Battery Power Supply

[jkeny]

Yes the batteries are very low self discharge - will sit on a shelf for years.

I still maintain that the problem with charging in series is that eventually they will go out balance - don't know how long it will take, though?

[jrling]

No sign of imbalance after several months here!

If one had to balance them manually every year or so, I can live with that.

[jkeny]

I really don't understand this - if 12V is input into 4 cells in a series, how does it split this 12V evenly over the 4 batteries - why doesn't the first battery in the series see the full 12V? Even if it doesn't see the full 12V, why don't these outside cells see a different charge voltage than the inside cells?

I'm obviously missing something - maybe if the charge current is low, it allows it to be distributed in a more balanced way?

This might explain it - if the batteries are put in series with equal voltages & near to the optimal voltage (i.e each one is 3V if using 4 in series & 12V charger) then there is very little current drawn from the charger (as long as the batteries are not let go low) & the current has a chance to distribute evenly?

Just some random thoughts!

[nige2000]

it was explained to me as a relation between battery internal resistance and cell voltage, cells are never exactly the same,
resistance varies a bit like transistor properties
the cell with the lowest resistance in an in series pack will be doing most of the work meaning it will deplete first
to get the best performance (Ah) out of a charge cells should be resistance matched so that all fully charge and fully discharge

my understanding is that in series cells kinda bais themselves even with a voltage imbalance and thats ok as long as the lowest resistance cell does not get over discharged

i see these guys do 12v packs with pcm

https://www.ev-power.eu/LiFeYPO4-batter ... l5ht03suv7

also cheap for 3ah cells :)

[jkeny]

it was explained to me as a relation between battery internal resistance and cell voltage, cells are never exactly the same,
resistance varies a bit like transistor properties
the cell with the lowest resistance in an in series pack will be doing most of the work meaning it will deplete first
to get the best performance (Ah) out of a charge cells should be resistance matched so that all fully charge and fully discharge

my understanding is that in series cells kinda bais themselves even with a voltage imbalance and thats ok as long as the lowest resistance cell does not get over discharged

i see these guys do 12v packs with pcm

https://www.ev-power.eu/LiFeYPO4-batter ... l5ht03suv7

also cheap for 3ah cells :)

Yeah, that would explain what I was saying - as long as the cells don't draw much current while charging (i.e trickle charging) then they will balance roughly but if this condition isn't always maintained & cells are allowed to reduce in charge (causing the charging to need to deliver more current) then the likelihood is that at least one cell will become unbalanced & this is a cascading event which will eventually kill one or more cells & that will overcharge the rest of them

[jrling]

Interesting discussion.

I suppose that in my case, I started off with all the cells within 0.01V of each other and I always have the float charger turned on, even before I fire up the PC, and so they never get a chance to discharge below 3V. So the perfectly benign environment for keeping in balance - which they do.

The GWL 12v packs look excellent value for money, although not A123 cells of course, which could be better construction/longevity. The PCM module presumably does not balance the cells, but rather cuts in if any cell or cells go below a specified voltage in discharge. As with battery balancing modules, I can't help feeling batteries with a protection module, are potentially going to be worse than the purity of cells on their own. But that's a purist's gut feel rather than being based on any empirical evidence.

[jkeny]

Interesting discussion.

I suppose that in my case, I started off with all the cells within 0.01V of each other and I always have the float charger turned on, even before I fire up the PC, and so they never get a chance to discharge below 3V. So the perfectly benign environment for keeping in balance - which they do.

The GWL 12v packs look excellent value for money, although not A123 cells of course, which could be better construction/longevity. The PCM module presumably does not balance the cells, but rather cuts in if any cell or cells go below a specified voltage in discharge. As with battery balancing modules, I can't help feeling batteries with a protection module, are potentially going to be worse than the purity of cells on their own. But that's a purist's gut feel rather than being based on any empirical evidence.

You are correct- all PCMs that I know of use a chip to monitor voltage & when the target voltages are hit (min & max) turns off the conduction of a Mosfet on the ground leg of the circuit - in other words each the battery's current is being delivered/returned through this mosfet which, depending on the mosfet used, changes the power delivery characteristics of each battery & of the series of batteries, as a whole.

However that is a good price for 4 LiFePO4 batteries - the PCM can be removed

[jrling]

As this is the place for cutting edge battery PSU discussion, has anyone tried Lithium Titanate batteries?

For 5V battery PSU, I am a little wary of float charging LiFePO4 at 2.5v, which is really near the bottom of their recommended working range. Apart from the obvious danger of running them down to a terminally low voltage and damaging them, I would guess they are not working optimally at 2.5V and could be 'noisier' and internal resistance may rise, thus defeating the instant reaction to current draw that is the theory of why they work so well in our type of digital equipment.

LTO batteries have a nominal working voltage of 2.3V and therefore would be in the sweet spot at 2.5V and doubled up in a series pair. They also have unbelievably low ESR of <1mOhm; the very expensive Toshiba SCiB LTOs are specced at 0.32mOhm! See here - http://www.ebay.co.uk/itm/NEW-Lot-6pcs- ... SwnF9Y7frb

Here is an example from nige's fave supplier -
https://www.ev-power.eu/LTO-technology/ ... -30AH.html

They do look like the answer to the maiden's prayers for 5V supplies; if only they weren't so expensive, but they should come down in price over time. Also I suppose you could say they are a steal compared to constructing a high-end linear PSU.

[jkeny]

Much cheaper here http://www.batteryspace.com/Lithium-Tit ... ttery.aspx

The snag is that it is categorised as class 9 shipping so only couriers will handle it & at a high cost of $70 - so battery price for 1300mAh 39Amp is $13 but shipping is $70 - it can only be ordered as sample so only one per shipping, I believe?

[jrling]

Yep. But EV Power do ship Class 9 'normally' and cost is here -

https://www.ev-power.eu/Frequently-aske ... _pricelist

Still expensive. But I wonder if the technology is worth the money. The specs would suggest it was. If I had a requirement for a 5V PSU, I would give it a go, but at present I don't. Will of course post my findings, if I do.

It would be great to hear from anyone who has used them.

Jonathan