PIP-4048MS and PIP-5048MS inverters

[offgridQLD]

Water is all I have at the ready if it all went up in flames. Worth mentioning to make sure that you will still have water pressure if your off-grid system is on fire. Assuming you have AC powered pressure pump.

I have a small tank off the shed (5000lt) with a fire pump always connected. Though all that only counts If I was at the house at the time of a fire event. though I think it could just burn for hrs and the solid cement block room would contain the fire. Would be a big mess when you opened the door though

Hmmm your points about lifepo4 ventilation and power electronics have me thinking Weber. My situation gets complex with a enclosure within a room that is some times vented and some times sealed (depending on refrigeration requirements). Though there is a separation shelf between the two battery and power electronics at times do share the same rooms volume of air that's sealed.

I wonder if there is some room volume or air exchange numbers that could come into play in the future rule to prevent the flammable air/fuel ratio.I have experienced the same (candy smelling solvent) odors after placing lifep04 cells in sealed boxes for a while then opening the lid. Though unsure if it has reached a point air/fuel ratio to trigger a flame from a spark. Though I haven't smelt it in my small power room. Perhaps the night time fan ventilation is enough to overcome it.

Perhaps I need to program my room ventilation timer to come on at a few intervals pr day to exchange the air...though that won't help with coinciding with a dramatic cell solvent venting event. So many possibility's.

Kurt

[Johny]

I wonder if there is a gas sensor that could detect venting early enough to be useful in overriding ventilation fans?
Would any of these work?
http://www.futurlec.com/Gas_Sensors.shtml

[offgridQLD]

Well I had a skim over the linked doc. I did see pure Argon mentioned as a way to snuff out a battery fire. I have a large G sized welding bottle of 100% pure argon in the shed at 200 bar. Perhaps a solenoid valve that injects Argon into the room based a high solvent trigger alarm.

This is getting silly. I think I will just let it burn .

Edit: I also noted a 500mm side clearance and or 100mm below the top of the cell mentioned for spark triggering of battery gas but this would have been rules for lead acid. I don't know how the lithium solvents would react.

[paulvk]

Its all great saying that lead acid batteries and electronics need to be separated but how come all the UPS have them packed into the same tight case?

Ok onto more useful PIP stuff.
Making good progress with my web-server to inverter project its working as a web-server and talking to the inverter sending the QPIGS command every 10seconds then displaying the data on a 64 x 128 LCD display.
It also calculates the CRC on both send and receive.
As it has a micro SD card (from this it serves pages) I am going to log the data in comma separated text format so it can be used in excel spreadsheet

[weber]

Its all great saying that lead acid batteries and electronics need to be separated but how come all the UPS have them packed into the same tight case?

Two reasons why those small UPS for a single computer are OK in such cases, where a system for an entire household would not be.
1. The maximum rate of hydrogen generation for a given lead-acid cell type is proportional to the maximum charge rate in watts, so the destructive power of small systems is limited.
2. A tight case means few spaces that can fill with an explosive hydrogen/air mixture.

[Northland]

3. They are cycled once a year if they are lucky and spend 99.99% of the time not charging
4. There are fans constantly pushing air through

[Northland]

2 questions :

1. A lot of people using LifePo4, but what cells are you using? I see there is a new 42120 cell, 12.5aH. Seems to be best ah/$. I was looking at 38120 which were 10th.
2. Why stop at 15 cells? What are the voltage limits of the pip? The MPPT seems to be 60A limit so why not maximize the voltage to get maximum watts?

[weber]

Those are important questions offgridQLD, about what percentages of the solvent in air are flammable, and what those percentages smell like, and what's the lowest percentage you can easily smell. I found the flammability percentages for dimethyl carbonate once, but nothing about smellable percentages (or parts per million). But that information may well be on the web somewhere.

I look forward to government funding for some expensive destructive experiments which are videoed. Like the great ones RISE did in the past for solar panel and DC circuit breaker fires.

There's a fascinating argument here about Lithium-ion battery fire fighting.
https://en.wikipedia.org/wiki/Talk:Lith ... attery_.3F
Particularly the section beginning:
"I have deliberately destroyed plenty of batteries of all types and ages for the purposes of type approval in various uses (and so that the aerodrome fire department could assess fire fighting techniques)."

Johny, I think there's a good chance the alcohol detector might detect the Lithium ion solvent. But would you trust it to still be working in 10 years time when the fault occurs? Same with fans. Clearly, passive ventilation systems are more trustworthy.

Hi Northland. Coulomb and I are using prismatic LiFePO4 cells of between 20 Ah and 180 Ah. Lishen brand for the former and CALB brand for the latter. The only argument for using 15 LiFePO4 cells with a PIP-4048MS (hey, I'm back on topic ) was before Coulomb did his masterful modifications to the firmware, which allows us to set the low voltage cutoff higher than 48 V. So there is no reason not to use 16 cells now. We have always used 16 cells, we just relied entirely on the BMS and a bunch of contactors for low voltage cutoff in the past.

[Northland]

Yes I understand the firmware mod, but why stop at 16? Why not 17-18-19 or more?

Why was it necessary to change to 16 cells? (Other than out of the voltage ranges for the pip, which was fixed with firmware)

Why did you choose those particular cells? other cells are better on ah per $ basis. I just came across a 24v 18ah pack for 100usd, that's 0.69USD per ah per cell.

[offgridQLD]

I think you will find there are voltage limitations on the internal components of the PIP inverter . Caps in standard form are limited to just around 63v (from memory) and I'm sure many other components have limits.

As for the style, brand of Lifepo4 cells used. Part of the reason to use large format prismatic cells like Calb is the practicality of a single 180 - 400Ah cell vs trying to string together 40 or so parallel 10ah cells.

180ah at 52v is about the lower limit I could see being practical for even a small frugal offgrid application. A larger home with more mod cons 400ah fits well.

The reputation of a brand is something to . Its a big Gamble with $5000 to say $15000 on a unknown manufacturer. Calb and a few others have some track record in home built EV's and stationary storage with local distributors and some warranty.

That said, I have a soft spot for recycling OEM EV traction packs. I just feel the quality is there and the price is great. It has to be a good thing. Recycling what would have gone to wast. I would be happy to go down that path for my next set of cells. Though I do feel the chemistry used in OEM ev's needs to be managed well. As it's less forgiving than lifepo4

Kurt

[Monkeytom]

I started with S16 LiFePO4 but I am still using some LA as well, The Li Pack would trip the 60A breaker from time to time on charging, so I have reduced the Li pack to 15S and it is much happier being mixed with LA.
The different chemistries accept different charge rates and I noticed when a cloud would come over the charge current would still flow to the LA from the LiFePO4.

[Northland]

What type LA do you have? I was thinking of adding lifepo4 to my agms but thought it would be impossible. Might be possible to balance using those voltage relays

[weber]

Yes I understand the firmware mod, but why stop at 16? Why not 17-18-19 or more?

Why was it necessary to change to 16 cells? (Other than out of the voltage ranges for the pip, which was fixed with firmware)

Why did you choose those particular cells? other cells are better on ah per $ basis. I just came across a 24v 18ah pack for 100usd, that's 0.69USD per ah per cell.

I agree with everything Kurt (offgridQLD) wrote in response to these questions.

The PIP hardware is running very close to the wind as regards maximum battery voltage so we must not go over the existing maximum charge voltage setting of 58.4 V. 58.4 / 17 = 3.44 V per cell. That's just a bit too low for reliable cell balancing. And the inverter caps are likely to last longer if we stay well away from 58.4 V.

So the only realistic options are 15 or 16 LiFePO4 cells.

[coulomb]

why stop at 16? Why not 17-18-19 or more?

Because it's a 48 V (nominal) inverter. It's designed for a high voltage cut-off of 60.0 V (PIP User manual, near the bottom of page 28). Even 17 cells would mean spending a lot of time at 3.45 x 17 = 58.7 V, and that's too close to the upper design limit for my liking.

I'd like to see a 96 V nominal inverter (I'm sure there are a few out there), but it seems that the battery would under certain circumstances be too close to the 120 V ELV limit. So perhaps a 72 V nominal version, but no-one seems to make them.

[weber]

I was going to do a home power system using second-hand Nissan Leaf cells (NMC chemistry), but both the customer and myself got cold feet after we heard about Jack Rickard's fire, even though it was due to (ab)using them without a BMS.

viewtopic.php?title=jack-rickard-no-bms ... 757#p59830

The only numbers of series NMC cells usable with a PIP are 13 or 14, and since they come in modules ("sardine cans") with pairs of cells already in series, 14 is by far the most convenient number.

[Monkeytom]

What type LA do you have? I was thinking of adding lifepo4 to my agms but thought it would be impossible. Might be possible to balance using those voltage relays

I have flooded cells two strings of 8 T105 6 Volt 220Ah type golf cart batteries for a 440Ah 21Kwh LA Pack. I have put 60Amp resettable breakers on each battery bank string so they are fused separately.
I bought a used set of 43 x 90Ah Thundersky cells from an abandoned EV project for our AC-50 EV project it was 45 originally and he lost two cells from a faulty BMS. I have just bought another two new ones to make it up to 45 cells 15S3P and was going to use them on the off grid system in the mean time, I have hooked up one set of 15s 4.32Kwh in parallel with the LA and dropped my charge voltage to 56v till I get the new BMS going I have bought a set of EV Power Cell top BMS modules and a couple of 100A SSR's to run back to back and once the voltage gets up they will disconnect the Li off the system and let the system top up the LA as soon as it goes to float at 54v the Li will come back online. I have my PIP set to go back to Grid at 47 volts on the low side but seldom see that now with the extra battery capacity. I am aiming for a 30S2P 90Ah pack in Lara's Karmann Ghia project and now I am thinking I want to split that in two and connect it to the OFF Grid system 2x PIP4048HS with 5 Kw of panels for solar charging and extra storage when not being driven.
By the way we went to Stocklands Burleigh Today and I could hardly believe my eyes, car park was nearly full except two empty spots right at the front for EV charging only.

[offgridQLD]

"I was going to do a home power system using second-hand Nissan Leaf cells (NMC chemistry), but both the customer and myself got cold feet after we heard about Jack Rickard's fire, even though it was due to (ab)using them without a BMS."

Yet we can sleep at night with a Nissan leaf parked in the garage. I understand the decision to put it on the back burner. That Jack Rickard reality check of what can happen would have shaken my confidence to.

Though I'm sure with a large OEM like Nissan. The last thing they want is a safety risk to the public and all the mess that would follow. It makes me think that all boxes would have been ticked and and no stone would have been left unturned befor installing them in the car.

Just insuring the same standards are maintained out of the car in a stationary storage setup is the trick.

Kurt

[paulvk]

Now have the server calling the QPIGS to the inverter when a web-page is requested sorting it out and replacing the parts of the page with the results of the QPIGS just a simple page but it can be made as jazzy as you want without changing the program in the flash of the AVR

[beaujewel]

For those considering the coulometer from a few pages back, I have one. Its been monitoring my lifepo4 for around 7 months now and appears to be very accurate. I asked for a 150a shunt and they were very obliging.

[edmundp]

Hi All,

I have searched - but could not find any particular reference to this.

Has anyone ever run the PIP-4048 with Li-Ion cells (Tesla style) - in something such as a 14S configuration?

Any experience with this? I am aware of the cell volatility, but given a large enough bank and thus low enough relative current draw temperatures should be kept under control.

Any thoughts?

Edmund

[coulomb]

Has anyone ever run the PIP-4048 with Li-Ion cells (Tesla style) - in something such as a 14S configuration?

Lots of us run 16S of the 3.2 V LiFePO4 cells, which technically are also Lithium Ion. But obviously you mean the various 3.7 V chemistries, where 14S would be similar in voltage to 16S of LiFePO4.

Firstly, I'd be terrified of the safety issue. But if this could somehow be addressed, I believe that most if not all of the non-LiFePO4 lithium chemistry cells have a far less flat voltage versus SOC curve, meaning that you can get a much better estimation of their SOC merely from their terminal voltage. So that might work out much better with the PIP/Axpert inverters, which have a strong "lead acid thinking" to them.

I'm not aware of anyone running these. We were going to use Leaf cells (which would be similar, and I believe it would have been 14S, 7S of 2S2P modules in a can) for a customer, but after the Rickard fire with these cells, even though he was basically asking for trouble, we got cold feet and decided to switch to CALB LiFePO4 cells instead.

Others may not be as skittish, and may be using or planning to use Leaf modules, but I'm not aware of any at present. There are also Chevy Volt modules available from an importer in Brisbane, but I don't know anything about these.

[Adverse Effects]

the Tesla packs are 57 Volt, 3kWh as far as i know and use 18650 cells in Modules( 15s 22p ) and i am sure there is someone out there somewhere that has worked out the BMS on them or worked out a way to add an aftermarket BMS to them

my personal thoughts are if the PIP-4048MS can be configured to work with the same voltages i think it would be a good setup


PS:- i do not have a PIP-4048MS but have read this full treard as it has grown


[EDIT]Lithium Ion 18650 EV Module - 57 Volt, 3kWh

LOL just found there BMS guide

[coulomb]

the Tesla packs are 57 Volt, 3kWh as far as I know and use 18650 cells in Modules( 15s 22p )

It's a pity these aren't 14S. 63 V max is too high; one cell less is still a bit high for my liking, but you could stay away from 100% SOC and squeak it in.

[Adverse Effects]

with my 18605's i never charge past 4.1 and never drain past 3.0 i have some with well over 5oo cycles and don't see any real capacity loss at all in them

[PlanB]

Those Tesla modules are something to drool over are they not! I could never buy them for a solar app, the temptation to try to shoehorn them into the Leaf somehow would be just too great.