, PIPs and Axperts are identical inside, and literally come from the same factory. As for whether an Axpert MAX and Axpert MAX II are the same internally, they mostly are (apart from the obvious changes to the display and LEDs). It's still not clear to me whether the firmwares are compatible or not. I'm leaning towards not.
If the Axperts are made by the original designer and manufacturer of the hardware internals and firmware (again, as i understand it, Voltronic), shouldn't they be better than MPP Solar's stuff? Or are they plagued with the exact same issues 1:1?
Same issues 1:1. I highly doubt that MPPSolar are making changes to the firmware; they would have good contacts in Voltronic Power to get these things done. That's part of the reason that it's so slow; first you have to convince the reseller (MPPSolar in this case) that you have a real problem, that you're not a clone maker, and so on. Then you have to get them to understand the problem in enough detail that they can request a fix from Voltronic. The fix is sent back to the reseller, and finally back to the user. If this doesn't fix the problem, the whole process has to be repeated. This appears to be so that Voltronic never have to deal with end users.

Edit: I suspect that for all their flaws, MPPSolar are probably one of the best (least worst?) resellers to deal with.
I may be able to do the test soon; will report results.

The schematic of the parallel board doesn't look too complex. I'll inspect the parallel boards of both inverters when i go to perform the battery test. Perhaps the parallel boards are messed up indeed, as the issue is only present in parallel mode.

So the Axpert MAX2 8KW is in fact a newer model than the PIP-MAX 8KW? If MPP just packages the design and sells it, then how come they have a new 11KW PIP-MAX but there is no corresponding Axpert model? In fact, looking at the spec sheet of the 11KW PIP-MAX, they increased the max solar charging current from 120A to 150A, but the PV inputs rating remain unchaged (2x 500Voc/18A). Where/how do they get the extra 3KW charging power in what appears to be the same inverter as the 8KW PIP-MAX?
http://www.mppsolar.com/v3/catalogs/PIP-MAX.pdf
coulomb wrote: Sat, 19 Feb 2022, 11:21
alexfsr wrote: Fri, 18 Feb 2022, 01:13 Is it possible to see in the code how the inverter decides to switch between modes (line/battery) and potentially fix it as this is still my number 1 issue that im currently getting around with timers but these timers need to regularly be adjusted manually on site...
I thought I'd have a quick look at this, but realised that I don't know exactly what behaviour you want fixed. Could you elaborate, please?
Ok the issue is as follows:
In the morning around 5-6AM when sun starts rising and the solar panels start giving voltage, the inverters throw a few PV loss warnings until the array voltage stabilizes above the 90V minimum. Then around 8-9AM on a clear skies day, the sun is high enough for the panels to produce enough power to supply the loads and charge the battery (that has been fully depleted the previous evening and was not charged overnight). However, the inverters don't charge the battery at that point for some inexplicable reason. Instead, they only draw enough power from the solar to supply the loads (about 3KW/inverter, 24/7 load) and that's it. The battery doesn't charge so the battery voltage stays below the "back-to-battery" threshold. Therefore, the inverters never switch into battery mode and stay in line mode all day (on a sunny day!!). If at that point i enter the settigns and change the threshold setting to be below the actual voltage of the battery at that moment, the inverters switch immedietly upon leaving the settings menu into battery mode, and the battery starts charging. The issue occurs both in single and parallel mode so it has something to do with the SCC programming. Obviously adjusting the voltage every morning is not an accpetable solution, especially on a permanent basis. Not to mention the loss of all the unutilized power thought the day that should be used to charge the battery, but isn't.

At the moment i get around it by using the output priority timer to force the inverters to stay in line mode until a time when i'm fairly certain on a sunny day the solar array will have enough power to supply the loads. I set the back to battery voltage to be very low (like 50V) to be sure the battery will be over that in all cases. This way when the inverters switch back to SBU output priority when the timer hits the set number, they switch directly into battery mode. This workaround has several issues however and is also not an acceptable long term solution:
1. Keeping the back to battery voltage low is very unhealthy for the battery on cloudy days as the inverters will be pretty much constantly switching back and forth as the voltage moves up and down quite rapidly with LFP batteries in low SOC under load.
2. The timers also need to be periodically adjusted throughtout the year and that can only be done on-site. The WatchPower app for PC and Android does not support input/output priority timer settings.
3. I have recently discovered the timers have an issue as well. The timer setting for when a given priority mode should end takes effect 1 hour later than the set number. E.g. if the USB priority is set to end at 6AM (to force line mode), it actually ends at 7AM. The other timer setting for when to start the mode works on time.

I reported all of this extensively to MPP Solar support (Andy) when i was also having the AC charging issue. As already established, they only patched the AC charging issue. I don't know if they didn't understand the explanation (did so in great detail, even gave them WatchPower logs with highlighted rows that point to the issue) or just didn't want to bother. Safe to say, their support is barely useful which is highly unfortunate as i'm working on several solar projects that i still need to get inverters for but all these issues are kind of a dealbreaker.
Post Reply