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

PIP-4048MS and PIP-5048MS inverters

Post by coulomb »

MODERATOR NOTES
This thread is not for repairs or hardware modifications, but is about other aspects of the 5 kVA 48 Vdc 230 Vac standalone inverter/charger with max 145 V MPPT and non-removable display, designed (in Taiwan) and manufactured (in China) by Voltronic Power, and sold under various names and model numbers.

Other model numbers:
  • PIP-4048MS and PIP-5048MS (from MPPSolar)
    Axpert MKS 5K
    Mecer (rhymes with "spacer") MKS 5K (from Mustek)
    Mecer SOL-I-AX-5M, SOL-I-AX-5P and SOL-I-AX-5P64
    IPS-4000WM (from Giant Power)
    Effekta AX-M 5 kVA
    Flinslim MPPT 5kVA
    OPTI-Solar SP5000 Brilliant
    MAXIC5000-48
    Pro Power 48V 4000W Integrated Power System
    SP5000-48MPPT80
    Proline Energy inverter 5kVA 48V w/MPPT SCC
    Inverex MKS 5KVA
    SOLARIX PLI 5000-48 (from Steca Electronik GMBH)
    Voltex MKS 5KVA
    RCT Axpert 5000VA
    SolarPower24 Primo MKS 5K-48, Primo MKS(PF1) 5K-48, Primo MKS(PF1) 5K-48H

There are other topics for the
PIP-5048MK inverter (also called the Axpert King 5K), and for the
PIP-5048GE/GK/MG/MGX/MAX inverters (also called the Axpert VM II 5000-48, Axpert VM III 5000-48, Axpert MKS II 5K, Axpert MKS III 5K, and Axpert MAX 7.2 or 8.0), and for
PIP inverter repairs and hardware modifications, and for the
Voltronic InfiniSolar inverters.
Please use them when relevant. We're trying to keep this topic to a reasonable size. It's already over 100 pages (2500 posts).


Index

Frequently asked questions (FAQ)

Documentation

Firmware Hardware
Serial Communications boards
Commands
Performance / problems
Black monoliths
Monitoring software
Off-topic discussions
Other

[ Edited Coulomb 1/May/2020: changed most internal links to https:// (except off-topic section). Hopefully works around the "remember me" problem. ]

END MODERATOR NOTES
====================================


[Post by coulomb]

OffgridQLD found the following 4000 W combination inverter/charger/MPPT from MPP Solar, as he describes here:

http://forums.aeva.asn.au/viewtopic.php?title=new-to-the-forum-new-imiev&p=52606&t=3714#p52606

Image

Weber decided to buy one of these for an offgrid solar job (now known as Black Monolith #1). One of the problems, as OffgridQLD has noted, is that despite advertising to the contrary, it doesn't charge LiFePO4 batteries properly, and also doesn't support a BMS properly. Most of this would be solved if we were able to change some charge parameters on the fly. For example, to stop it charging when a single cell starts getting too high in voltage, we could increase the Bulk and Float voltage settings. Hopefully, we can get the last Cell Management Unit (CMU) to send a short command to the unit to effect the change.

The unit comes with PC software to talk to it over a serial port. We thought we'd try and sniff the commands from the PC software to the unit when the voltage was adjusted. There is a password associated with the PC software, presumably to prevent unauthorised or accidental changes, and we didn't know how this would affect things.

The first surprise was that the serial communications are at 2400 bps; we were expecting 19.2 kbps because there is a setting (for ModBus) that mentioned this figure. Perhaps it has cheap opto-isolators, like the Elcon/TC chargers.

We found mostly ascii commands from the PC software to the unit, e.g.
QSIDx
QMODIx

The "x" characters above were actually graphic characters. Aha! These could be checksums. But when we tried to figure out what the checksum algorithm was, it made no sense. On closer looking, there were actually two random-looking characters at the end of each line, before a carriage return to terminate it. In the first example above there was an invisible control character and in the second example the "I" was apparently part of the checksum. Oh-oh: these could be 16-bit CRCs (Cyclic Redundancy Check bytes).

Well, fortunately Weber and I had "cracked" the CRC algorithm for some Nissan Leaf CAN bus messages. In fact, some Leaf enthusiasts referred to our post on this as the "golden post" that enabled future work to proceed: http://www.mynissanleaf.com/viewtopic.php?f=44&t=18120&start=10#p392323 (original post here). All we had to do was dig up the details, modify the algorithm for 16-bit CRCs (the Leaf uses 8-bit CRCs)...

To cut a long story short... we found it! The CRC polynomial is 0x1021, using left shifts, and processing the bytes in the order transmitted (not reversed). As with the Leaf CRC, there is no initial or final XOR value. We used a spreadsheet to confirm the polynomial, using a bitwise XOR function written in Visual Basic, "iterating" by repeating lines on the spreadsheet (80 rows with the same formulae). We had a huge amount of help from a paper by Greg Ewing of New Zealand.

The commands are as follows:
PCVV56.5<crc1><crc2><cr>   Set the bulk charge voltage to 56.5 V
PBFT54.1<crc1><crc2><cr>   Set the float voltage to 54.1 V
PSDV42.1<crc1><crc2><cr>   Set the inverter low voltage cutoff to 42.1 V

What about the password? It turns out that the password seems only to be used to authenticate the user to the PC software, not to the unit itself. (Though there are some mysterious commands that come across at about the time that the password is entered). In any case, the unit doesn't seem to care whether the password is entered or not; it still responds to the same commands to change the setpoints.

So it looks good for getting this unit to respond appropriately to commands from our battery management system, and Weber may not even need a battery management master unit (i.e. the cell-top units should be enough to do the battery management on their own).

This may also be useful for people like OffgridQLD, using a simple BMS. A simple Arduino or similar may be able to detect the end of bulk charging, and change the bulk/absorb voltage setting to be the same as the float setting.                    
Attachments
SMV III 5K 232 Axpert KS&MKS&V&KING RS232 Protocol 20200102.pdf
Explains model codes from QGMN command
(1.21 MiB) Downloaded 1030 times
ForumEA_B_mppt-1.pdf
(98.97 KiB) Downloaded 2971 times
Communication Protocol-20150924 - Customer.pdf
Communications Protocol 20150924
(600.41 KiB) Downloaded 4370 times
Last edited by weber on Tue, 07 Nov 2017, 14:12, edited 2 times in total.
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
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

Thanks for sharing your findings. It sure didn't take you guys long to crack into the unit. A skill set that I am a little envious of.

So at this stage,

PCVV56.5<crc1><crc2><cr>   Set the bulk charge voltage to 56.5 V
PBFT54.1<crc1><crc2><cr>   Set the float voltage to 54.1 V
PSDV42.1<crc1><crc2><cr>   Set the inverter low voltage cutoff to 42.1 V

So I am assuming you can send the same commands using different (voltage)numbers and the PIP4048 would respond appropriately? I guess this is what happens when you use the (user) tab under battery type. bulk and float Voltage is adjustable so It sends the above commands with your new voltage numbers.

So the thing we don't have user control over is the timers.

Given the absorb /constant voltage stage of the charge is preset to use a timer. The timer or (time in absorb) is based on how long the battery's took from start of bulk to reach absorb voltage (56.5v, or what ever you set it to) That recorded time lets say in this example it was 2hrs would then be used to set the time that the battery will be held at 56.5v but with a upper and lower limit of Min 10 mins min & 8hrs max.

I was wondering if the Min & max time limits could be changed? Setting the max to 0 mins (or perhaps a user defined short period). I guess there is a timer in the software the counts the time to reach absorb voltage if that could be tricked into registering a shot time so that we always trigger a 10min absorb

Perhaps going about it that way is more difficult and of no advantage.

I take it you can only sniff out commands that you can trigger from the software and sniff as it's being sent to the pip4048 and the timers wouldn't be taking that path.

Excuse my primitive understanding of device hacking.


On a side note, A member using a smaller version of the unit (3kw) has reported that he has had a issue with the unit playing up when on float. His float voltage was set at 54v and the unit has been holding float for a short time then suddenly allowing the float voltage to increase. Left to its own it reaches 63v! then trips out on (over voltage) . Turning the unit off then on again results in a brief period in absorb 54v before the voltage climbs again to 63v. It could just be a dud unit but then another member of the form came on saying he had a MPPT charge controller from the same company that did the same thing.

To date I have have only had my unit holding float a few hrs testing though it seemed stable. Just something to keep tabs on in case it's a wide spread software bug.

Kurt

Last edited by offgridQLD on Sun, 26 Oct 2014, 05:48, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber »

Yes, that was a very satisfying days work -- or was it play -- it's hard to tell. Image

Some more thoughts on the PIP-4048MS inverter/charger/MPPT:

The worst thing about it is its no-load power consumption -- a whopping 50 watts! In designing an off-grid power system with it, I have had to include this in the load calculation. It's equivalent to having a second fridge! But this one won't keep your beer cold. It has meant oversizing the battery and the PV array by about 25%. Or putting it another way, what could have been a 4 kWh/day system is only a 3.2 kWh/day system. I'm guessing this no-load loss is mostly hysteresis loss in the core of the high frequency transformer.

The second-worst thing is its unknown life-span or reliability. Check back here in 5 years time. To deal with this, I have included the cost of two of these units in costing the job.

So, with these drawbacks, why use this device at all?

Firstly, I'm using a Lithium battery. In fact, premium CALB-brand cells from EV-Power Australia. 16 x 180 Ah cells. So I need to protect them from over-charge and over-discharge based on the voltage of the most stressed cell at any given time. It's just asking for trouble to only use the overall pack voltage. On discharge you could have every cell getting low around 3.25 V except for the runt of the litter which is down to 2.55 V and about to hit rock-bottom, and the overall battery voltage is still a respectable 51.3 V.

Using the protocol that Coulomb and I reverse-engineered yesterday, we can still only raise the inverter cut-off voltage, on-the-fly, to 48 volts. This is one way in which this inverter (and pretty much every other inverter on the planet) fails to cater for LiFePO4 batteries. So we will also need to modify the inverter's voltage-sensing to make it read some voltage above 51.3 V as if it was 48 V.

So one reason for using this unit is that it is far easier to hack protocols and voltage sensing in a single combination unit rather than in three separate units: inverter, genset charger and solar MPPT charger.

The other reason is that this unit costs about one fifth as much as well-known reliable combination units with low no-load consumption like the Selectronic SP Pro.
Last edited by weber on Sun, 26 Oct 2014, 08:09, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

PIP-4048MS inverter

Post by PlanB »

Interesting box of goodies. I've been having a little play with one via an RPi. Just using EV works cell top modules for balancing at the mo. So if you plug your BMS into the comm port where am I gonna plug in the Pi?
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by coulomb »

offgridQLD wrote: So I am assuming you can send the same commands using different (voltage)numbers and the PIP4048 would respond appropriately?
Yes. And of course for every different voltage request, you need a different CRC. For a while I thought we'd be able to get away with just a few voltage settings, so we could just store the CRCs in a small table, but that would not allow fine enough control. So that's why the CRC algorithm is important.
So the thing we don't have user control over is the timers.
So far, no. There is a tiny chance that the firmware has a command ready to respond to that would change the timer, that the WatchPower software just doesn't bother to give you an option to change, but it seems unlikely.
I was wondering if the Min & max time limits could be changed?
Well, unless there is an undocumented command as posited above, this would involve a change to the firmware. It looks to me that the inverter may use a TI DSP chip, and there is a 14-pin 2.54 mm spacing header with pin 6 missing just like TI use for their JTAG ports. The problem will be that they will have set the security bits; being lax with this as with the Elcon/TC chargers seems unlikely. If there are firmware updates possible, this is a possible way in. Reverse engineering the firmware of an inverter is a huge task.
Perhaps going about it that way [ adjusting timers ] is more difficult and of no advantage.
Since they have chosen not to allow the timers to be changed, yes, it is quite difficult. It's not of no advantage, just not likely possible.
I take it you can only sniff out commands that you can trigger from the software and sniff as it's being sent to the pip4048 and the timers wouldn't be taking that path.
Right. We sniff the serial line between the PC and the inverter unit. Calculating the absorb time is something that would be done inside the inverter's firmware. Or actually, there may be a separate DSP or microcontroller chip with separate firmware that lives on the MPPT (solar charging) board.
On a side note, A member using a smaller version of the unit (3kw) has reported that he has had a issue with the unit playing up when on float. ...
Yikes. Without knowing how the various processors talk to each other in that unit, it's difficult to speculate on how this might happen. I wonder if continually sending new float voltage setpoints might have fixed this. If it's just that the MPPT processor crashed, then resetting it might fix it; if it's some rare bug that overwrote the float setpoint, or a glitch affecting the RAM, then just sending a new setpoint might overwrite the bad value with the good value and fix it. But you would have to wonder what else got overwritten if that's the case, and what effects those might have.

Certainly worth watching out for.
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:

PIP-4048MS inverter

Post by weber »

PlanB wrote: Interesting box of goodies. I've been having a little play with one via an RPi.
That's great news. The more we can share information on these systems, the sooner we can win the war that the coal-fired power companies, network owners and their political lap-dogs have declared on us, and on the planet. Bring on the network death spiral, I say. Image

So what have you been trying with it?
Just using EV works cell top modules for balancing at the mo.
Do you mean EV Power Australia's cell-top modules? I thought EV Works only had a spaghetti BMS that doesn't do balancing.
So if you plug your BMS into the comm port where am I gonna plug in the Pi?

Same as if you used them in an EV. Use one of these USB to RS232 converters so you have two UART ports on the RPi. Talk to our BMS at 9600 baud with one and talk to the inverter/charger at 2400 baud with the other.

The more I think about it, the more I think I might like to use an RPi for this too. I don't think I can do it direct from last CMU to inverter. Kurt's description of scary overvoltage faults with these inverter/chargers, reminds me that I need to have the BMS (or a PVCU on its behalf) control a last-ditch contactor that isolates the battery if the inverter/charger refuses to listen to commands to back off.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Adverse Effects
Senior Member
Posts: 1172
Joined: Sat, 01 Jan 2011, 03:30
Real Name: Adverse Effects
Location: Brisbane

PIP-4048MS inverter

Post by Adverse Effects »

this guy is talking about updating the systems so there must be a way in to the firmware

http://youtu.be/_FJhHsDZBrk
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

" I need to have the BMS (or a PVCU on its behalf) control a last-ditch contactor that isolates the battery if the inverter/charger refuses to listen to commands to back off"

I think this double redundancy (last-ditch) contractor is a smart option. In fact I don't think I would trust any device charging or loading a lithium house bank without it.

Regarding the 50W consumption did you put I meter on it. From memory at was less than 50W 40 - 45? . Still not great but I don't think its that bad for a 4000w inverter charger. Sure its not the super efficient 8w of the Sp-pro but My one generation older (selectronic PS1) consumes about 60w idle on.

50W idle x 24hrs 1.2kw two extra 200w panels should cover it.

The unit has a power saver mode that will drop the consumption a considerable amount but I find with a modem home with modems, 240v smoke alarms, electric garage door openers and the like you just cant use this feature.

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

PIP-4048MS inverter

Post by coulomb »

Adverse Effects wrote: this guy is talking about updating the systems so there must be a way in to the firmware


Yeah, but you have to have a legally blind friend to remotely do the install for you Image

Seriously, it sounds like there is some support out there. I might even consider a pair of these myself.
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.
T1 Terry
Senior Member
Posts: 1952
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry »

Our system isn't quite as complex. We use Junsi cell log8's in tandem, a timer circuit triggered by the alarm port at 3.6v in any cell with an SSR to cut the solar for 3 mins and a second one to trigger the hot water element for the same 3 mins, then the timer resets and the whole thing starts again.
It's worked fine for 3 1/2 yrs, although we have found there is virtually zero gain from the MPPT facility, the charge rate is the same direct connected through the SSR, so it's kind of a waste of time really bothering with the inbuild charger function at all. Add to that, the unit runs much cooler with the solar not connected through the MPPT.

As far as the stand by load, it won't even power the microwave clock without going into full standby load, so we either put up with the losses or turn everything off that has a parasitic load. The other option is to run a small inverter to power the parasitic loads, but this requires quite a bit of rewiring and frankly, too much of a pain in the butt to be worth thinking about. The auto defrost, door seal heater and timer clock in the fridge is enough to send the things into full standby mode, so we just learned to live with it.
The older style heavy weight transformer units don't seem to be as heavy on the standby load, their overload capacity is quite high and they can maintain it for quite some time, like 20 mins or more. The high frequency units have an over load capacity that lasts about as long as the blink of an eye.

T1 Terry   
Green but want to learn
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

" we have found there is virtually zero gain from the MPPT facility, the charge rate is the same direct connected through the SSR, so it's kind of a waste of time really bothering with the inbuild charger function at all. Add to that, the unit runs much cooler with the solar not connected through the MPPT"

What voltage is the PV your feeding to the unit is it near to battery voltage?

I know with some of the real cheap and nasty chargers listed as MPPT they have just used the words (MPPT) as marketing hype and they are not true MPPT chargers. Perhaps the PIP is the same.

The issue with using two of the units in parallel is you double that idle consumption to 100w.

Personally I just purchased one as a light weight inexpensive inverter that could handle 2200w load for a few hrs (the charger was just a bonus)So if it proves to be a reliable inverter then that's a good start and I think still getting good value as a stand alone inverter. Even if it turns out a additional MPPT charge controller is needed to achieve a reliable package.

Hey if you jest went out and purchased a Sp - pro (sure it would just work) but were is the fun in that Image

Kurt
Last edited by offgridQLD on Sun, 26 Oct 2014, 16:47, edited 1 time in total.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber »

offgridQLD wrote:Regarding the 50W consumption did you put I meter on it.
No but we were astounded at the amount of heat it pumped out while doing absolutely nothing (no solar input, no 230 Vac output). It had to run its fans while doing nothing. You could have dried your socks and undies with it!

Yes, we soon turned on the load-detecting power-save mode, but agree it will be totally useless in real life.
50W idle x 24hrs 1.2kw two extra 200w panels should cover it.

That's the thinking of someone who has an off-grid system with 5 to 7 days of battery storage, Image because solar panels used to be expensive and batteries (lead-acid) couldn't be deeply cycled.

Now that panels are a dollar a watt and batteries (LiFePO4) can be cycled to 70% DoD 3000 times, the old battery and array sizing rules-of-thumb, based on AS4509, are out the window.

I'm using pretty much the new rules-of-thumb kindly explained by Positronic Solar.

[ Edit Coulomb: this seems to have moved here: https://www.dropbox.com/sh/vjqgmy62umy46q0/AACn-ye6zcsKOvWcVJ6Irf2ua/Positronic%20ESS%20Booklet.pdf ]

[Edit: Weber: This seems to be the closest thing still available online. See PDF page 13 or search on "positronic" in https://renew.org.au/wp-content/uploads ... ipbook.pdf]

So, although this system is designed to deliver 4 kWh per day (including the PIP-4048MS's bonus "clothes drier" function), It will have 3.5 kW of PVs (3S6P x 195 W) and 9 kWh of storage (16 x 180 Ah LiFePO4). So there's a little over 1 day of storage if I cycle between 25% and 75% DoD, and a 3.5 times oversupply factor from the PV array.

So in simple terms, your 2 panels for the bonus clothes drier would need to become 7 panels.
Last edited by coulomb on Mon, 30 Nov 2015, 11:05, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

Fans running without doing anything. I don't remember my unit doing that with it just plugged in. Perhaps it was hotter ambient temp today. Will try that tomorrow.


Yes true can agree with that thinking. Much the same as our offgrid system now 8.2kw of PV (currently 63kwh of lead) Though it will be replaces with roughly 20kwh of lifepo4. Even though at times like today we consume more than 30 kwh over 24hrs charging EV's and running air conditioners. Though predominantly most large loads are during the daylight hrs. Less money tied up in battery's has to be a good thing.

4kwh/day system. that's a very modest 24hr load for a house.

It makes me think it funny how Everyone on our street is offgrid about 8 homes and it's interesting to see all the different Sized and configured systems so cover the different lifestyle's.

Across the road the couple just have 1500w of PV (well placed no shade) but they are frugal in there consumption and they get by without starting there generator in all but the most gloomiest of weather. Actually I cant remember the last time I heard it.

Next door at a guess they have around 4000 - 5000w of PV though it's large home with all the mod cons and I just found out they actually power another family members house down the road from there system to.(must be a 500m cable run!) Though I often hear a very faint hum of there generator in the afternoon on days when I have hit float at 10AM So I think they are under done with the PV.

The newest house on the street just went up this year has installed lithium battery's and around 3000w of PV. They haven't seen there a winter yet and I have a feeling they to are a little under done with the PV to see it through without fossil fuel support but they might just be frugal.

Everyone's energy needs are so different but you soon learn to just your lifestyle around the days weather.

I find in QLD we never really get to many days in a row that a real shockers for PV output without the sun breaking through. Unless we get the washout from a cyclone up north now and then.

Kurt
Last edited by offgridQLD on Sun, 26 Oct 2014, 17:31, edited 1 time in total.
User avatar
PlanB
Senior Member
Posts: 471
Joined: Sat, 16 Jan 2010, 15:24
Real Name: Kris McLean
Location: Freemans reach
Contact:

PIP-4048MS inverter

Post by PlanB »

Just mucking about here really, relatively cheap entertainment. Sorry I meant EV power cell tops

Does your CRC generator work with the POP02 (battery output source priority)command? Mine gives E2 0A as the CRC but the box expects E2 0B.
weber wrote: Sun, 26 Oct 2014, 17:11 Using the protocol that Coulomb and I reverse-engineered yesterday, we can still only raise the inverter cut-off voltage, on-the-fly, to 48 volts. This is one way in which this inverter (and pretty much every other inverter on the planet) fails to cater for LiFePO4 batteries. So we will also need to modify the inverter's voltage-sensing to make it read some voltage above 51.3 V as if it was 48 V.
If you use a resistive divider to fool the battery voltage measurer into thinking 53v is 48v for cutoff won't that change the bulk & float values too?
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber »

offgridQLD wrote: Fans running without doing anything. I don't remember my unit doing that with it just plugged in. Perhaps it was hotter ambient temp today. Will try that tomorrow.
Maybe it had something to do with the fact that the inverter was lying on its back for the comms sniffing experiments rather than mounted on a wall. But it had clearance all around.
4kwh/day system. that's a very modest 24hr load for a house.

It's a typical off-grid consumption for a couple with no kids, a house designed not to need aircon, wood for space heating, no swimming pool, no EVs (unfortunately), efficient single combination fridge-freezer, gas cooking, and solar hot water.

I say "solar hot water", but not necessarily in the way you might think. With a 3.5 to 4 times oversized PV array I'm now recommending heat-pump hot water with substantial water storage. The heat-pump will only be powered when the battery is full.

These folks have been quite happy for the last 15 years with only 2 kWh per day from 500 W of PVs and 11 kWh of lead-acid battery. Although with that they used gas for both fridge and hot water.
Last edited by weber on Sun, 26 Oct 2014, 19:05, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber »

PlanB wrote:Does your CRC generator work with the POP02 (battery output source priority)command? Mine gives E2 0A as the CRC but the box expects E2 0B.
Ours gives E2 0A as well. So the box is weird. Thanks for letting us know.
If you use a resistive divider to fool the battery voltage measurer into thinking 53v is 48v for cutoff won't that change the bulk & float values too?

Yes, it will. But I think we can tolerate this.
[Edit: Changed from "But it turns out we need this too" to "But I think we can tolerate this"]

Whatever gets mapped to 48 V will have to serve as both:
(a) the lowest charging voltage at which a single cell could be overvoltage, &
(b) the highest discharge voltage at which a single cell could be undervoltage.

52.8 V (3.3 V per cell) seems like a good voltage for that. That's a 10% increase over 48 V. I'm guessing we'll have to change two separate voltage dividers, one in the inverter/charger and one in the MPPT.

Actually we're going to attempt balancing at 76% SoC, at 3.320 V per cell, where there is a bit of a step in the voltage vs SoC curve. This will need precise cell voltage measurement.

T1 Terry,

Thanks for the heads-up that this might not be a proper MPPT and a separate one may be needed. How many panels did you have in series (in each string) feeding the MPPT? And how many cells are in each panel, 72 or 54?
Last edited by weber on Mon, 27 Oct 2014, 05:32, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

As I mentioned I would yesterday. I ran my PIP 4048 for two hrs today laying on it's back and that is how I have always run it anyhow. The fans never came on and the case was cool to the touch.

Perhaps I missed something. You were running the unit without a battery connected or PV connected and just powering it from the AC input?

Kurt

T1 Terry
Senior Member
Posts: 1952
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

PIP-4048MS inverter

Post by T1 Terry »

6 x 12 so 72 cell panels, 2 in series, each panel had an STC Vmp of 35.9v so series string Vmp should have been 7.1v, but we all know STC figures are nonsense, in full sun with the panels at around 70*C the true Vmp was around 64v, could have been as low as 60v at times, that one is a tad tricky to determine in genuine solar conditions as they are never stable.
We found the best long life formula was to aim for 3.5v per cell boost and 3.45v float and not allowing a cell to remain at 3.6v. We have over 1,000 cycles on the test batteries and they still deliver their advertised capacity while remaining above 3v per cell. Unfortunately my set of test batteries were held at full solar output from 12v nom. solar panels, around 17.5v every day for 6 weeks while we were away due to my stupidity. They are now quite pregnant, have high internal resistance, but they still seem to work. I haven't managed to squeeze them flat yet so I can fully charge them and see what remaining capacity they have, been busy as a one arm wall paper hanger with a dose of the crabs.

T1 Terry
Green but want to learn
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

"6 x 12 so 72 cell panels, 2 in series, each panel had an STC Vmp of 35.9v so series string Vmp should have been 7.1v, but we all know STC figures are nonsense, in full sun with the panels at around 70*C the true Vmp was around 64v, could have been as low as 60v at times"

So really only 4 - 9 volts head room to MPPT with I can see why you wouldn't see a huge benefit in the MPPT feature.

That said I'm not saying the MPPT charger in the PIP4048 is a true MPPT as I haven't played with the charger side of the unit much at all yet.

Kurt
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

PIP-4048MS inverter

Post by weber »

offgridQLD wrote: As I mentioned I would yesterday. I ran my PIP 4048 for two hrs today laying on it's back and that is how I have always run it anyhow. The fans never came on and the case was cool to the touch.

Perhaps I missed something. You were running the unit without a battery connected or PV connected and just powering it from the AC input?
Ours was connected to a battery, and only a battery (and the comms port). Did you turn off the power-save (load-detect) feature for this test?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
offgridQLD
Senior Member
Posts: 1827
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

PIP-4048MS inverter

Post by offgridQLD »

"Ours was connected to a battery"

I think the above is the difference that matters. I just powered the unit via it's AC input. So we can forget that comparison as no battery was connected.

Though when testing the unit out a few weeks ago with a small battery connected I had the unit powered up for about 1hr on the bench (not in low consumption mode) and didn't get the fan running. Only when I loaded it with some workshop power tools for a while did the fan come on.Perhaps I just didn't let it sit for long enough unloaded and on.

40 - 50w with no where to go in a steel box will heat up I guess.

Kurt



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

PIP-4048MS inverter

Post by weber »

That's another big problem with this device. As I wrote in Kurt's everything-thread some weeks back: It has a very narrow MPPT range (60 V to 115 V, less than 1:2) which appears designed to take only 3S of the 72-cell modules (around 200 W each) or 4S of the 54-cell modules (around 250 W each). (216 cells in series in both cases)."
viewtopic.php?title=new-to-the-forum-ne ... 714#p52693

I would not expect it to do much useful MPP tracking into a 16 x LiFePO4 battery with only two 72-cell panels in each string. An MPPT is typically a non-isolated buck converter. So with 2S, if even one cell of one panel gets shaded, you'll get nothing from that string, despite the bypass diodes.

And given that this installation has significant shade moving across the array at some parts of the day, I am forced to use three 72-cell panels per string. But then I run into the PIP-4048MS's array open circuit voltage limit of 145 V. The array open-circuit voltage may well exceed this on a frosty midwinter's dawn.

I'm using these Suntech 195 W panels.
http://www.solar360.com.au/files/Suntec ... 0Specs.pdf

With their Voc of 45.4 V at 25°C and tempco of -0.34%/°C they only have to be below 6°C for 3S to exceed 145 V.

Any suggestions?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

PIP-4048MS inverter

Post by Johny »

How much would you guys give me if I could provide document with the serial comms protocol for the PIP-4048MS...
Never mind - I'm feeling generous.

Edit: Added file to forum HS_MS_MSX_RS232_Protocol_20140822_after_current_upgrade.pdf
Last edited by Johny on Mon, 27 Oct 2014, 10:09, edited 1 time in total.
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

PIP-4048MS inverter

Post by Johny »

While I'm typing this I'm assuming the next question will be where I got it.
I found the manufacturer and sent them an email this morning when I saw the chatter on the forum. I do this a lot for various things with not much success.
They actually answered.

Dear John

You may be looking for the communication protocol - please see attached.

thanks
Eric

On 2014/10/27 08:55, John XXXXX wrote:

Name John XXXXX
Email john.xxxx@xxxxxxxxx.com
Subject PIP-4048MS
Message Hi. We are adapting this inverter for use in charging Electric Vehicles and using Lithium batteries. Is there as document available that list all the serial commands like the ones that the PC application uses to "talk" to the inverter? We have figures out setting voltages but are wanting to also change absorb timers etc.
Site http://www.mppsolar.com/v3

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

PIP-4048MS inverter

Post by weber »

Ha!

Johny, you're a genius! Thanks!

To answer your question: I would give you my gratitude. But you already had that. Image

Before your second post, I thought "Why didn't I find that?". On Saturday I googled a bunch of the commands we saw when we first sniffed it, "QID QMODI QPIGS QPIWS" (without the quotes), and turned up nothing, nix, nada.

But that was before we realised the trailing "I" on "QMODI" was part of the CRC. So I just now googled "QID QMOD QPIGS QPIWS" (without the quotes), and turned up this in Google's cache:
http://webcache.googleusercontent.com/s ... =firefox-a
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
Post Reply