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.
nonlinear
Noobie
Posts: 3
Joined: Tue, 02 Mar 2021, 16:46
Real Name: Sven

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

Post by nonlinear »

The lack of release notes is indeed vexing.

I have noticed the following which may be coincidental but could also point to changes: on my previous system when I engaged either my pressure washer. off-cut saw, or chipper, the system would immediately switch to grid irrespective of the Kettle comp setting. This no longer occurs, it is able to ride out surge currents better than in the past. Anecdotal at best, but who knows?
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 »

nonlinear wrote: Thu, 04 Mar 2021, 16:51 on my previous system when I engaged either my pressure washer. off-cut saw, or chipper, the system would immediately switch to grid irrespective of the Kettle comp setting. This no longer occurs, it is able to ride out surge currents better than in the past.
Good to know. I guess it's because the battery to output path is now 5 kW all the way, instead of being limited to 4 kW battery to bus.
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
Greendrake
Noobie
Posts: 12
Joined: Sun, 09 Jun 2019, 19:55
Location: New Zealand
Contact:

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

Post by Greendrake »

I beg pardon if this has been discussed before: voltage between earth and neutral/line AC output.

I have an off-grid/floating system. No earth cables actually go deep into the ground. I simply wire all metal cases across the property together and call it "earth".

The inverters (two in parallel) have earth wire attached as per the manual. When there is no load, I observe 78V AC between neutral and earth, and 154V between line and earth (230V between line and neutral). With load connected, neutral/earth jumps to 85V, line/earth drops to 146V.

Is that normal? Shouldn't the neutral be just bound to earth inside the inverters?

Moreover, the battery/DC circuit is also not indifferent to earth. When the battery is say 50V, The positive terminal is at roughly 30V to earth, the negative at −20V.

Obviously, since there is voltage difference, earth is not bound to any of those AC/DC terminals. What I really struggle to get is how earth could possibly be part of either AC or DC circuit at all. If there is voltage difference, it means there would be current flowing should the wires be connected? Where would it flow through, and how big would it be?

The inverters are these clones:

nonlinear
Noobie
Posts: 3
Joined: Tue, 02 Mar 2021, 16:46
Real Name: Sven

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

Post by nonlinear »

I have not reviewed the AS/NZ standards in a long time but I am relatively familiar with SANS and NEC and I'm fairly sure that your installation is not according to code and poses a danger. As a first step you should probably ensure that your earthing is referenced to your home earth which should be bonded according to code (usually at your incoming DB). In NEC and SANS installations (i.e. USA and SA) this will be bonded to (usually) the supply transformer earth which (usually/often) is also referenced to the neutral (in residential installations).

The bonding of the neutral in an Axpert to the earth/neutral of the system during on/off grid operation is a very thorny topic and I don't think I have the necessary reputation on this forum to get involved in that fight. Maybe other forum members want to comment what they believe is the best strategy.
User avatar
Greendrake
Noobie
Posts: 12
Joined: Sun, 09 Jun 2019, 19:55
Location: New Zealand
Contact:

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

Post by Greendrake »

nonlinear wrote: Fri, 05 Mar 2021, 14:57 I have not reviewed the AS/NZ standards in a long time but I am relatively familiar with SANS and NEC and I'm fairly sure that your installation is not according to code and poses a danger. As a first step you should probably ensure that your earthing is referenced to your home earth which should be bonded according to code (usually at your incoming DB). In NEC and SANS installations (i.e. USA and SA) this will be bonded to (usually) the supply transformer earth which (usually/often) is also referenced to the neutral (in residential installations).
The thing is, I am committed to follow pure reason and laws of physics rather than to blindly follow the standards. Those standards were created to accommodate many various scenarios at once, and may not necessarily be technically justified in my not-so-common completely off-grid scenario. (I admit the standards will provide safety, but that is certainly not the only way to provide it and certainly not the most efficient one in my case). Let's also set aside any (non-)legality of not following the standards. That is a separate topic and is not what I am asking about.

In an off-grid system like mine, the physical earth/ground is no part of the circuit. There are no transformers. The neutral does not get bonded to any buried metal rod anywhere. The is actually no "home earth" at all as the house has never been connected to the grid. (Lightning safety is provided by a lightning arrester — totally separate from the house and from any electrical circuits, although very close to it).

What there is instead are metal boxes/cases, sheds, shipping containers etc. They are all electrically connected together and are what I call "earth".
nonlinear wrote: Fri, 05 Mar 2021, 14:57 The bonding of the neutral in an Axpert to the earth/neutral of the system during on/off grid operation is a very thorny topic
That is exactly why I am raising it. Have to find the truth about it, the bottom of it, dot the i's and cross the t's.
nonlinear wrote: Fri, 05 Mar 2021, 14:57 I don't think I have the necessary reputation on this forum to get involved in that fight.
Certainly no need to fight. But if you have any sound technical thoughts/reasons based on the laws of physics (as opposed to citing the standards), then they are very valuable regardless of your reputation.
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 »

Greendrake wrote: Fri, 05 Mar 2021, 13:14 I have an off-grid/floating system. No earth cables actually go deep into the ground. I simply wire all metal cases across the property together and call it "earth".
The problem is that the potential of all the metal cases could be different to the potential of a human or other animal with their legs on the ground (real earth). This could cause injury or death.
The inverters (two in parallel) have earth wire attached as per the manual. When there is no load, I observe 78V AC between neutral and earth, and 154V between line and earth (230V between line and neutral).
That's because of the required EMI "Y" capacitors, between neutral and earth, and active and earth. With earth effectively floating, it forms a potential divider. These capacitors have a small value (usually 1 nF), so their impedance at 50 Hz is quite high (around 3 MΩ if my rusty maths is correct). Your multimeter (usually with an input impedance of around 10 MΩ) drags the voltage from the centre of active and neutral.

There are usually also "X" capacitors, these connect across the line (i.e. neutral to active). These have no effect on earthing. Both X and Y capacitors have to be safety rated. You'll see a lot of fine writing on them, with all their ratings and safety certificates.
With load connected, neutral/earth jumps to 85V, line/earth drops to 146V.
I think that means that some of your loads also have these "Y" capacitors, and the values must be such that it pulls the earth towards active a little, and away from neutral a little.
Is that normal?
The neutral to earth value should be very low, less than about 10 VAC. But it's normal to see this sort of behaviour if the earth isn't connected properly.
Shouldn't the neutral be just bound to earth inside the inverters?
No. That would create problems with earth leakage detection systems. There should be one and only one bonding of neutral to earth at one premises.
Moreover, the battery/DC circuit is also not indifferent to earth. When the battery is say 50V, The positive terminal is at roughly 30V to earth, the negative at −20V.
There are "Y" capacitors from the battery positive and negative to earth as well, but of course they should not affect DC measurements. I have no idea what's inside your clones.
If there is voltage difference, it means there would be current flowing should the wires be connected? Where would it flow through, and how big would it be?
It flows "through" the "Y" capacitors, and is generally a fraction of a milliamp. But there are several of them, and the current adds up. People sometimes have trouble with residual current breakers if they have too many appliances with these capacitors in them, causing nuisance tripping.

To be safe, neutral out of your inverter has to connect to one of these places:
  • Neutral of AC-in, when in line mode. This neutral should be earthed somewhere upstream.
  • Directly to earth inside the inverter, when not in line mode. Some models (e.g. the genuine models that came with main firmware version 73.00) have this feature. This connection must only be when the inverter is not in line mode (there should never be more than one connection from neutral to earth in any one premises).
  • Directly to earth outside the inverter, again only when not in line mode. Some inverters have the ability to signal when the inverter neutral needs to be tied to earth, and this can drive an external relay that makes this connection.
In all cases, the connection to earth has to be strong enough to clear an active to earth fault in an appliance.

One way around this is to connect AC-out neutral to AC-in neutral, for cases when the inverter never makes the connection. AC-in neutral should then bond to earth (and to a proper earth stake) at the one and only point. I assume that even though you are off-grid, you occasionally need to run a generator and connect it to AC-in of the inverter, so AC-in will be connected to something. You should have your loads protected by residual current breakers with overload (RCBOs).
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
Greendrake
Noobie
Posts: 12
Joined: Sun, 09 Jun 2019, 19:55
Location: New Zealand
Contact:

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

Post by Greendrake »

I bow my head, coulomb. Immensely helpful info!
coulomb wrote: Fri, 05 Mar 2021, 17:43 The problem is that the potential of all the metal cases could be different to the potential of a human or other animal with their legs on the ground (real earth).
As I understand, the only cause of that would be static electricity somehow accumulated on the metal cases, as referenced to the real earth. For that to happen, I guess the metal cases would need to be rubbed against the real earth; the static charge against it won't accumulate simply by way of having currents running in floating circuits inside those boxes, will it? So, is the scenario where that static potential gets big enough to cause injury/death really realistic, especially that those metal boxes do stand on the ground (via sleepers, concrete slabs etc.) — although the connection may not be as good as bare feet standing on wet soil?
coulomb wrote: Fri, 05 Mar 2021, 17:43 That's because of the required EMI "Y" capacitors, between neutral and earth, and active and earth. With earth effectively floating, it forms a potential divider. These capacitors have a small value (usually 1 nF), so their impedance at 50 Hz is quite high (around 3 MΩ if my rusty maths is correct). Your multimeter (usually with an input impedance of around 10 MΩ) drags the voltage from the centre of active and neutral.
Extremely good to know! Thank you. Will dig it.
coulomb wrote: Fri, 05 Mar 2021, 17:43 The neutral to earth value should be very low, less than about 10 VAC. But it's normal to see this sort of behaviour if the earth isn't connected properly.
Sorry, I still do not get how a properly connected earth would make any difference to that voltage — given that, unlike on-grids, the real earth is no part of the circuit in this off-grid. Are you saying that the Y-capacitors' charge will be interested to go into the real earth should it be properly connected?
coulomb wrote: Fri, 05 Mar 2021, 17:43
Shouldn't the neutral be just bound to earth inside the inverters?
No. That would create problems with earth leakage detection systems. There should be one and only one bonding of neutral to earth at one premises.
I am actually skeptical about getting leakage detection systems work in off-grid floating systems. If I was aiming for it, where would I make that "one and only one bonding of neutral to earth"? Would I say just run the neutral from the inverter a few dozen metres away and attach it to a properly buried earthing rod? And then use properly earthed rods where the inverters are installed, thereby connecting the earth and the neutral not directly but via the whole volume of soil in my backyard?

Also, just as a thought experiment. How would your answer be different if we were talking about a literally floating system (e.g. a spaceship) where there is no ground to stand on, let alone bury earthing rods into? Just the frame of the ship.
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 »

Greendrake wrote: Fri, 05 Mar 2021, 18:47 As I understand, the only cause of that would be static electricity somehow accumulated on the metal cases, as referenced to the real earth.
No, static electricity is another problem.

The problem with a floating system is that when there is a leakage (e.g. neutral or active to earth, even your pretend earth), there is no immediate indication of a fault. So when you get a second leakage (e.g. a human touching an active conductor), there is a potentially hazardous or even lethal situation. So with a fully floating system, you have to have monitoring to make sure it stays floating. Elevtric vehicles' traction batteries are an example of this. Ever time they start up, you hear the "dance of the contactors", part of which is testing insulation monitoring. If a leakage is detected, it's flagged as a fault and the car goes nowhere. That way, a double leakage leading to a shock never happens, unless the technician investigating the fault does something stupid. You could run a fully floating system, as long as you did the insulation monitoring. But it's usually easier just to do the earthing system, as if it was connected to the grid.
coulomb wrote: Fri, 05 Mar 2021, 17:43 The neutral to earth value should be very low, less than about 10 VAC. But it's normal to see this sort of behaviour if the earth isn't connected properly.
Sorry, I still do not get how a properly connected earth would make any difference to that voltage — given that, unlike on-grids, the real earth is no part of the circuit in this off-grid. Are you saying that the Y-capacitors' charge will be interested to go into the real earth should it be properly connected?
I'm saying that with a correctly wired grid connected system, there will be less than 10 V from neutral to earth, because of the bonding between neutral and earth near the incoming mains. So that keeps the voltage below 10 VAC (all you get is the voltage drop due to neutral load current). With an off-grid system, it should be the same. With your "floating" system, earth is actually connected to a weak voltage divider between neutral and active, by the "y" capacitors. So you'll get way more than 10 VAC from neutral to earth, so touching neutral could give a tingle (the strength of the tingle depending on the number and capacitance of the capacitors involved). Contact with any neutral is an unusual situation, sure. But I'm not comfortable having that neutral to earth voltage. I believe that RCBOs are just as important off-grid as on-grid; an inverter has many many times the power needed to electrocute a human.
I am actually skeptical about getting leakage detection systems work in off-grid floating systems.
I don't agree with your scepticism here.
If I was aiming for it, where would I make that "one and only one bonding of neutral to earth"?
I would make it part of the connector for the generator. That leads to AC-in of the inverter. If your inverter never connects neutral and earth (with the genuine models, this is easy to check: they will have continuity from AC-out neutral to earth when powered down if they have this), connect AC-out neutral to AC-in neutral. It's cheating a little; shorting AC-in N and AC-out N defeats one of the safety relays.
Would I say just run the neutral from the inverter a few dozen metres away and attach it to a properly buried earthing rod? And then use properly earthed rods where the inverters are installed, thereby connecting the earth and the neutral not directly but via the whole volume of soil in my backyard?
No, all the earths should be connected together with copper, not "via the earth" (as in soil).
How would your answer be different if we were talking about a literally floating system (e.g. a spaceship) where there is no ground to stand on, let alone bury earthing rods into? Just the frame of the ship.
In that case, the frame of the ship (presuming it all connects together) acts as the local "earth", because the potential of any human will almost certainly be the same or very close to that of the frame of the ship. That's why we talk about "earth"; we almost can't help being at earth potential.
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.
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 »

My #2 PIP clone finally died so I retired #3 and got new genuine MPP units #1 and #2 out of storage. Replacing the old units was fairly straight forward but the mounting points are slightly different which required some new holes. Because the new units are genuine I was finally able to load Coulomb and Weber's patched firmware. Many many thanks guys!

Amusing problem: remember the POP command CRC problem with some clones? I had that problem and I forgot to remove the work around from my RaspberryPi code that monitors and occasionally adjusts settings.

Actually I didn't have this problem but I do need to manually override the CRC16 code that the python crc16 library returns.
ser.write('\x50\x4f\x50\x30\x32\xe2\x0b\x0d')
works but
ser.write('\x50\x4f\x50\x30\x32\xe2\x0a\x0d')
does not. The second is the XMODEM CRC16 returned by https://crccalc.com.

jsl
E85
Noobie
Posts: 6
Joined: Wed, 11 Sep 2019, 14:16

MPPT -> Battery Current

Post by E85 »

I have a question that I hope the experts on this forum can answer.

I would like to use a current transformer to measure the current flowing from the MPPT solar charger into the battery.

I was hoping I could do something like this. The lower ”battery discharge current clamp” would measure all current in and out of the battery but also measure the solar charge current backwards so that the netto would be all discharge current minus any charge current from AC, the upper ”battery charge current clamp” would only measure the solar charge current
123SmartBMS_PLI_5000_dual_clamp_setup.jpg
123SmartBMS_PLI_5000_dual_clamp_setup.jpg (43.96 KiB) Viewed 4705 times
I thought I could use the easily accessible red or black wires in the inverter more or less this way:
Solarix+PLI_5000_open+text+current-transformer.jpg
Solarix+PLI_5000_open+text+current-transformer.jpg (94.11 KiB) Viewed 4705 times
But now I think this is not possible because it seems that the red and black wires that I though where suitable for this are before a DC-DC converter and having a higher voltage then the battery?

If that is the case, is the voltage in the black and red wire a fixed voltage, related to the battery voltage or related to the MPPT voltage of the solar panels?

Could there be another point inside the inverter where I could make an extra detour to measure (only) the current from the solarcharger into the battery or is this not at all possible?

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

Re: MPPT -> Battery Current

Post by coulomb »

E85 wrote: Thu, 18 Mar 2021, 18:42 I would like to use a current transformer to measure the current flowing from the MPPT solar charger into the battery.
Current transformers only work with AC, and the solar charge controller (SCC) is basically DC. You could use a Hall effect device, but you'd have to deal with the drift problem.

Or you could use the shunt resistors that are already there. Or you could query the inverter with suitable commands, and have it tell you what it's already measured for you, with noise filtering, calibration, etc.
I thought I could use the easily accessible red or black wires in the inverter more or less this way:
Solarix+PLI_5000_open+text+current-transformer.jpg
There is a post I wrote years ago, accessible from the PIP inverter repairs and hardware modifications topic index, titled "The red and black wires are NOT SCC output." It indicates the two 0.33 mΩ shunts which could be used to measure the current through the two halves. But as I mentioned above, they already make this measurement available to you via commands such as QPIGS (Query General Status).

[ Edit: 0.33Ω -> 0.33 mΩ ]
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.
E85
Noobie
Posts: 6
Joined: Wed, 11 Sep 2019, 14:16

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

Post by E85 »

Thank you for your reply!
coulomb wrote: Thu, 18 Mar 2021, 19:35 Current transformers only work with AC, and the solar charge controller (SCC) is basically DC. You could use a Hall effect device, but you’d have to deal with the drift problem.
Yes correct, I realy meant a hall effect type non evasive DC current sensor that has some circuitry inside that you see on the photo, it is part of the 123smartBMS I’m using.
coulomb wrote: Thu, 18 Mar 2021, 19:35Or you could use the shunt resistors that are already there. Or you could query the inverter with suitable commands, and have it tell you what it's already measured for you, with noise filtering, calibration, etc.
I thought I could use the easily accessible red or black wires in the inverter more or less this way:
Solarix+PLI_5000_open+text+current-transformer.jpg
There is a post I wrote years ago, accessible from the PIP inverter repairs and hardware modifications topic index, titled "The red and black wires are NOT SCC output." It indicates the two 0.33 mΩ shunts which could be used to measure the current through the two halves. But as I mentioned above, they already make this measurement available to you via commands such as QPIGS (Query General Status).

[ Edit: 0.33Ω -> 0.33 mΩ ]
Yes, I had found that post before, but I didn’t fully understand it. Now when I look at it again I understand that both the black and red wires are connected to the battery plus, each via its 0.33 mΩ shunt and a relay?

If that is the case I could have both wires together pass through my current sensor and thus separately measure the battery charge cuurent from the MPPT charger.

Building an interface that emulates the 123smartBMS’ current sensors and using the QPIGS command to read the PV-InputCurrentForBattery, BatteryDischargeCurrent and BatteryChargingCurrent would be an elegant alternative. But I think for me that would be quite complicated too.

Thanks.
EnnO1
Noobie
Posts: 1
Joined: Sat, 16 Jan 2021, 09:40
Real Name: Enrico

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

Post by EnnO1 »

Thanks Coulomb and Weber for the great work.

I own 3 Steca Solarix PLI 5000-48 Rev.2 (2019) with Pylontech US3000 and use LF1_72.20e.
The original U1_72.40 was on the Steca.
There is now an update to version U1_72.40 on the Steca homepage (https://www.steca.com/frontend/standard ... 000_48.zip)
Could you use this update to create an LF1_72.40e?
Firmware 72.20e works fine and I couldn't see any difference from 72.40. (KettleKomp is very good)
I thought it would help one or the other here who is looking for the original firmware 72.40.

Have a nice 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-4048MS and PIP-5048MS inverters

Post by coulomb »

EnnO1 wrote: Sat, 20 Mar 2021, 17:16 There is now an update to version U1_72.40 on the Steca homepage
Thanks!
Could you use this update to create an LF1_72.40e?
We could. But it's a lot of work.
Firmware 72.20e works fine and I couldn't see any difference from 72.40.
A quick perusal indicates that there are BMS commands in there. I wonder what they have in mind. I suspect that they are preparing for a version with a removable display and BMS support. But that won't affect most if any users with current hardware, and the BMS support may not even be finished or complete.
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.
E85
Noobie
Posts: 6
Joined: Wed, 11 Sep 2019, 14:16

Re: MPPT -> Battery Current

Post by E85 »

coulomb wrote: Thu, 18 Mar 2021, 19:35
E85 wrote: Thu, 18 Mar 2021, 18:42 I would like to use a current transformer to measure the current flowing from the MPPT solar charger into the battery.
Current transformers only work with AC, and the solar charge controller (SCC) is basically DC. You could use a Hall effect device, but you'd have to deal with the drift problem.

Or you could use the shunt resistors that are already there. Or you could query the inverter with suitable commands, and have it tell you what it's already measured for you, with noise filtering, calibration, etc.
I looked more into this last option as I have a Raspberry Pi doing such queries to the inverter already, so I should be able to connect a MCP4725 12-bit DAC converter to emulate the 0-5V output signal that the 123smartBMS current sensors have. That would give ample resolution for a virtual -100A - +100A sensor (0.5V = -100A , 2,5V = 0A and 4.5V = 100A).

For the ”virtual” current sensor to work I would need to have the Raspberry PI ground equal to battery negative battery terminal as the BMS current sensors (signal) ground is also connected to the negative battery terminal.

Unfortunately the ground of the inverter (and thus the ground of the Raspberry PI when connected through a USB cable) doesn’t seem to be floating. I measured the negative battery terminal having a potential of ca half the battery voltage (below) the inverter ground and the positive battery terminal potential seemed equally high above the ground, if that makes any sense ?

EDIT: The red text above is not correct and the battery terminals do seem to be floating after all, as soon as I connect a testresistor between the negative battery terminal and the inverter chassis/ground the potential between these disappears (and there is no measurable current flowing through the resistor).

So this could work after all.
Last edited by E85 on Thu, 25 Mar 2021, 20:02, 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 »

I finally got around to adding buttons to my Node Red battery management system to send commands to my ancient PIP-4048s to change the battery calibration voltage. I also added the PSAVE command, as mentioned at the bottom of the post with all the details: Calibrating your inverter's battery voltage reading . I can now confirm that using the PSAVE command works, and the calibration survives a power down of the inverters. The commands only work for the inverter connected to the BMS via its serial port. In my case, I had to move the serial data cable from one inverter to the other; I have two in parallel.

Most of it I could do in comfort indoors, but of course I had to have my trusty Fluke multimeter out in the heat where the inverters are to verify whether 5 bumps was enough (I settled on 6 bumps). I held the multimeter up near the LC Display so that I could verify that they were reading close to the same all the time, as my "quick pot" in the kitchen changed the load by a little over a kilowatt twice a minute or so.

In case you care, this is what my buttons look like. I wish I could correct the mega volts (MV) to millivolts (mV), but the Node Red buttons refuse to render lower case. I'm pretty sure it can be changed, but I can't be bothered. I've also shown the Javascript code for the Assemble BatCal+ node.

Battery Cal and PSAVE buttons.png
Battery Cal and PSAVE buttons.png (18.1 KiB) Viewed 4496 times
Bat Cal and PSAVE flows.png
Bat Cal and PSAVE flows.png (45.9 KiB) Viewed 4496 times
BatCal+ node.png
BatCal+ node.png (13.22 KiB) Viewed 4496 times
For anyone that wants to copy the code: sorry. It all depends on micro-Forth commands like Pp (for PIP string serial send) that automatically send strings with CRC characters to the connected inverter. I just wanted to give an idea of what it took to implement this on my system. (About 10 minutes work today, though I had done the PSAVE command some time ago, and it took about half an hour with figuring out what I needed to change).

All of the Node Red "code" to implement these 3 buttons was adapted from other areas; Weber wrote 99.8% of it. Thanks, Weber, I use it daily!

Edit: for those that care, the Node Red white-on-grey arrows shoot off to other nodes on another page of the Node Red flows. One enables or disables "autosending of data"; that's the constant sending of QPGS0/1 and other commands to update graphs and logs. The other is "send command to BMU". Those nodes at the other end of the arrows were implemented long ago, and are needed for any settings changes.

[ Edit: "serial port" → "serial data cable". ]
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.
cybersyx
Noobie
Posts: 6
Joined: Mon, 14 May 2018, 23:40

Re: PIP-4048MS inverter

Post by cybersyx »

Hi, i've flash this firmware on my MPP SOLAR PIP4048 with dual SCC but i've a big problem.
The problem is that the batteries remain in bulk and do not float. Does anyone have the original firmware or do they know how I can solve this problem?
Thanks
weber wrote: Thu, 04 Jan 2018, 07:38 Patched Firmware 75.31a for dual and triple SCC models

This patched firmware fixes the premature-float bug in the manufacturer's firmware for the PIP-4048MSD, PIP-4048MST, Axpert Plus Duo 5K, Axpert Plus Tri 5K and equivalent. It does not contain the patches for our dynamic current and load control (DCLC) or AussieView™ additions, and we have not provided a version with our voltage and current threshold changes for LFP cells. This is primarily due to the fact that we don't have a suitable inverter to test them on.

Here are the zip files with all the software you need, to reflash your PIP-4048MSD/MST or Axpert MKS 5K-48 Duo/Tri, and to revert to standard 75.31 firmware if required. See these installation instructions (but ignore the zip files there).

Note: This firmware is only suitable for the 48 V models with dual or triple low-voltage MPPTs and a power factor of 0.8 (4 kW / 5 kVA), not those with a single MPPT or a power factor of 1.0 (5 kW / 5 kVA).

For lead acid (24S), lithium cobalt-blends (LCO, NMC, NCM, NCA) (14S) and lithium titanate (LTO) (21S)
dsp_LC1_75.31a.zip


It should go without saying, that you use this at your own risk. The original unpatched 75.31 is included in the above zip-file, in case there's a problem with the patched version. But you will not be able to revert to whatever version of the firmware is in your inverter now (unless it is 75.31).

Please let us know if you use it successfully.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: PIP-4048MS inverter

Post by weber »

cybersyx wrote: Mon, 29 Mar 2021, 05:47 Hi, i've flash this firmware on my MPP SOLAR PIP4048 with dual SCC but i've a big problem.
The problem is that the batteries remain in bulk and do not float. Does anyone have the original firmware or do they know how I can solve this problem?
Could it just be that you are used to the faulty manufacturer's firmware that goes to float mode prematurely?

With the patched firmware (LC1_75.31a), it will only go to float if the battery voltage has been above the absorb voltage setting [26] minus 0.5 V and the charge current has been below 1/5 of the maximum charge current setting [2] (or below 5 amps if that is greater) for 30 seconds continuously. That's when the absorb-time setting [32] is "Aut". Otherwise the voltage may have to be above setting [26] minus 0.5 V for the number of minutes given by setting [32].

Here's a more detailed description of the charge termination conditions.

Here's how to restore factory firmware if you really need to.
https://forums.aeva.asn.au/viewtopic.php?p=75494#FAQ6 or https://forums.aeva.asn.au/viewtopic.ph ... 349#p80349.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
cybersyx
Noobie
Posts: 6
Joined: Mon, 14 May 2018, 23:40

Re: PIP-4048MS inverter

Post by cybersyx »

Perfect, thank you very much for the correct explanation but in my inverter menu number 32 is not present so I have no way to set the time and therefore the batteries remain in bulk
weber wrote: Mon, 29 Mar 2021, 06:17
cybersyx wrote: Mon, 29 Mar 2021, 05:47 Hi, i've flash this firmware on my MPP SOLAR PIP4048 with dual SCC but i've a big problem.
The problem is that the batteries remain in bulk and do not float. Does anyone have the original firmware or do they know how I can solve this problem?
Could it just be that you are used to the faulty manufacturer's firmware that goes to float mode prematurely?

With the patched firmware (LC1_75.31a), it will only go to float if the battery voltage has been above the absorb voltage setting [26] minus 0.5 V and the charge current has been below 1/5 of the maximum charge current setting [2] (or below 5 amps if that is greater) for 30 seconds continuously. That's when the absorb-time setting [32] is "Aut". Otherwise the voltage may have to be above setting [26] minus 0.5 V for the number of minutes given by setting [32].

Here's a more detailed description of the charge termination conditions.

Here's how to restore factory firmware if you really need to.
https://forums.aeva.asn.au/viewtopic.php?p=75494#FAQ6 or https://forums.aeva.asn.au/viewtopic.ph ... 349#p80349.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: PIP-4048MS inverter

Post by weber »

cybersyx wrote: Tue, 30 Mar 2021, 04:28 Perfect, thank you very much for the correct explanation but in my inverter menu number 32 is not present so I have no way to set the time and therefore the batteries remain in bulk
For us to be able to help you further, please tell us the type and capacity of your battery, the make and model of your solar panels and their configuration, whether you have utility or generator, your inverter settings [2], [5], [11], [16], [26], [27], [29] and any higher numbered settings you may have, your original main CPU firmware version number and your second, third and fourth CPU firmware version numbers.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
cybersyx
Noobie
Posts: 6
Joined: Mon, 14 May 2018, 23:40

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

Post by cybersyx »

Well, 8 AGM battery 200ah (connected in series of four 48V). Settings are:

Charge source priority: Solar Only
Output source: SBU
Battery Type: AGM
Bulk Voltage: 56.4
Float Voltage: 54.2
Max charge current: 40A
Back to Charge voltage: 48
Back to discharge voltage: 54
Main CPU version: 75.31
SCC1 CPU version: 0.00
SCC2 CPU version: 0.00
SCC3 CPU version: 0.00

I've 16 solar panel(250W 24V) connected in series of two. 10 on the first SCC and 6 on the secondo SCC.
Attachments
battery volts.png
battery volts.png (26 KiB) Viewed 4437 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-4048MS and PIP-5048MS inverters

Post by coulomb »

cybersyx wrote: Wed, 31 Mar 2021, 04:14 Max charge current: 40A
I think that's too low. You have a 400Ah battery, two strings of 200Ah modules. So that's 0.1C. Usually, lead acid is charged at 0.15C, 50% faster. If you increased the maximum charge current to 60A, that also means that the terminating charge current goes up from 8A to 12A, which it would reach sooner as the charge current tapers.

At 0.1C, it would take 10 hours to charge the battery from dead flat. Hopefully you only discharge to about 50%, but that's still about 5 hours, longer if loads take away some of the charge current. So your battery might be just about to go to float when you run out of sun.

We'd have to see the battery charge current curve to be sure that this is what's happening.
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.
solamahn
Groupie
Posts: 163
Joined: Thu, 18 Jun 2015, 00:09
Real Name: Julian Leach
Location: PNG

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

Post by solamahn »

I would make sure batteries are 2P4S and not 4S2P. Set battery type to USE, bulk to 58.4, float to 55 - 56 depending on battery temperature, maximum charge to 80A and panels to 3S. I assume the 2 panels arrays are facing different directions so the 3S configuration would require adding 2 more panels to the 10 panel array.
Solamahn PNG
24x300w, 2x4048ms, 75kw AGM
24x280w, JFY6000
12x300w, 4048ms, 20kw Winston
30x280w, 2x4048V, 12kw AGM
9 x 280w, 3024msxe, 10kw CALB
24 x 300w, 5048msd, 20kw Winston
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 »

solamahn wrote: Wed, 31 Mar 2021, 21:55 Set battery type to USE, bulk to 58.4...
You suggest a high voltage. That's 14.6 V per 12 V module, well above the gassing voltage. But I see that my Optima (a fancy AGM auxiliary battery for my Leaf) can be charged to 14.7 V with no current limit. That means that Voltronic's default of 14.1 V per 12 V module for the AGM battery type is very low. Perhaps the "AGM" option for the battery type should really be "GEL".

I have very little lead acid experience, so I'm not arguing. Just curious.
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.
solamahn
Groupie
Posts: 163
Joined: Thu, 18 Jun 2015, 00:09
Real Name: Julian Leach
Location: PNG

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

Post by solamahn »

They are the settings I use with AGM’s. No venting because the charging switches to float at 14v before the battery is full, especially if the battery is not hot. Also I notice that my 3 x 72 cell panels PV in voltage never goes over 120v. Higher PV in voltage seems to make the SCC perform better. Might even be able to use 4S with 60 cell panels.
Solamahn PNG
24x300w, 2x4048ms, 75kw AGM
24x280w, JFY6000
12x300w, 4048ms, 20kw Winston
30x280w, 2x4048V, 12kw AGM
9 x 280w, 3024msxe, 10kw CALB
24 x 300w, 5048msd, 20kw Winston
Post Reply