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

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.
User avatar
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

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

Post by rthorntn »

My house lost power again, I think it might only happen when the inverters have been continuously powering the house with battery for over 24h, that is why it happens a lot more in summer. I have doubled my base load and the weather has been pretty meh so I'm rarely getting through the night on battery.

@coulomb does that seem feasible, if so, does it shed any more light on what could be happening?

Thanks for all the help that I've had in here!
7.4KW solar (2 strings of 10)
Two paralleled MPP 5048MG inverters (71.80)
19.2KWH battery (10 x Zenaji Aeon LTO)
Growatt solar inverter
Goodwe SBP G2 battery inverter
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

rthorntn wrote: Tue, 21 Dec 2021, 14:32 I have doubled my base load and the weather has been pretty meh so I'm rarely getting through the night on battery.

@coulomb does that seem feasible, if so, does it shed any more light on what could be happening?
By "what could be happening", I assume you are referring to your batteries tripping off-line.

My guess is that your batteries aren't sharing the load evenly, so that one battery gets more of the load than the others, at least short term. It may be that they equalise over a period of minutes to hours. But if the share of the current is enough to trip them, then it's obviously too late for equalisation. If that's what is happening, then yes the extended base load and poor weather could lead to more frequent tripping.

How are the battery modules connected to each other? Perhaps a photo of the wiring would be useful. It can be difficult getting the modules to share properly.
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
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

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

Post by rthorntn »

Thanks @coulomb.

Sorry I should have been more specific, the main issue is the inverters (two in the load sharing config) silently "crap" themselves and my power goes out for a couple of seconds, then it comes back on again.
7.4KW solar (2 strings of 10)
Two paralleled MPP 5048MG inverters (71.80)
19.2KWH battery (10 x Zenaji Aeon LTO)
Growatt solar inverter
Goodwe SBP G2 battery inverter
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

rthorntn wrote: Wed, 22 Dec 2021, 06:42 the main issue is the inverters (two in the load sharing config) silently "crap" themselves and my power goes out for a couple of seconds, then it comes back on again.
Ah. It's possible that the battery modules are disconnecting briefly due to over-voltage†. Perhaps you could watch the battery voltage when this happens, or better yet post a graph of battery voltage from your monitoring software (if any). Perhaps a brief over-voltage just causes a brief disconnect, and a more sustained and/or higher voltage surge causes them to trip off.

What is your absorb/CV battery voltage setting?

How does this compare with the maximum battery module voltage?

If these two are too close, overshoots from the poor Axpert control algorithms may be the root cause. Reducing the absorb/CV voltage setting would be the fix if so.

† Edit: I've recently become aware that some LFP BMS are capable of "disconnecting" the battery for charging, whilst somehow enabling it for discharge. If this applies to Pylontechs and other inverter-scale battery modules with built-in BMS, and if this acts quickly enough, then what I described won't happen, as the BMS should reconnect the battery as soon as the current starts wanting to flow out of the battery.
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
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

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

Post by rthorntn »

Thanks @coulomb

I think I may have had problems with overshoot in the past so I tweaked the settings and it improved living with the system :lol:

But I'm not so sure this is the batteries, attached is the emonCMS voltage graph from the incident as well as the "missing data", just after 7.30am the logging disappears and it pops back to tell me the battery voltage is 0 over an hour later and at 9.30am the voltage jumps back to normal.

Its like the management CPU starts crapping itself and after an hour I lose power for a few seconds.

ICC and my internet router are on a UPS so I should have data.

Any hints anywhere that an inverter firmware update could fix something like this, I mean I can't be the only one in the world with this problem? I'm on stock 71.80 on the two MPP 5048MG inverters. I'm just reading https://powerforum.co.za/topic/6620-axp ... ugust-2020 for example...?
Attachments
missing data.png
missing data.png (124.55 KiB) Viewed 8802 times
volts.png
volts.png (167.11 KiB) Viewed 8802 times
7.4KW solar (2 strings of 10)
Two paralleled MPP 5048MG inverters (71.80)
19.2KWH battery (10 x Zenaji Aeon LTO)
Growatt solar inverter
Goodwe SBP G2 battery inverter
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

rthorntn wrote: Wed, 22 Dec 2021, 10:27I'm on stock 71.80 on the two MPP 5048MG inverters.
There is also 71.86 here.

Sorry, I don't know much about the higher PV voltage models, so I have no idea whether a firmware update such as the above will fix your problem.
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
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

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

Post by rthorntn »

Thanks @coulomb and have a great holiday!
7.4KW solar (2 strings of 10)
Two paralleled MPP 5048MG inverters (71.80)
19.2KWH battery (10 x Zenaji Aeon LTO)
Growatt solar inverter
Goodwe SBP G2 battery inverter
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

Hello guys,
i have problem with with my PIP8048MAX, MPPT does not take full advantage of the maximum possible power from PV panels when it is cloudy. Here is video: https://youtu.be/xYGh9namYVg .
Has anyone encountered this problem and is there a solution? Thank You

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

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

Post by coulomb »

@Berosh, I moved your post here, and renamed this topic to cater for the MAX model. I think Weber and I hoped to avoid discussion of all but the 5 kVA models, but such discussion seems inevitable for the 7.2 and 8.0 kW models.
Berosh wrote: Thu, 06 Jan 2022, 08:35 i have problem with with my PIP8048MAX, MPPT does not take full advantage of the maximum possible power from PV panels when it is cloudy. Here is video:
Thanks for the video, it answers a lot of questions.

At first, I thought it had to be the dreaded premature float bug, but from the video I can see that the charge LED is still flashing, indicating that it's still in the bulk/absorb stage.

You have very recent removable display firmware, and besides this seems to be a DSP problem.

I don't have any PIP-8048MAX firmware, but I see that the PIP-7248MAX firmware is up to at least version 45.07. My guess is that the 45.xx and 46.xx (where the xx are the exact same) are recompilations of the same firmware with different constants for 7200 W max and 8000 W max respectively; if so there should now be main firmware version 46.07 or later for the 8 kW models.

Since there aren't any PIP-8048 firmwares available for download, at least that I'm aware of, you'll have to ask your supplier for the update file. It's not certain that this will fix it, but they do seem to have fiddled the solar charging code a lot recently. At least, that's my impression without having had time to study it in any detail. Send a photo of your sticker, and a pointer to the above video, in your email to your supplier.
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

coulomb wrote: Thu, 06 Jan 2022, 11:24 @Berosh, I moved your post here, and renamed this topic to cater for the MAX model. I think Weber and I hoped to avoid discussion of all but the 5 kVA models, but such discussion seems inevitable for the 7.2 and 8.0 kW models.
Berosh wrote: Thu, 06 Jan 2022, 08:35 i have problem with with my PIP8048MAX, MPPT does not take full advantage of the maximum possible power from PV panels when it is cloudy. Here is video:
Thanks for the video, it answers a lot of questions.

At first, I thought it had to be the dreaded premature float bug, but from the video I can see that the charge LED is still flashing, indicating that it's still in the bulk/absorb stage.

You have very recent removable display firmware, and besides this seems to be a DSP problem.

I don't have any PIP-8048MAX firmware, but I see that the PIP-7248MAX firmware is up to at least version 45.07. My guess is that the 45.xx and 46.xx (where the xx are the exact same) are recompilations of the same firmware with different constants for 7200 W max and 8000 W max respectively; if so there should now be main firmware version 46.07 or later for the 8 kW models.

Since there aren't any PIP-8048 firmwares available for download, at least that I'm aware of, you'll have to ask your supplier for the update file. It's not certain that this will fix it, but they do seem to have fiddled the solar charging code a lot recently. At least, that's my impression without having had time to study it in any detail. Send a photo of your sticker, and a pointer to the above video, in your email to your supplier.
Hello Mr. Coulomb,
thank You for your reply. It it not float bug, because my setup is float voltage=absorbtion voltage and home apliances always consume min. 300W. I have my own hardware (ESP32 board) connected via RS232 to PIP8048MAX and xiaoxiang BMS connected via bluetooth and control battery charging with my custom ESP32 firmware via commands to PIP8048MAX. Now i am working on xiaoxiang bluetooth to pylontech RS485 protocol convetor... I sended my problem to mpp solar and to my supplier but no response yet :-( I apologize for my imperfect English. With regards Berosh
kjm
Noobie
Posts: 8
Joined: Mon, 02 Aug 2021, 23:57

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

Post by kjm »

Hi

this is only a firmware problem that not optimised because the mppt loop was not able to return to an optimal point and that is why if the current exceeds 0.6A and beelow 1A and the voltage is beelow 120 volts on the two mppt inputs I disconnect and reconnects 10 seconds later and often the mppt voltage remains between 180 and 210 volts for my installation (2 x 6 for mppt1 and 2x 6 for mppt2) and the control is done automatically every 30 seconds
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

kjm wrote: Fri, 07 Jan 2022, 01:29... that is why if the current exceeds 0.6A and below 1A and the voltage is beelow 120 volts on the two mppt inputs I disconnect and reconnects 10 seconds later and often the mppt voltage remains between 180 and 210 volts for my installation...
Ah! I didn't catch the fact that the PV1 voltage in the video was 99 V. So it looks like the MPPT has gone over the maximum power point hump, or been confused about the double hump with the two PV strings (if present).

This is PV101 stuff! Finding the maximum power point is what MPPTs do. Getting stuck at low panel voltage is just crazy.

Hopefully it will give me a clue to find out which firmwares (of the ones I have access to) have this problem or don't.
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-5048GE/GK/MG/MGX/MAX inverters (with unsafe SCC)

Post by coulomb »

coulomb wrote: Fri, 07 Jan 2022, 10:55 Ah! I didn't catch the fact that the PV1 voltage in the video was 99 V. ... Hopefully it will give me a clue to find out which firmwares (of the ones I have access to) have this problem or don't.
It turns out that I have Axpert MAX / PIP-MAX firmware versions 45.04, 45.06, 45.07, and some outliers like 45.84. All of these are 7200 W firmwares, but my guess is that the 8000 W firmwares will be very similar. At first glance, the MPPT tracking code is identical in three that I looked at (all the aforementioned except 45.06), and a little disappointing. Power is tracked only once per second, for example, and there doesn't seem to be any full scanning, just logic that reverses direction as needed. This is not proper maximum power point tracking, and can get stuck in local maxima. It looks like 99 V was a local maximum for the above system. Interestingly, the lower MPPT voltage limits before PV loss is declared is indeed 90-450 V, which agrees with the manuals (for once). So it was just 10% over the voltage where it would have given up anyway, presumably restarted, and presumably fixed itself.

I wish I could say that the 145 V max models use a completely different algorithm. However, without comparing the code in detail, it is reminiscent of what I saw with the separate SCC firmware. (145 V max models use a small MCU on a the Solar Charge Controller board to manage PV power; the 450 V and 500 V max models control Solar Charge Power with the same DSP that controls the rest of the inverter hardware (the DC-AC converter, the boost transistor, relays, bus voltage management, charge management, etc).

So I'm not hopeful of a significant improvement in the PV handling in the near future. I'd he happy to be proved wrong, of course.
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
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

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

Post by rthorntn »

@coulomb could this have something to do with the outages I'm seeing?

Thanks!
7.4KW solar (2 strings of 10)
Two paralleled MPP 5048MG inverters (71.80)
19.2KWH battery (10 x Zenaji Aeon LTO)
Growatt solar inverter
Goodwe SBP G2 battery inverter
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

rthorntn wrote: Fri, 07 Jan 2022, 14:13 @coulomb could this have something to do with the outages I'm seeing?
I guess it could be the reason behind a lot of problems, yes. Especially if you have a system with paralleled strings; these will presumably have multiple maxima which I assume will confuse simple hill climbing firmware the most.

There's still a chance that I'm misreading the code; maybe there's a scan function (where the whole range of MPPT voltages is occasionally scanned to find the true maximum power point) that I've not come across. But then, if there is, how does it get stuck at 99 V?
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

Thank you all for your answers. I hope a fix will come soon. Getting the maximum out of the PV panel is important for me in island mode. The MPPT tracker can be unlocked by manually disconnecting the PV panels or by switching the USB-> SBU mode.
kjm
Noobie
Posts: 8
Joined: Mon, 02 Aug 2021, 23:57

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

Post by kjm »

Hi coulomb

I have make the test with only one string and the problem is same , with the 7200W or the 8000W same problem , if you have 90V in the mppt input with 0,5A after disconnect the pannel the mppt voltage begin with 200V and go slowely from 0 to 0,5A and after 10 seconds it get down to 90V and the courant stay to 0,5A , if the courant is over 0,8A the voltage go betwen 180 and 200V. With victron smartsolar 250/60 charger it find very fast the MPPT point with 4 x 6 panel (2 x 6 in one direction and 2 x 6 in another direction) in input and the voltage is never below 160V but not the same price
Capture1.JPG
Capture1.JPG (31.33 KiB) Viewed 8568 times
in the picture you can see that was 5 atempt to disconnect and reconnect the solar pannel and at the 6 atempt the mppt is stay to 180V
this day was a very cloudy day
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

kjm wrote: Sat, 08 Jan 2022, 06:00 I have make the test with only one string and the problem is same , with the 7200W or the 8000W same problem,
Ok, that shoots down the multiple peak power point theory.
if you have 90V in the mppt input with 0,5A after disconnect the pannel the mppt voltage begin with 200V and go slowely from 0 to 0,5A and after 10 seconds it get down to 90V and the courant stay to 0,5A , if the courant is over 0,8A the voltage go betwen 180 and 200V.
Ah, that puts new light on it, thanks. Now I'm wondering if they've neglected the effect of self consumption, which on these models I'm guessing would be about 70 W. (Anyone have a real figure?) I note that 90 x 0.8 = 72 W, so with up to that power level, when they think they are sending power to the battery, it's actually running out of the battery. But now that I type this, that doesn't seem like it should affect the peak power tracking.

Oh wait, unless their measurement of PV power is a calculation, and they didn't take self consumption into account for that calculation. I'll go have a read soon.
With victron smartsolar 250/60 charger it find very fast the MPPT point with 4 x 6 panel (2 x 6 in one direction and 2 x 6 in another direction) in input and the voltage is never below 160V but not the same price
Yes, Victrons have a proper scanning MPPT algorithm, which takes real engineering to get just right. All that development costs money. There does seem to be significant engineering put into the Voltronic Power designs, but once it's in production, there doesn't seem to be much effort at fixing any design flaws. Almost like maintenance is relegated to the junior engineers, who don't have the experience to fix anything fundamental.
in the picture you can see that was 5 attempts to disconnect and reconnect the solar panel and at the 6 attempt the mppt is stay to 180V
this day was a very cloudy day
Again, this puts new light on the problem. I had the impression that the problem was a fairly rare and random one, when it would sometimes get stuck and you'd have to reset the solar charging to unstuck it, whereupon it would work correctly for probably the rest of the day. The above indicates a systemic problem, one which only seemed to be corrected when the available PV power increased above a certain level, presumably with more sunlight and/or less cloud.
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

Yes, Victrons have a proper scanning MPPT algorithm, which takes real engineering to get just right. All that development costs money. There does seem to be significant engineering put into the Voltronic Power designs, but once it's in production, there doesn't seem to be much effort at fixing any design flaws. Almost like maintenance is relegated to the junior engineers, who don't have the experience to fix anything fundamental.
At the same time, correcting this error should be relatively simple :-(
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

coulomb wrote: Sat, 08 Jan 2022, 09:18 Oh wait, unless their measurement of PV power is a calculation, and they didn't take self consumption into account for that calculation. I'll go have a read soon.
No, it seems that they have an actual sample for PV voltage and PV current, and average the product of these.
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

So far, I've solved it with a ugly fix. When the PV power is between 60-150W and PV voltage is less 110V, then my esp32 via RS232 performs SBU-> USB-> SBU mode switching. It works, but it's far from ideal.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

I think I found the suspect code. There is a test for the present PV current strictly less than 0.50A (measured in hundredths of an amp, so 0.49A or below), and the PWM value less than about 50%. If so, it targets the present PV voltage less 5.0V, regardless of what the MPPT tracking function returns. This seems to be a misguided attempt to "get things moving" when PV available power is low, but it seems to me to have the effect of boosting too much, locking the panels into a low voltage state well away from the maximum power point. It would come good on its own only when in this unusual operating condition there was enough sun to get the current over 0.49 A.

Once the target voltage falls below 90 V, it will be pegged at 90 V (the lower limit of MPPT tracking). This doesn't seem to trigger any resetting that would get it out of this low voltage frozen condition.

I'll have a think about how this could be safely worked around. I'll need a beta tester, since I don't have a MAX or other model that might have this fault.

Edit: there is another function I need to understand first; it sets the boost (MPPT) PWM value.
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

coulomb wrote: Mon, 10 Jan 2022, 13:55 I think I found the suspect code. There is a test for the present PV current strictly less than 0.50A (measured in hundredths of an amp, so 0.49A or below), and the PWM value less than about 50%. If so, it targets the present PV voltage less 5.0V, regardless of what the MPPT tracking function returns. This seems to be a misguided attempt to "get things moving" when PV available power is low, but it seems to me to have the effect of boosting too much, locking the panels into a low voltage state well away from the maximum power point. It would come good on its own only when in this unusual operating condition there was enough sun to get the current over 0.49 A.

Once the target voltage falls below 90 V, it will be pegged at 90 V (the lower limit of MPPT tracking). This doesn't seem to trigger any resetting that would get it out of this low voltage frozen condition.

I'll have a think about how this could be safely worked around. I'll need a beta tester, since I don't have a MAX or other model that might have this fault.

Edit: there is another function I need to understand first; it sets the boost (MPPT) PWM value.
Interesting, do you have source codes for inverters? I'm a little afraid of being a beta tester :-) My inverter is still under warranty, but I may risk it.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

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

Post by coulomb »

Berosh wrote: Mon, 10 Jan 2022, 17:30 Interesting, do you have source codes for inverters?
No, I'm an old-style reverse engineer that can work with DSP assembly language.
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.
Berosh
Noobie
Posts: 9
Joined: Thu, 06 Jan 2022, 08:16
Real Name: Jiri

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

Post by Berosh »

Good work! What type of DSP is in PIP inverters? Will the code be compatible for MAX, that has two MPPT PV inputs?
Post Reply