PIP-4048MS and PIP-5048MS inverters

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.
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

Hi,

sorry for disturb again.

First good news my system is working very well since flashing the firmware. Very much thanks to all.

Today I expand my battery system to the following. 3 * US2000 and 1 US3000. Works very well until now. The US3000 is the new one.

My question. What about the max charge current I should set in the inverters. There are two in parallel. Can I set both to 100A or each on 50A. It is not clear for me, if each inverter put the max charge curent into the batteries or is the maximum devided through number of inverters?

Thanks very much
Markus
paulvk
Senior Member
Posts: 457
Joined: Fri, 23 Oct 2015, 23:45
Real Name: Paul
Location: Sydney

Re: PIP-4048MS and PIP-5048MS inverters

Post by paulvk »

" each inverter put the max charge curent into the batteries " Yes so if both set at 50A then 100A will be the charge.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

MarkusRE wrote: Sun, 26 Jul 2020, 03:39 Can I set both to 100A or each on 50A.
Since you have 3 x US2000 and 1 x US3000, your maximum total charge current can be 3 x 25 A + 1 x 37 A (the US3000s can take more charge current than the US2000s). So that's 112 A max. So you could set one to 50  A and the other to 60 A (a total of 110 A). Whichever has the best solar should be set to 60 A.
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.
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

Hi,

It seems I killed one of my inverters just now. I got an overvoltage error. The US2000 got to 100% the US3000 had 96%.

After restarting the system one inverter made a loud „bang“ and stayed on Error 09.

Now it seems I run into trouble as I flashed the firmware and warranty should be gone. Any idea?

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

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

MarkusRE wrote: Sun, 26 Jul 2020, 03:39 Today I expand my battery system to the following. 3 * US2000 and 1 US3000. Works very well until now. The US3000 is the new one.
I wish I'd said something about this now. I was wondering how much trouble you have to go to to parallel a new module with existing, to ensure that they are at the same SOC.

Also, thinking about it now, how does it work with a US3000 in parallel with US2000s? How do they share the load so that they all run at about the same SOC? But others seem to have paralleled US3000s with US2000s without issue.

Which battery module was the master? The US3000 or one of the US2000s?

I take a guess at what happened. Perhaps somehow when one module was filling up, the master battery module somehow didn't get to know about it, until the module that filled up first panicked and the whole battery had to disconnect from the inverter to protect the cells in that module. It seems to me that this should not happen; the master should be aware of the problems with all the modules, and gradually reduce the charge current so that it doesn't have to disconnect as a last resort.

Perhaps it was an overshoot by the inverter firmware. What absorb and float battery voltage settings were you using at the time?

Fault code 09 after a loud bang is definitely not good news for the inverter, as I'm sure you've guessed.

Was this patched firmware? I doubt that this would have had anything to do with it, but it's always possible.
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.
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

I think it was charging to fast, I had 52.1V and 50.8V set up. I monitored the batteries the whole morning, only this damned 10min I had a shower. Max charge power was set to 50A per inverter, but it was not balanced enough. Thats what I think.

Now the state of charge is more or less the same. 88 - 90% on all batteries.

The US2000 are the first batteries, the US3000 is the last. In batterywatch I could see, that the third US2000 went over.

Right now the US2000 have 45Ah with 90% and the US3000 65,7Ah with 88%.

Now I will go down for 51V and 50V for safty reason.

Is there a way to go back to the original firmware somehow because of warranty?
kylebotha
Noobie
Posts: 2
Joined: Sun, 26 Jul 2020, 22:54
Real Name: Kyle

Re: PIP-4048MS and PIP-5048MS inverters

Post by kylebotha »

Hi everyone. I can`t seem to figure out how to send commands to a Axpert inverter, I've been trying to use Teraterm and Realterm to send a QPIGS command but when I send the command I get a NAKss response. This is the first time I`m trying to do any sort of serial communication commands so I believe I`m doing something wrong, and I don`t think the problem is the connection to the inverter because Watchpower works just fine. In Teraterm and Realterm I set up the communication format as the protocol manuals describe, baud rate of 2400 etc. then after I've setup the serial ports communication format and selected the coms port that the inverter is connected to, I type out the command+crc which I get from Scott's little CrcGen application so i type in "5150494753B7A9" and then press enter, then I receive the NAKss response, I did notice in the command protocol documentation the format of the commands is <command><crc><cr>, but I don`t know what the <cr> value is, and I read Coulombs post on how to directly send commands using Tera-term and I edited the teraterm.ini file as described, but when I try to enter characters with a high bit set such as 0xA7 which Coulomb describes as pressing (right-alt-7) nothing happens inside of Tera-term, even with local echoing enabled. So if anyone could help me, what would I type into Tera-term to send the QPIGS command and get a valid response ?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

kylebotha wrote: Mon, 27 Jul 2020, 00:53 I type out the command+crc which I get from Scott's little CrcGen application so i type in "5150494753B7A9" and then press enter, then I receive the NAKss response,
Scott's app gives the result in hex characters, so you have to have your terminal program in hex mode. As far as I know, Teraterm doesn't have a hex mode, so to use the result as-is you'd have to use Realterm. I haven't used that for some time; I think you have to set it up to accept hex characters, and convert pairs of hex characters to send single characters. For example, the 'Q' of "QPIGS" has to be sent as a single character; Realterm has to convert the pair '5' and '1' into a single 'Q' to send. Similarly, it combines the 'B' and '7' into a single 8-bit character. So when you turn on local echo, you should see 'Q' 'P' ... not '5' '1' '5' '0' ...
I did notice in the command protocol documentation the format of the commands is <command><crc><cr>, but I don`t know what the <cr> value is,
Heh. CR stands for Carriage Return, a relic from the days of typewriters and teletypewriters, where there was an actual mechanical carriage with the print head that had to return to the left margin when it received that control character. Modern keyboards call that control character "enter". That's probably a relic of the days before graphical operating systems, when you had to type commands to the operating system, and the enter key told the system that the command was finished, i.e. there is no more to type for this command. It's still needed with the inverter commands, since they are of variable length, some take a parameter or parameters, and so on. The carriage return control character has the hex value 0D (= decimal value 13). I believe that even in hex mode, Realterm adds the carriage return to the end of your command, though you may have to adjust some settings for this to happen.
I read Coulombs post on how to directly send commands using Tera-term and I edited the teraterm.ini file as described, but when I try to enter characters with a high bit set such as 0xA7 which Coulomb describes as pressing (right-alt-7) nothing happens inside of Tera-term, even with local echoing enabled.
Then the right-alt key isn't set up yet; see the instructions on editing the TERATERM.INI file, and change the MetaKey and Meta8Bit settings. Restart Teraterm to take effect. It's a pain to get it set up, but you only have to do it once. If using right-alt, make sure that nothing else is "capturing" that key. For example, I use WinCompose, and it captures that key to compose UTF8 characters, such as √ (I type right-alt v / to get that to come out). So I need to use MetaKey=left .
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.
paulvk
Senior Member
Posts: 457
Joined: Fri, 23 Oct 2015, 23:45
Real Name: Paul
Location: Sydney

Re: PIP-4048MS and PIP-5048MS inverters

Post by paulvk »

To send hex in Realterm you use this format 0x7E 0x2F
you do this in the send tab
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

MarkusRE wrote: Sun, 26 Jul 2020, 22:54 Is there a way to go back to the original firmware somehow because of warranty?
You can't automatically revert to the firmware the unit came with; new firmware replaces the old; there isn't enough flash space to keep the original firmware.

The reflash tool (for the PIP-MS series) always uses the fixed file name dsp.hex. But you can always revert to factory firmware that a patched firmware is based on (i.e. the unpatched version). Every patched firmware zip file contains the unpatched file, dsp_original.hex (as well as the patched file, dsp_L.hex). So you can delete dsp.hex, then copy-and-paste dsp_original.hex and rename the copy to dsp.hex.

Start the reflash tool, and it will reflash to factory firmware version 74.40. That's official factory firmware for your machine, so there should be no problem with that.

You should be able to reflash your machine, even with it showing fault code 09. You should also be able to verify after the reflash that it is in factory firmware by pressing the up key; there should be a screen with "U1 74 40" instead of "LF1 74 40E".
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.
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

Good to hear, that itis possible, but I installed the 72.20e version and original was 72.40. So in the zip is no 72.40. Does anybody maybe have this? Inverter is the 64V version.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

MarkusRE wrote: Mon, 27 Jul 2020, 19:59 Good to hear, that itis possible, but I installed the 72.20e version and original was 72.40. So in the zip is no 72.40.
Ah. 64 V. Then the zip will have 72.20, obviously.
Does anybody maybe have this? Inverter is the 64V version.
My 2015 machine came with a 72.40. But obviously that's a different version; yet another version number collision.

I'm not aware of any 72.40 (PF1 for 64 V) firmware available for download. I doubt that they would notice that your machine came with 72.40 if you put 72.20 into it.
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.
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

coulomb wrote: Mon, 27 Jul 2020, 20:17
MarkusRE wrote: Mon, 27 Jul 2020, 19:59 Good to hear, that itis possible, but I installed the 72.20e version and original was 72.40. So in the zip is no 72.40.
Ah. 64 V. Then the zip will have 72.20, obviously.
Does anybody maybe have this? Inverter is the 64V version.
My 2015 machine came with a 72.40. But obviously that's a different version; yet another version number collision.

I'm not aware of any 72.40 (PF1 for 64 V) firmware available for download. I doubt that they would notice that your machine came with 72.40 if you put 72.20 into it.
Exactly that is my problem. Before doing a warranty claim it need to be the rigt firmware I think. So I cross fingers to find a solution.
bigjsl
Noobie
Posts: 19
Joined: Wed, 07 Oct 2009, 03:08
Real Name: John Lindsay
Location: Adelaide
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by bigjsl »

It's been a while. My pair of Chinese knockoff PIP4048MS have been running for almost two years. I had an early PIP failure and had to use my spare but otherwise everything has been OK. Recently I noticed three out of four fans had stopped. I did what any sane person who bought Chinese knockoffs would do and ordered two new PIP5048MS units from MPP via eBay. I have one of them and other almost arrived thanks to DHL today but I should get it tomorrow. Some of you may remember that after two second hand laptops I concluded I would not be able to install the wonderful hacked firmware on my sad clone inverters. Recently I updated by python monitoring software that uploads stats to pvoutput so it switches between UTL & SBU modes to minimise how much mains electricity I buy to charge the batteries. This seems to be working fairly well. Next addition should be something to force a full bulk charge but I'm waiting for bug free charging first. Well now I have proper Taiwanese PIPs so here I go. (I will sort out the old units so the fans work (and blow up in both units, only changed both the original deployed units. I may bring one back into service as a solar charger.)
kylebotha
Noobie
Posts: 2
Joined: Sun, 26 Jul 2020, 22:54
Real Name: Kyle

Re: PIP-4048MS and PIP-5048MS inverters

Post by kylebotha »

paulvk wrote: Mon, 27 Jul 2020, 08:05 To send hex in Realterm you use this format 0x7E 0x2F
you do this in the send tab
Thank you :D It works
MarkusRE
Noobie
Posts: 27
Joined: Sun, 03 May 2020, 18:34
Real Name: Markus Schulz

Re: PIP-4048MS and PIP-5048MS inverters

Post by MarkusRE »

As my EASUN blew up, I rdered a new one for 660€, which was planned for next year.

Now my question. I‘m sure the answer is NO, but I have a litte bit of hope.

Is it possible to download the original firmware from the new one and flash it to my broken one? I‘m in contact with EASUN, trying to get the original, but this engeneer I‘m talking with is a bit stupid.

Or an other idea, can I modify the original 72.20, that it just tell it is a 72.40? I think they won‘t check in detail.
greyworm
Noobie
Posts: 4
Joined: Sun, 11 Feb 2018, 04:26
Real Name: Chris
Location: Ledziny, Poland, JO90ND11
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by greyworm »

Hello everyone.
Greetings from Poland - this is such a country in Europe :)
Sorry for my (maybe) not very Shakespearean English, unfortunately I don't use it very often ...
I would like to add my five cents to the MS series inverters knowledge base.
Currently, I am in possession of two copies of these inverters.
One - purchased in 2017 and branded as ORVALDI MKS5K MPPT3000W 48V, sold by Orvaldi in Poland.
The second one was purchased a few days ago from EASUN on Aliexpress - 5048MS 64VDC.
In the next post I will add photos of stickers.
Both of them were able to reprogram your firmware without any problems.
Ultimately intended for parallel operation.
Both look identical from the outside, except for the names of the distributors on the housings.
During the assembly of the card for parallel operation, I had the opportunity to see the new one in more detail and I must say that, apart from the fan brand, it does not differ from its older "brother".
So you can add to the list of brands these two series of products that can be flashed by firmware developed by you.

I have to admit that the first inverter works impeccably throughout its lifetime.
It works in island mode with emergency bypass from the network, powered by a set of panels with a total power, at the moment, about 5100W.
At night, it "feeds" on LTO batteries in the 21S8P system, made of 66160 cells with a capacity of 40Ah each (168 units).
At a nominal voltage of 48 V it gives about 15.3 kWh of power.
Temporary power shortages recently forced the purchase of a second copy to increase the possibilities of the system.
The system supplies the entire single-family house 24/7/365 with the few cases when you need to "support" recharging the battery at night (cheaper tariff) in the winter.

For the first post, I think it's enough ... :P .

Best regards,
Vy 73 de SP9XCC,
Chris (Krzysztof)
Last edited by greyworm on Thu, 30 Jul 2020, 15:21, edited 1 time in total.
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

greyworm wrote: Wed, 29 Jul 2020, 18:19 Greetings from Poland
Welcome to the forum.
Currently, I am in possession of two copies of these inverters.
One - purchased in 2017 and branded as ORVALDI MKS5K MPPT3000W 48V, sold by Orvaldi in Poland.
The second one was purchased a few days ago from EASUN on Aliexpress - 5048MS 64VDC.
These are quite possibly somewhat different units, despite appearances.
In the next post I will add photos of stickers.
That will be good.
Both of them were able to reprogram your firmware without any problems.
What versions of patched firmware are you using in them?

The first ones's model number ("MKS5K MPPT3000W 48V") suggests strongly that it has a 3000 W (60 A max output) solar charge controller. That makes it the same as mine, manufactured in 2015, and PF0.8 (4000 W / 5000 kVA). It would also be a 58.4 V model (not a 64 V model).
Ultimately intended for parallel operation.
I don't think you're going to be able to parallel these, sadly. It's a problem for a lot of owners in your situation (as I assume it; we'll know for sure with the label photos), where you have a non-64 V PF0.8 machine, which is no longer manufactured, and you want to parallel it with a new model.

I'm pretty sure you can't parallel a 64 V model with a 58.4 V model.
Both look identical from the outside, except for the names of the distributors on the housings.
It's still a mystery to me whether the PF0.8 models and the PF1.0 models differ in hardware at all. In other words, is the difference purely firmware, so that a PF0.8 model could be upgraded to a PF1.0 model merely with different firmware.

The difference between a 64 V model and a 58.4 V model will be subtle, for example the voltage rating of capacitors and MOSFETs, and the turns ratio of the transformer (which has no part number to distinguish between models). The transformer may be different in other ways to allow it to pass 25% more power (5 kW versus 4 kW), but these might not be visible unless you really know what you are looking for.
During the assembly of the card for parallel operation, I had the opportunity to see the new one in more detail and I must say that, apart from the fan brand, it does not differ from its older "brother".
I assume that you did not take off the main cover (only the smaller cable cover), which would break the small sticker and probably void the warranty. So you are really only looking at superficial differences.
So you can add to the list of brands these two series of products that can be flashed by firmware developed by you.
Thanks.
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.
greyworm
Noobie
Posts: 4
Joined: Sun, 11 Feb 2018, 04:26
Real Name: Chris
Location: Ledziny, Poland, JO90ND11
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by greyworm »

coulomb, thank you for you reply,


In the older 74.40e copy, and in the new 72.20e (I am considering an attempt to upload the 74.40e version to it).

Image Image

I suspect that the difference is in the permissible DC voltage and lowering it by limiting the firmware should not affect its operation. Unfortunately, it does not display information about the firmware version of the SCC chip.
In the older one I have there on 4.12
...where you have a non-64 V PF0.8 machine,...
No, I have two units with PF1.0.

Image Image
...but these might not be visible unless you really know what you are looking for...
I know what to look at.
On a daily basis, I deal with (among others) a bit larger of their brethren working at voltages of 3.8kV or 6.3kV.
I work as a commissioning engineer for the electrical part of a power plant ... currently on a 910MW unit...

I attach photos of stickers and displays.
I hope they will be visible (various forums adopt ...)

Generally, I would like to thank you for your commitment to the cause.
I have already learned many things thanks to you about this series of inverters, and I suspect I will find out more ...
There is never too much knowledge.

Regards,
Vy 73 de SP9XCC,
Chris (Krzysztof)
Attachments
MKS5K_sticker.png
MKS5K_sticker.png (372.29 KiB) Viewed 3755 times
MKS5K_display.png
MKS5K_display.png (177.84 KiB) Viewed 3755 times
Isolar_sticker.png
Isolar_sticker.png (340.99 KiB) Viewed 3755 times
Isolar_display.png
Isolar_display.png (86.64 KiB) Viewed 3755 times
----------------------------------------------
5,1 kW PV in use (10 x 125W in 2S5P + 15 x 255W in 3S5P),
2 x 5048MS series,
168 cells of 40Ah 66160 LTO battery in 21S8P ~15,3kWh at 48V DC,
4,5kW DC water heater with PWM control,
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

greyworm wrote: Thu, 30 Jul 2020, 15:43 In the older 74.40e copy, and in the new 72.20e
OK. So they definitely won't parallel as is; they have different paralleling constants.
(I am considering an attempt to upload the 74.40e version to it).
Consider carefully. You'll be attempting to run firmware assuming a 1:8 transformer ratio on hardware with a 1:7 ratio transformer. There are parts of the code that e.g. set limits for the bus voltage, and they involve the battery voltage times the assumed turns ratio. Whether that will just disallow an otherwise usable area of operation, or will endanger the hardware, I don't know.
Unfortunately, it does not display information about the firmware version of the SCC chip.
In the older one I have there on 4.12
The version of SCC firmware should not matter too much, but you should be able to see it by putting a power supply on the PV input so that the SCC powers up.
No, I have two units with PF1.0.
OK. Your old machine takes the prize for the most confusing and inconsistent model name ever.
On a daily basis, I deal with (among others) a bit larger of their brethren working at voltages of 3.8kV or 6.3kV.
I work as a commissioning engineer for the electrical part of a power plant ... currently on a 910MW unit...
Respect! 🧐
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
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by weber »

We have just bumped patched firmware 74.40e from beta to release.

This is firmware for some 5 kW / 5 kVA / PF1 models without the 64 V option.

The only change was the first letter of the version number (from B to L). This was after 5 months and 107 downloads of the LFP version and 82 downloads of the Lead aCid/Lithium Cobalt version, with zero bug reports.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
greyworm
Noobie
Posts: 4
Joined: Sun, 11 Feb 2018, 04:26
Real Name: Chris
Location: Ledziny, Poland, JO90ND11
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by greyworm »

Hi,

coulomb... unfortunately you are right.
You'll be attempting to run firmware assuming a 1:8 transformer ratio on hardware with a 1:7 ratio transformer.
I checked organoleptically - the voltage on the DC bus in the new inverter is lower - approx. 375V at 54.4V on the battery instead of approx. 440V in the old one ...

But no one promised it would be easy...

With my high impulse loads (starts of electric motors, water pumps and refrigerators) and their inrush current times up to 12 x the nominal current (!), I will probably start looking towards the inverter at low frequency ...

Currently, unfortunately, when the stove (approx.2.4kW) or the washing machine (heater approx.2.2kW) is running and the water pump is turned on, the inverter stops.
The pump has only 720W of nominal power, but the measured impulse during the start (registration on the network analyzer) is 7900W with a duration of over 150 miliseconds.
As a result, it gives an instantaneous load of the inverter over its limit of 10kW surge load and its shutdown.
What's most interesting, the inverter does not register any alarm or error ...
It just goes dark for a few seconds and restarts (auto restart active).

It is not dependent on the power on the panels or the voltage on the battery.
My battery has never dropped below 44V for the entire time of use.
Despite the active power supply, the bypass does not start, and in my opinion it should switch to the bypass in this case ...
Unless the bypass here is a pseudo bypass connected to the DC bus ...

In this case, nothing will help, it will drop off anyway at each overload impulse beyond the set limit (self protection).

One more theory to check - does the BMS limit access to the battery ...
I have a version up to 200A, but the response time may be too long and the voltage on the battery input side of the inverter drops.
I have the kettle correction on, but the impulse can be so short that the correction will not catch it, although the display generally shows the information when the load is started that it raises the battery voltage indication (at least it shows that it is raising the indication).

I have a separate active balancer installed on the battery, due to its capacity, the one in the BMS was unable to maintain the voltage on the cells, so I can bypass the BMS for a few days and watch what will happen ...

The same thing happened with the original firmware, so it doesn't depend on that either.

It is true that I tried the new one with the changed firmware, but I suspect that the same as in the old one, it does not matter.

The weekend is coming, I will observe and measure ... this time on the battery side.

Regards,
Vy 73 de SP9XCC,
Chris (Krzysztof)
----------------------------------------------
5,1 kW PV in use (10 x 125W in 2S5P + 15 x 255W in 3S5P),
2 x 5048MS series,
168 cells of 40Ah 66160 LTO battery in 21S8P ~15,3kWh at 48V DC,
4,5kW DC water heater with PWM control,
Malan
Noobie
Posts: 1
Joined: Sun, 09 Aug 2020, 16:10
Real Name: Malan van Zyl

Re: PIP-4048MS and PIP-5048MS inverters

Post by Malan »

The shore power lead/AC in (unplugged) measures 220V between Neutral and Earth in a 5KVA unit. Why could that be? For safety it has been removed and the unit is running on solar only. Could a small UPS downstream from the inverter play a role??
User avatar
rthorntn
Groupie
Posts: 135
Joined: Sat, 06 Jun 2020, 08:36

Re: PIP-4048MS and PIP-5048MS inverters

Post by rthorntn »

Hi,

I'm interested in what KettleComp is actually compensating for, the science of it, why does adding 2KW of load to my big battery lower the voltage by 0.5V, does this voltage drop happen on a Victron BMV?

Thanks.
Richard
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
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: PIP-4048MS and PIP-5048MS inverters

Post by weber »

rthorntn wrote: Tue, 18 Aug 2020, 07:40 I'm interested in what KettleComp is actually compensating for, the science of it, why does adding 2KW of load to my big battery lower the voltage by 0.5V, does this voltage drop happen on a Victron BMV?
Hi Richard. It is compensating for the internal resistance of the battery. An ideal battery would keep a constant voltage no matter what load you put on it. But real batteries aren't like that. A real battery can be modelled as an ideal battery with a small resistance in series. We can even calculate (approximately) how many ohms that resistance is, from the figures you gave. Lets call your battery 50 volts to keep the numbers simple. So 2000 watts of load corresponds to 2000/50 = 40 amps.

If your battery sags 0.5 volts when the current is 40 amps, its internal resistance is 0.5/40 = 0.0125 ohms, or 12.5 milliohms.

This also causes your battery voltage to rise when being charged. Of course we're talking about the instantaneous (or within a few seconds) sag or rise in voltage purely due to changes in current. Not the slow change in voltage due to state of charge. So by compensating for the changes in voltage caused purely by changes in current, KettleComp lets the inverter more accurately estimate the state of charge from the compensated voltage. But its SoC estimate is still not very accurate, maybe +-25%. The main point of KettleComp is to ensure that the inverter does not shut off, or switch back to the utility, merely because there is a heavy load. It ensures that a low value of compensated voltage will really represent a low state of charge.

Yes, your Victron BMV sees those changes in voltage too, but it doesn't need to compensate for them because it is using a far more accurate means of measuring the state of charge of the battery, called "coulomb counting" or "amp-hour counting". If it samples the current every second, then because there are 3600 seconds in an hour it can add 1/3600 of every current reading into an accumulating register. And you've told it what the battery capacity is, so it can calculate the percentage state of charge at any moment. But this value can slowly drift out of sync with the actual SoC if there is any offset in the current measurement, so it relies on voltage to recognise when the battery is fully charged, so it can resync to 100%.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
Post Reply