Lithium, Mixing Sinopoly LFP and Winston LFYP Cells??

[native pepper]

Terry can you explain why a 90ah cell in parallel with a 100amp cell will not continue to accept charge when it has reached its capacity and the 100amp hasn't. My ignorance says the 90amp cell would still be getting charged because the 100amp is still charging, so an explanation would be nice to hear.

As far as cells bulging, my house bank of 700ah lifepo4 will be 10 years old later this year, has never been strapped and has never bulged, nor have they lost any of their capacity. The only thing holding them together is the copper bar connections. For some reason my 40ah cells are averaging 42ah now, when they were installed they averaged 40ah. Slowie claimed if you used proper charging regimes, they would improve with age, how he worked that out is beyond me but not surprised he seems to have got it right.

Cells over 50ah, he claimed would have shorter lives because they charge from the center and create internal heat, whilst the smaller cells take charge evenly throughout the cell. He also said large cells tend to slump and bulge, which he felt would lower their lives considerably. Have yet to run across anyone with large cells much over 5 years and the ones I've seen not that old have lost capacity already and are bulging, but they use lead acid chargers supposedly configured for lifepo4, but aren't in reality.

That could be why none of my cells ever get warm, even when the pack are accepting over 100amps. From what I've been told and observed in testing, it is cells over 50ah and plastic encased, that suffer bulging and heating. All my cells including bus, boat and workshop, are 40-50ah cells, industrial metal cased.

The only time I've seen a metal encased small cell bulge, was when they were testing for overheating and no matter what they did, nothing would make the cell get hot or catch fire, but it did bulge a bit when it reached 4.6v. It was only when a neighbour of slowhands when making his pack box, drilled into a cell which created a spark and the cell emitted loads of smoke then flame. Didn't see it myself but according to the bloke who did it, there was no explosion, just smoke until slowie arrived and put it out by covering the cell. Have seen photo's of the cell and believe none of the surrounding cells were effected in any way.

Also note the OP, seems to be using charge parameters well outside the safety zone for lifepo4 along with float settings. But it seems most are using lead acid charging parameters and completely neglect real lifepo4 requirements. Isn't taking a 12v pack over 14v, detrimental and the same taking it below 12v. The charge settings we use I was told are on the limit, 3.5v per cell upper and 3v per cell lower, which results in the pack settling at 13.8v, restart charging at 13.6 and not really cycling. Or do I have it wrong.

[BruceS]

Before this thread gets outta hand maybe we can just accept your post as a statement rather than a question!
I think your cells NP are totally different than any other ones I've ever seen or heard of in Australia (metal cased) and therefore to compare apples with oranges and comparing charging rates etc is also not going to be of much benefit either.

I'm glad what you are doing with YOUR cells works for you NativePepper ... I really am.

[bagmaker]

Thats interesting stuff Pepper.
From a Winston chart their 40Ah cell has a thickness of 47mm and an internal resistance of 0.7milliamp (I think the reference is milliamp anyways )
The next size is 60Ah, considerably thicker at 61mm but roughly the same size. Resistance has gone down to 0.55milliamp.
The chart
http://www.evlithium.com/Winston_Battery.html
then shows a progressive increase in size and a progressive decrease in internal resistance as the capacity rises.
There does seem to be an abberation of the smallest cell.

[Busman]

I just had a look at the Victron BMS, it does not alarm until 3.7 and on the low side 2.5, seems there is a bit more room in there ?

[T1 Terry]

For NP regarding the mismatched cell capacity connected in parallel. No matter what cells you use, even from the same manufacturer, the same batch and same advertised capacity, there will be a different capacity in real life, just one of those things when mass production is involved. All cells no matter what the chemistry will only charge when the voltage exceeds the voltage already held within the cell plus the internal resistance. Fortunately the Li cells we use have a very low internal resistance, 0.05v difference from rested to charging or discharging is required for energy to flow into or out of the cell.
As most already understand, these Li cells accept a charge current with very little voltage rise until they are saturated, then the voltage rises rapidly. If one cell in a parallel pair has reached saturation point but the other cells in the parallel group have not, the other cells will absorb any further current until the voltage rises above that of the saturated cell plus 0.05v. This means eventually all the cells will be fully charged plus 0.05v within the group.
A fully charged rested cell has a voltage of 3.4v, 4 cells in series equals 13.6v, charging to 3.45v per cell means every cell in that parallel group must have reached 3.4v and the additional 0.05v resistance lifts the terminal voltage to 3.45v, this is where the 13.8v end of charge in float mode comes into play. Any cell within the group lower than 3.4v will continue to accept current because the terminal voltage is at least 0.06v higher than the cell rested voltage. This will continue until the cell can not absorb any more current without the terminal voltage exceeding 3.45v. Hopefully I explained that and didn't muddy the water even further. It often takes quite a few re-reads before an old concept is put aside and a new concept can be taken in.

This 0.05v per cell adds up when the cells as linked in series, in a 4 cell battery that adds up to 0.2v. This means 2 four cell 12v batteries linked in parallel can end up with a genuine rested voltage up to 0.2v apart. Even though the terminal voltage reads the same when they are linked, the individual cell voltages if the links between the 2 batteries were removed would show a rested voltage of a cell in the same position in pack 1 to be different to the same corresponding cell in in pack 2.
That bit probably has a lot of eyes rolled back in the heads and the reader in a coma or near enough to it Once memory charge is added into the equation it gets real deep so I'll stop here
I had another 2 paragraphs one finger typed by when I re-read it even my eyes started to roll back

T1 Terry

[T1 Terry]

Hi Terry,
Thanks for the detailed reply. The cells are in a wooden box that my husband made designed to compress, and well bolted to van floor. I know wood is not ideal for cooling, but so far even after 5 mins at over 100amp discharge the cell terminals are still cold....now to work out how to add the pics......Well after some resizing that worked! Sorry re image quality. Battery and inverter live under the seat but the perspex cover reflects too much. You will also have to squint a lot to see the threaded rods on each side. The BMS and relays are in the grey box in the under table shot. Charging bits live in front box. The big CB is just for roof solar and used mostly just as a convenient switch so I can turn it off if charging from car or from portable panels when van is in heavy shade. Locations mostly so I could easily remove the lithiums and return the AGM system if we ever change vans. Inverter is not directly included in LVC, but via a relay from load bus and its existing remote controller, avoiding using a jolly big contactor. This is really just an added safety in case we forget to manually turn it off immediately after use.

A few questions Leslie, hope you don't mind.
What is the aluminium box mounted to the big heatsink just below the Victron MPPT controller?
How do you control the battery charger end voltage?
Are the circuit breakers DC rated or AC type?
Where did the balance boards that shunt at 3.5v come from?
Can these balance boards be turned off?
What current do they pass at max capacity?

T1 Terry

[Barboots]

I was interested enough in the aforementioned approach of applying a stepped up vehicle DC recharge to the solar controller that I did a bit of searching.

This?
http://www.ebay.com.au/itm/302138176071

Definitely interested in further discussion.

Cheers,
Steve

[native pepper]

Thanks Terry, had an idea in the back of my mind what you've described would be the case with lifepo4 cells because of how their voltages work compared to lead acid, but didn't know the physics behind it or if it was fantasy, that's why I asked. Sometimes old knowledge can infiltrate new understanding and become confusing, especially when dealing with the physics involved. The paragraph you say may make eyes roll, makes perfect sense to me now.

Would that be the case with cells of greater capacity difference, eg 2 x 100amp linked to 2 x 50amp in parallel. When not travelling, can connect my bus 480ah to the house and get more use, but have been disconnecting the 700ah house system and connecting it to the bus the couple of times I've tried it and always disconnect the 200ah workshop pack to use my 120ah portable pack. If it was safe to connect 480ah bus pack in parallel with my house pack, or portable pack, that would save time and make it easy. The bus pack is in an aluminium box with wheels on it, so it can be easily removed from the bus.

It frustrates me no end to have a mind that has never been capable of grasping maths or physics, even though I've done a number of courses and had someone with a physics and electronics degrees trying to explain it to me for many decades. Have lots of books on electronics, physics and computer technology and still don't get or understand the concepts until I see it in action.

Same with music, have spent more than 4 decades playing professionally, yet can't grasp musical notation or scores. Even though played in bands backing big international shows, others read, I learn it all be ear and just make out I'm reading the score. To give you an idea, last year was the first time I actually understood tablature, it has never gelled in my mind, then one day watched a youtube which followed the tab in a lead break and suddenly it clicked, I could see the notes on both the guitar and corresponding tab. Felt like a real dill as it was so simple, yet blind to my mind.

[T1 Terry]

Ah, that explains the giant heatsink, must get rather warm when operating. I wonder what the incoming current is when the output reaches 20 amps @ 19v. The 380w output with an input of say 11v allowing for cable voltage drop would put the incoming current at around 35 amps, the wiring in the photo sure doesn't look heavy enough to carry 35 amps so???? When you consider the heat that must be generated to require such a large heatsink you have to add in the % inefficiency to that 380w so maybe closer to 40 amps.......
The other thing against this approach is the MPPT controller can only function as either a DC to DC charger or a solar controller at any one time. There are already plenty of DC to DC chargers that have the same capability, add this unit cost to the Victron MPPT controller cost http://www.ebay.com.au/itm/VICTRON-ENER ... SwYvFZInfL at $257 and the price tag is now Roughly $303 and still requires switches, relays etc to make it functional. The Projecta IDC25 http://www.ebay.com.au/itm/PROJECTA-12V ... SwVCRZcYKG is only $255.....

T1 Terry

[T1 Terry]

Hi NP, if each battery pack has its own charge control, cell monitoring and over cell voltage/under cell voltage control.... and they are basically the same voltage, they can be coupled together without much of a problem. Keep in mind if one pack is deeply discharged the current flow from one pack to the other can be huge. I demonstrated this many yrs back at a Taggerty get together, a 5 cell 15v 90Ah battery recharging a 4 cell (8 cells really) 180Ah 12v battery that I'd just used as a demo to make a loaf of bread from the basic flour to baked. The battery recharged in around 15mins but the current flow was enough to make the 2/0 gauge (68mmsq) welding cable smoke at the jumper lead clamps. The clamp meter wasn't in the hunt as it topped out at 400 amps, something to keep in mind when paralleling lithium battery packs.

T1 Terry