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.
User avatar
lopezjm2001
Groupie
Posts: 100
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

Re: PIP-4048MS inverter

Post by lopezjm2001 »

weber wrote: Thu, 03 May 2018, 11:46 Alternatively, you could tell them that yes, the problem was apparently caused by updating to main firmware 73.00. But it is fixed now, because you have reflashed with Coulomb and Weber's patched firmware 73.00b. Or if you want to keep your life simple, you could just say it is fixed now because you have reflashed with "other firmware". :)
Thanks for the advice. I have suggested to Peggy from MPPSolar that they should look at your improved firmware and make these improvements to their own firmware. I am assuming it is free software.
2016 BMW i3 94AH
13.08 Batt.Kapa.max 29.7 kWh(13/08/2018)
13.08 Batt.Kapa.max 28.7 kWh(22/06/2020)
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 »

lopezjm2001 wrote: Thu, 03 May 2018, 20:15 Thanks for the advice. I have suggested to Peggy from MPPSolar that they should look at your improved firmware and make these improvements to their own firmware. I am assuming it is free software.
Yes, our patches are free and open-source. Yes, we would prefer that Voltronics would fix the bugs and implement the improvements in their source code for their factory firmware.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
lopezjm2001
Groupie
Posts: 100
Joined: Mon, 02 Feb 2015, 20:37
Real Name: John Lopez
Location: Sydney

Re: PIP-4048MS inverter

Post by lopezjm2001 »

weber wrote: Thu, 03 May 2018, 20:29 Yes, our patches are free and open-source. Yes, we would prefer that Voltronics would fix the bugs and implement the improvements in their source code for their factory firmware.
They might do that.
Hello Sir

Thanks for your reply and suggestion.

We will provide these information to our technical research and development department. They will test and modify the original firmware.

Thank you.

Best regards,
Albert Y. ¦ Support Team, MPP Solar Inc. 
support@mppsolar.com ¦ www.mppsolar.com
2016 BMW i3 94AH
13.08 Batt.Kapa.max 29.7 kWh(13/08/2018)
13.08 Batt.Kapa.max 28.7 kWh(22/06/2020)
simat
Noobie
Posts: 12
Joined: Thu, 15 Feb 2018, 21:16
Real Name: Simon Matthews
Location: Harvey, WA

Re: PIP-4048MS inverter

Post by simat »

weber wrote: Sat, 21 Apr 2018, 19:59 Hi Simon, it is considered to be in absorb stage when it is within 0.5 volts of the absorb voltage setting. You can read more details of the absorb to float conditions here:
viewtopic.php?f=31&t=4332&p=64095#p64095
Scroll down to the green heading "Charge Termination Conditions".
Hi Weber, I have finished my software that talks to the PIP and BMS and finally enough sunshine to fully charge the battery.

I have logged three occurrences of the battery going to float. Running your patched software 73.00b with "Bulk charging time" (option 32) set to 450 minutes gives the following
The PIP consistently changes from state 11-12 (Bulk to absorb) when the voltage is 0.6V less that the Bulk CV set voltage.
The PIP changes from state 12-13 (absorb to float) when the PIP "Battery charging current" was zero for a period of 30 seconds. Absorb times were 33 minutes, 60 minutes and 21 minutes and total bulk time was well under the programmed 450 minutes. "Real" Battery Current when PIP switched to float were -0.1A, -4.5A and 1.1A (minus is current into battery).

As far as I can see, this matches the "Charge termination conditions" you mention in your post. If using a timed absorb period rather than automatic absorb terminated by charge current I would prefer it if the absorb period was not terminated by charge current but by time alone. The reason I want a long absorb time is to balance an LFP battery. It is not a big deal if I can achieve the same result by raising the float voltage high enough to achieve the same result.

Does any one know if there is a serial command to set the absorb time?

I have found that you can set the PIP back to bulk charging from float by raising the the "battery re-charge voltage" above the current battery voltage via the PBCV command. This only seems to work when the "Bulk charging time" is set to "Aut". With this in mind it would not be hard to set up my battery monitoring software to act as a watchdog to get around the premature float bug for all those with Voltronic/PIP inverters that can't use your patched software.

I have programmed my battery monitor to issue commands to the PIP to drop the charge voltage in the event that any of the individual cells goes out of its safe operating voltage range.

Simon
Off-Grid with LFP (LiFePO4) battery
15x180Ah CALB cells 15s, 2xMPPSolar MS5048s, ~5.5kW solar panels
System controller and demand manager https://github.com/simat/BatteryMonitor
J1772 EV charger https://github.com/simat/Variable-J1772-Charger
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 inverter

Post by coulomb »

simat wrote: Sat, 05 May 2018, 19:02 As far as I can see, this matches the "Charge termination conditions" you mention in your post. If using a timed absorb period rather than automatic absorb terminated by charge current I would prefer it if the absorb period was not terminated by charge current but by time alone.
This is the original behaviour. When a non-zero absorb time is set (setting 32 is non-zero), the threshold becomes the number of paralleled machines times two, i.e. 2 A per machine. I've changed the description in the Charge Termination Conditions to reflect that fact that achieving this isn't "practically impossible" (thanks for the data).
Does any one know if there is a serial command to set the absorb time?
Yes, there is a PCVTnnn command, where nnn is the number of minutes. It must be a multiple of 5 and no more than 900. This will change setting 32.
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.
al76
Noobie
Posts: 14
Joined: Sun, 14 Jan 2018, 11:29

Re: PIP-4048MS inverter

Post by al76 »

weber wrote: Wed, 02 May 2018, 10:08
al76 wrote: Wed, 02 May 2018, 06:08 Anyway I have the replacement pip installed (day 2) and its had issues each morning.
I log output from QPIGS and soon after sunrise I see a brief spike from the panels then nothing reported from panels. The battery volts reported by QPIGS seems to indicate charging is happening but the pip display also shows that no solar charging is happening. After a cold start ( disconnect panels, turn off unit, disconnect batteries, wait a bit, reconnect batteroes, turn on unit connect panels) the solar charging seems to report correctly via QPIGS and the panel display.
Anyone offer any clues?
I understand you are using 72-cell panels. Is your PV array presently configured as 3 strings of 2 panels or 2 strings of 3 panels? As mentioned previously, strings of 3 may produce excessive open circuit voltage when cold.

You know about the premature-float bug right? The solution to that is to update to the latest patched firmware. See the index post.
pip has been working fine since I posted.
I have panels configured as 3strings of 2 panels. I'm reading up on the premature float bug I suspect this may be an issue.
al76
Noobie
Posts: 14
Joined: Sun, 14 Jan 2018, 11:29

Re: PIP-4048MS inverter

Post by al76 »

I have a pip4048 battery config question
I have 8 x 12v RITAH RA12-120SD configured as 4S2P for a 48v 240Ah system

The battery spec sheet
https://www.rpc.com.au/pdf/ritar_12V_120Ah.pdf

states
Float charging Voltage 13.6 to 13.8 VDC/unit Average at 25 degC
and
Equalization and Cycle Service 14.6 to 14.8 VDC/unit Average at 25 degC


Is using AGM setting on the pip4048 sufficient or can I do better with a USER setting?


also I believe my pip suffers from the premature float bug
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 »

Release Version of Patched Firmware 73.00c for some PF0.8 models


[Edit: This has been superseded. Please see the Firmware section of the first post of this topic to find the latest patched firmware for your machine.]
[Edit: Was the beta version. Between beta and release, only the first character of the version number changed (from B to L).]

This is the third version of our patched firmware based on factory firmware version 73.00 for the PIP-4048MS and equivalents. This patched firmware has all the same patches as 73.00b and earlier patched firmwares, including Dynamic Current Control and AussieView™. This includes fixing the infamous premature float bugs.

In addition, 73.00c makes the following improvements/bug fixes:

1. We fixed the bug where, after changing a setting once, you could not change it again without first pressing the up and down buttons. We also fixed the bug where, while changing a setting, you could not go more than 135 steps up or 120 steps down. This affected the Absorb time setting [32]. These were pointed out by user OomD on the South African Power Forum.

2. In versions 72.70c and 73.00a/b we attempted unsuccessfully to improve the accuracy of low current readings by the simple expedient of subtracting 1 amp to account for losses. But forum-contributor @sinux, pointed out that the losses during AC Charging are much less than 1 amp and so we had in fact made AC charge current readings less accurate.

In this version (73.00c) we have improved the accuracy of all charge and discharge current readings, by using more complicated formulae. A detailed description can be found near the end of the Dynamic Current Control manual, as linked above. These changes do not affect the relationship between requested currents and actual currents, only between actual currents and displayed or reported currents.


Note: This firmware is only suitable for the 48 V models with a single low-voltage MPPT and a power factor of 0.8 (4 kW / 5 kVA), not those with dual or triple MPPTs or a power factor of 1.0 (5 kW / 5 kVA), and not those with a maximum PV array open circuit voltage greater than 145 V DC. Parallel or phased machines must all run the same patched firmware version.

For lithium ferrous phosphate (LFP) (16S or 15S)
dsp_LF1_73.00c.zip
(1.54 MiB) Downloaded 471 times

For lead acid (24S), lithium cobalt-blends (LCO, NMC, NCM, NCA) (14S), and lithium titanate (LTO) (21S)
dsp_LC1_73.00c.zip
(1.53 MiB) Downloaded 1595 times

For reflashing instructions see the 72.70b reflashing instructions, but ignore the zip files there.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
gremlinman
Noobie
Posts: 11
Joined: Wed, 02 May 2018, 09:37

Re: PIP-4048MS inverter

Post by gremlinman »

Is it normal for the PIP to not have a float or bulk/absorb voltage ? I notice like right now, the PIP is indicating the batteries are "full" and the voltage reading is 52.4, 600w of solar is incoming and inverter is on but not being used. Shouldn't the PIP be attempting to be at 54v or 56v depending upon which mode it is in?
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 inverter

Post by coulomb »

gremlinman wrote: Tue, 08 May 2018, 10:50 Is it normal for the PIP to not have a float or bulk/absorb voltage ?
No. It should be aiming for either float or absorb voltage, depending on the mode. Check the charge LED (middle, green LED) to see if it is not charging at all (dark), bulk/absorb (flashing), or at float (solid green).
Shouldn't the PIP be attempting to be at 54v or 56v depending upon which mode it is in?
The exact voltages depend on settings. The defaults, set for lead acid batteries, are 54.0 and 56.4 for float and absorb. Lithium batteries need different voltages, depending on the number of cells, their chemistry, and a little on the whim of the operator.

You might be seeing one of the infamous premature float bugs, where a passing cloud causes the standard factory firmware to believe that because the battery is drawing little current, it is full.

Of course, it will take a while for 600 W of solar to charge a large battery. Do you have more solar power available?
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.
xenonhost
Noobie
Posts: 10
Joined: Sun, 23 Apr 2017, 00:19
Real Name: Marian Titieni
Location: Romania
Contact:

Re: PIP-4048MS inverter

Post by xenonhost »

Hi!
Is there a way to disable battery charging for 4 pcs from 5 parallel connected inverters? I want to use the batteries just to power up the inverters and use just solar power in the day and utility when solar not available.
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 inverter

Post by coulomb »

xenonhost wrote: Tue, 08 May 2018, 12:35 Is there a way to disable battery charging for 4 pcs from 5 parallel connected inverters?
Sure, don't connect solar to the four, and they won't charge. I don't know why you'd want to do this, however; why not spread the charging load?

Charging current limits are separate for paralleled inverters. You can use the LC display and buttons, or WatchPower, or other monitoring software to set the current limits for each inverter. Sorry, my earlier answer to a question about this was wrong; now corrected.
I want to use the batteries just to power up the inverters and use just solar power in the day and utility when solar not available.
Ok, but it's not necessary to restrict charging to one inverter-charger unit to achieve this.
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.
xenonhost
Noobie
Posts: 10
Joined: Sun, 23 Apr 2017, 00:19
Real Name: Marian Titieni
Location: Romania
Contact:

Re: PIP-4048MS inverter

Post by xenonhost »

But I have only 40Ah remaining cells
gremlinman
Noobie
Posts: 11
Joined: Wed, 02 May 2018, 09:37

Re: PIP-4048MS inverter

Post by gremlinman »

coulomb wrote: Tue, 08 May 2018, 11:26
gremlinman wrote: Tue, 08 May 2018, 10:50 Is it normal for the PIP to not have a float or bulk/absorb voltage ?
No. It should be aiming for either float or absorb voltage, depending on the mode. Check the charge LED (middle, green LED) to see if it is not charging at all (dark), bulk/absorb (flashing), or at float (solid green).
Shouldn't the PIP be attempting to be at 54v or 56v depending upon which mode it is in?
The exact voltages depend on settings. The defaults, set for lead acid batteries, are 54.0 and 56.4 for float and absorb. Lithium batteries need different voltages, depending on the number of cells, their chemistry, and a little on the whim of the operator.

You might be seeing one of the infamous premature float bugs, where a passing cloud causes the standard factory firmware to believe that because the battery is drawing little current, it is full.

Of course, it will take a while for 600 W of solar to charge a large battery. Do you have more solar power available?
I am on 73.00b though. I have 8 x 12v 120ah batteries. The led was flashing.

It is just strange that I thought the PIP should be attempting 56v regardless of amps going into it, I thought it would attempt to hold 56v whilst in bulk/absorb. I notice when I charge with generator it usually goes up from ~50v to 56v and stays at 56v for a while before hitting 54v and staying there putting 1a into the batteries from the generator. Am I wrong in assuming when in bulk/absorb mode the voltage should be 56v ?
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 inverter

Post by coulomb »

gremlinman wrote: Tue, 08 May 2018, 15:31 I am on 73.00b though. I have 8 x 12v 120ah batteries. The led was flashing.
Ah, I failed to remember that you were on 73.00b. Flashing LED, so you are in bulk or absorb. All good.
I thought it would attempt to hold 56v whilst in bulk/absorb.
Sure, it will have 56.4 V as its setpoint. It can't get there straight away, though.
I notice when I charge with generator it usually goes up from ~50v to 56v
Your generator charging must be at a fairly high rate then; a lot more than 600 W of charge power. That will get the internal battery voltage up higher sooner, but will also raise the battery terminal voltage due to internal resistance.
and stays at 56v for a while before hitting 54v and staying there putting 1a into the batteries from the generator.
Right. The time it's at 54 V and ~1 A it would be in float mode.
Am I wrong in assuming when in bulk/absorb mode the voltage should be 56v ?
In bulk mode, assuming a constant charge current, and ignoring internal resistance for a moment, the battery voltage should ramp up roughly linearly (assuming lead acid batteries) from the initial voltage till the 56.4 V setpoint, at which point it changes to absorb mode. It should stay in absorb mode until the battery current drops low, and/or the absorb time has expired (if the absorb timer is used). The battery voltage should be constant at 56.4 V during the absorb stage.

Taking into account internal resistance, the battery voltage would be higher than the figures indicated above, by perhaps half a volt to over a volt, depending on many factors. The bulk stage will still finish at or near the 56.4 V setting, however. As the charge current reduces during absorb mode, the extra voltage due to internal resistance times charge current will reduce to near enough to zero.

So for the generator to have the voltage shoot up to around 56 V nearly straight away, you really must be overcharging the lead acid batteries. Here I mean charging at too high a current, not to too high a state of charge. They should only be charged at a rate of some 13% of their Ah rating. For example, a 200 Ah lead acid battery should only be charged at about 0.13 x 200 = 26 A. Another way to look at is that it should never be charged faster than 7.7 hours (1/0.13), from empty to full. But of course, lead acid should never be discharged more than about 20%, so more realistically, it should take at least 0.2 x 7.7 = 1.5 hours to take it from 80% to 100% SOC. Don't rely on the PIP's SOC reading for state of charge; it's extremely approximate for lead acid batteries, and completely useless for other chemistries.

Now with the variability of solar charging, the battery voltage will be all over the place, but you should be able to discern a rough ramp from initial voltage to absorb setting voltage during the bulk stage, then a rough plateau near 56.4 V during the absorb stage, followed by a stay at around 54 V after that when it stays in the float stage (unless the battery voltage drops a lot, triggering another bulk/absorb stage pair). Unfortunately, the only way to distinguish between bulk and absorb modes is from the last field of the undocumented Q1 command, though a few monitoring systems display it. But you can't tell from the front panel.

[ Edit: added "all good". ]
[ Edit: added "The battery voltage should be constant at 56.4 V during the absorb stage." ]
[ Edit: "higher than the figure indicated" -> "higher than the figures indicated above". ]
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-4048MS inverter

Post by coulomb »

xenonhost wrote: Tue, 08 May 2018, 15:19 But I have only 40Ah remaining cells
Ah. So that's why you only want one unit charging.

I believe that even with no solar connected, you still can't set the maximum total charge (setting) to less than 10 A, so they will still contribute towards the total maximum charge current for the system. This may be more than you want. I think you will need to turn off some of the paralleled machines; I assume that if they are turned off (using the rocker switch under the units), they won't contribute to the total charge current total (for all machines). With only 40 Ah of battery, you can't be using the power of 5 paralleled machines anyway. I'd use only one, perhaps two at most.
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.
xenonhost
Noobie
Posts: 10
Joined: Sun, 23 Apr 2017, 00:19
Real Name: Marian Titieni
Location: Romania
Contact:

Re: PIP-4048MS inverter

Post by xenonhost »

but I saw in the last firmware update that I can set maximum charging from zero, is it not true?
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 inverter

Post by coulomb »

xenonhost wrote: Wed, 09 May 2018, 00:26 but I saw in the last firmware update that I can set maximum charging from zero, is it not true?
Sigh. I'd forgotten that patch; it's one of the earlier ones. So yes, you could use that instead of turning the machine off.

I don't know if the firmware will "idle" some inverters when more paralleled machines are available than needed to fulfil the present load. If not, each paralleled machine is some 50 W each of extra loss that you likely can't use with a 40 Ah battery. That's over 200 W for 4 unused machines. Switching them off will remove this load, whereas setting their maximum charge current to zero will not.
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.
gremlinman
Noobie
Posts: 11
Joined: Wed, 02 May 2018, 09:37

Re: PIP-4048MS inverter

Post by gremlinman »

coulomb wrote: Tue, 08 May 2018, 18:16
gremlinman wrote: Tue, 08 May 2018, 15:31 I am on 73.00b though. I have 8 x 12v 120ah batteries. The led was flashing.
Ah, I failed to remember that you were on 73.00b. Flashing LED, so you are in bulk or absorb. All good.
I thought it would attempt to hold 56v whilst in bulk/absorb.
Sure, it will have 56.4 V as its setpoint. It can't get there straight away, though.
I notice when I charge with generator it usually goes up from ~50v to 56v
Your generator charging must be at a fairly high rate then; a lot more than 600 W of charge power. That will get the internal battery voltage up higher sooner, but will also raise the battery terminal voltage due to internal resistance.
and stays at 56v for a while before hitting 54v and staying there putting 1a into the batteries from the generator.
Right. The time it's at 54 V and ~1 A it would be in float mode.
Am I wrong in assuming when in bulk/absorb mode the voltage should be 56v ?
In bulk mode, assuming a constant charge current, and ignoring internal resistance for a moment, the battery voltage should ramp up roughly linearly (assuming lead acid batteries) from the initial voltage till the 56.4 V setpoint, at which point it changes to absorb mode. It should stay in absorb mode until the battery current drops low, and/or the absorb time has expired (if the absorb timer is used). The battery voltage should be constant at 56.4 V during the absorb stage. Taking into account internal resistance, the battery voltage would be higher than the figure indicated, by perhaps half a volt to over a volt, depending on many factors. Bulk stage will still finish at or near the 56.4 V setting, however. As the charge current reduces during absorb mode, the extra voltage due to internal resistance times charge current will reduce to near enough to zero.

So for the generator to have the voltage shoot up to around 56 V nearly straight away, you really must be overcharging the lead acid batteries. Here I mean charging at too high a current, not to too high a state of charge. They should only be charged at a rate of some 13% of their Ah rating. For example, a 200 Ah lead acid battery should only be charged at about 0.13 x 200 = 26 A. Another way to look at is that it should never be charged faster than 7.7 hours (1/0.13), from empty to full. But of course, lead acid should never be discharged more than about 20%, so more realistically, it should take at least 0.2 x 7.7 = 1.5 hours to take it from 80% to 100% SOC. Don't rely on the PIP's SOC reading for state of charge; it's extremely approximate for lead acid batteries, and completely useless for other chemistries.

Now with the variability of solar charging, the battery voltage will be all over the place, but you should be able to discern a rough ramp from initial voltage to absorb setting voltage during the bulk stage, then a rough plateau near 56.4 V during the absorb stage, followed by a stay at around 54 V after that when it stays in the float stage (unless the battery voltage drops a lot, triggering another bulk/absorb stage pair). Unfortunately, the only way to distinguish between bulk and absorb modes is from the last field of the undocumented Q1 command, though a few monitoring systems display it. But you can't tell from the front panel.

[ Edit: added "all good". ]
[ Edit: added "The battery voltage should be constant at 56.4 V during the absorb stage." ]
Thanks for the info, so in "bulk" mode you will see a rise to 56v and then it is in absorb. I guess the thing which seems weird to me is aren't the batteries full at ~52v for AGM . In bulk mode is the actual batteries voltage still lower than 52v assuming no load? Do you have any links to the technical workings of charging lead acid batteries because I can only seem to find casual explanations for this. Why is there the linear voltage ramp in bulk mode instead of just pumping 56v into 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-4048MS inverter

Post by coulomb »

gremlinman wrote: Wed, 09 May 2018, 11:12 so in "bulk" mode you will see a rise to 56v and then it is in absorb.
Yes.
I guess the thing which seems weird to me is aren't the batteries full at ~52v for AGM .
Battery terminal voltages are a bit weird. 52 V (13.0 V per 12 V module, or 2.167 V per 2 V cell) is indeed a good 100% SOC voltage, but that's a resting voltage. You only see this voltage after several hours of no charge or discharge. Some AGMs may only rest at a lower voltage when full; it depends on many factors including the battery temperature. As soon as you start charging even a little, the terminal voltage shoots up to around 54 V; that's why the float voltage setting is at this voltage. Only a very small current flows into the battery at this voltage. Note that the voltage is a result of the current; to achieve float charge, the charger puts in a small current, say 1 A, and doesn't increase the current any more when the battery reaches 54 V. If there is a load on the battery of 10 A, the charger finds it has to push 11 A into the battery to maintain 54 V. Ten of those amps flow to the load, and the one amp keeps the battery at 54 V instead of 52 V.
In bulk mode is the actual batteries voltage still lower than 52v assuming no load?
No. In bulk mode, the battery terminal voltage rises from 52 V to around 56 V (I have been ignoring the fact that the charger will stop about 0.5 V short of the setting voltage).
Why is there the linear voltage ramp in bulk mode instead of just pumping 56v into it?
You don't "pump in" a voltage; you pump current (amperes, usually called amps) into a battery. You can think of the battery cables as "pipes" carrying electrons instead of water or gas. It's not quite accurate, but near enough. So many electrons have an electric charge called a coulomb; one coulomb of charge per second is an amp. The transfer of that charge to the battery represents energy; this energy causes chemical changes which result in the ideal battery changing its voltage. So bulk mode for the charger is the same as absorb mode; it keeps increasing the current until either it hits its limit or the battery voltage reaches the target. Float mode is much the same, with a different target voltage.
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.
gremlinman
Noobie
Posts: 11
Joined: Wed, 02 May 2018, 09:37

Re: PIP-4048MS inverter

Post by gremlinman »

coulomb wrote: Wed, 09 May 2018, 16:55
gremlinman wrote: Wed, 09 May 2018, 11:12 so in "bulk" mode you will see a rise to 56v and then it is in absorb.
Yes.
I guess the thing which seems weird to me is aren't the batteries full at ~52v for AGM .
Battery terminal voltages are a bit weird. 52 V (13.0 V per 12 V module, or 2.167 V per 2 V cell) is indeed a good 100% SOC voltage, but that's a resting voltage. You only see this voltage after several hours of no charge or discharge. Some AGMs may only rest at a lower voltage when full; it depends on many factors including the battery temperature. As soon as you start charging even a little, the terminal voltage shoots up to around 54 V; that's why the float voltage setting is at this voltage. Only a very small current flows into the battery at this voltage. Note that the voltage is a result of the current; to achieve float charge, the charger puts in a small current, say 1 A, and doesn't increase the current any more when the battery reaches 54 V. If there is a load on the battery of 10 A, the charger finds it has to push 11 A into the battery to maintain 54 V. Ten of those amps flow to the load, and the one amp keeps the battery at 54 V instead of 52 V.
In bulk mode is the actual batteries voltage still lower than 52v assuming no load?
No. In bulk mode, the battery terminal voltage rises from 52 V to around 56 V (I have been ignoring the fact that the charger will stop about 0.5 V short of the setting voltage).
Why is there the linear voltage ramp in bulk mode instead of just pumping 56v into it?
You don't "pump in" a voltage; you pump current (amperes, usually called amps) into a battery. You can think of the battery cables as "pipes" carrying electrons instead of water or gas. It's not quite accurate, but near enough. So many electrons have an electric charge called a coulomb; one coulomb of charge per second is an amp. The transfer of that charge to the battery represents energy; this energy causes chemical changes which result in the ideal battery changing its voltage. So bulk mode for the charger is the same as absorb mode; it keeps increasing the current until either it hits its limit or the battery voltage reaches the target. Float mode is much the same, with a different target voltage.
That is some good info, thanks for writing it up. I have one more question, when in bulk mode it goes up to 56v from a lower voltage, so it crosses the float voltage setting. Why doesn't it think when it hits ~54v in bulk mode that the batteries aren't full at this stage? How does the charger know the state of the battery and what it should be doing when it seems quite unreliable? Is it related to the internal resistance of the battery or something like this? ie When the batteries are full it gets harder and harder to put energy into it and the charger can measure this?

How accurate can a lead battery SOC be reported?
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 inverter

Post by coulomb »

Edit: Please trim your quotes to just what is needed to establish context. For two posts, you quoted my entire post, including quotes from your questions, when only a sentence or two was needed. It makes it easier to other readers to find information that way. Thanks.
gremlinman wrote: Thu, 10 May 2018, 11:16 ... when in bulk mode it goes up to 56v from a lower voltage, so it crosses the float voltage setting. Why doesn't it think when it hits ~54v in bulk mode that the batteries aren't full at this stage?
When in bulk or absorb modes, the 54 V float setting should be completely ignored; it's aiming for the absorb voltage setting. However, it happens that this is part of the problem with one of the original firmwares' premature float bugs. Although aiming for the absorb setting voltage, at any time when the current is low enough, the firmware checks to see whether it is "eligible to go to float". In this eligibility check, the battery voltage is compared with the setting, and it has to be within half a volt of this setting for a certain period of time. Problem is, there is a typo in their firmware, and they compare against the lower of the two settings (i.e. the float and absorb settings), not the higher of the two as they should. So with original firmware, if the float setting is 54.0 V, you could go to float at 53.5 V if there is a passing cloud for long enough.
How does the charger know the state of the battery and what it should be doing when it seems quite unreliable?
For charging, it only has to know when the battery is full. Usually, it uses the battery current at a certain voltage, but you can use setting 38 (absorb time) to modify that somewhat. With the patched firmware, and I expect the same from original firmware if they ever fix that typo, charging seems quite reliable to me.

If you mean that the state of charge indication is unreliable, then I agree. But it only has to know when the battery is full to stop charging; at other times, it simply doesn't need to know the state of charge.
Is it related to the internal resistance of the battery or something like this?
It doesn't measure the internal resistance.
ie When the batteries are full it gets harder and harder to put energy into it and the charger can measure this?
Yes. This is how the charging algorithm decides when the battery is full. This difficulty of adding more energy is because the voltage of the internal, ideal battery ("inside the real battery", if you like) rises to the point where the charger is forced to reduce the current to a low value to maintain the battery voltage near the absorb voltage setting. In fact, the internal resistance of a fully charged battery is higher than when it is still being charged. That's why a small current into or out of the battery causes a larger voltage change. Once the battery is discharged a little, the same current change will cause a smaller voltage change, so the internal resistance is lower. If the battery voltage ever gets very low, around 20% SOC for lead acid (which should be a very rare occurrence for good life), the internal resistance rises again.
How accurate can a lead battery SOC be reported?
It's quite tricky to get a good reading without measuring the specific gravity of the acid in the battery; this is a chore that seriously large (>= 1000 Ah) flooded lead acid batteries need. But since the battery usually operates only between 100% SOC and about 80% SOC, if you count coulombs (mathematically integrate the battery charge over time), with a few adjustments, it turns out you can get a quite useful indication of "fullness". The "adjustments" include taking into account the Peukert effect, which says that for lead acid especially, a battery loses capacity faster the faster you take energy out of it (i.e. the higher the discharge current).

State Of Charge is about how much energy there is till empty, but you typically don't want to get near empty anyway. To integrate the battery current, you need to have an accurate measure of the battery current. Unfortunately, PIP inverter chargers don't have such a measurement. Reported battery current is inferred (guessed, but it's a somewhat educated guess) from two things that are actually measured: the AC output power, and the AC input power. Some of the recent patches to the patched firmware are to improve this inference, but it's still not good enough for tracking battery capacity. So unfortunately, PIPs as they are now will never be able to track battery capacity accurately, no matter how fancy the firmware. External battery monitoring devices such as these are designed to perform this task properly. They come with a current shunt for accurate battery current measurement, and perform the "adjustments" I mentioned automatically, after you provide them with a few settings.

Some inverter/charger monitoring software, such as ICC, will work with certain brands of battery monitors to display and record accurate SOC, and allow you to perform certain changes when SOC reaches a certain level.

[ Edit: added sentence about battery monitors coming with shunts, and performing adjustments. ]
[ Edit: pH -> specific gravity ]
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.
beccsjb
Noobie
Posts: 3
Joined: Fri, 11 May 2018, 05:53
Real Name: Justin Burnett

Re: PIP-4048MS inverter

Post by beccsjb »

Great reading that has been :)

***Coulomb, do you know if your custom firmware would work on a 'POWER MPS-5K, Model: MPS 5K-48VDC'.***

These inverters are identical to the Axpert. Purchased from Ebay from a company called SWI Power i believe (they claim to manufacture them).

Currently running 3 x in parallel with control from ICC running on Raspberry Pi 3+, though having some stability issues. Feel the existing firmware just doesnt give you enough fine tuning on the low and cut off volts to properly be able to run a LiPoe4 battery (Pylontech).
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 inverter

Post by coulomb »

beccsjb wrote: Fri, 11 May 2018, 06:03 Coulomb, do you know if your custom firmware would work on a 'POWER MPS-5K, Model: MPS 5K-48VDC'.
It's difficult to know.
These inverters are identical to the Axpert.
As long as they are genuinely Voltronic Power made, it should work. Do you have a link to the Ebay advertisement? I could not find 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.
simat
Noobie
Posts: 12
Joined: Thu, 15 Feb 2018, 21:16
Real Name: Simon Matthews
Location: Harvey, WA

Re: PIP-4048MS inverter

Post by simat »

If you want to make a cheap accurate SOC meter and have some electronic skills you can use the TI ADS1115 16 bit Analog to Digital converter.

Here is the circuit for a simple battery monitor using this chip
Image

There is a daughter board based around the ADS1115 for the RaspberryPi boards on ebay. There are also a number of small PCBs based around the ADS1115, just google ADS1115 pcb.

Adafruit has kindly written some python open source drivers if you want to write your own SOC meter software or you could use my software

Simon
Off-Grid with LFP (LiFePO4) battery
15x180Ah CALB cells 15s, 2xMPPSolar MS5048s, ~5.5kW solar panels
System controller and demand manager https://github.com/simat/BatteryMonitor
J1772 EV charger https://github.com/simat/Variable-J1772-Charger
Post Reply