Question on charging ebikes with server psu

bike4life

10 W
Joined
Apr 10, 2020
Messages
82
Ok so i have purchased 4, 12 volt server power supplys at 10 dollars each....

Now I have a 12s lifepo4 battery pack. lifepo4 reccomends 3.8 volts per cell x 12 = 45.5 volts....

Now problem is my 4 x 12 volt server psu in series makes around 50 volts....

Can I charge my lifepo4 batteries with a slightly higher voltage.... I make sure the cells dont go over 3.8 volts or what not, its just during charging the voltage is higher.... Normally chargers for a 12s lifepo4 are around 45 volts... Its it safe for it to charge voltage go over by 4-5 volts.

Thank you....

I tried hooking 3 x 12v psu with a boost converter to 48 volts except somehow my boost converter doesnt really work coupled with the server psu and my batteries dont charge at all :( Weird cause I measure 45 volts on the output of my boost converter but it just doesnt charge them up.
 
If you want LFP to have optimal lifespan, stop charging at 3.45Vpc, a little higher like 3.5V if high C-rate.

No need to hold CV stage, just "charge to" if a PSU can use HVC to automate.

3.6Vpc is "do not approach" maximum too stressful shortens life, does not add much range/capacity anyway.

So, 41-42V at pack level. Fine if the source is set to put out a little higher, just stop when the circuit hits 41V.

But really, you should be monitoring cell level voltage like a BT BMS

stop when the **first cell** hits 3.5V

The Meanwell HLG series is a good PSU to use, be sure to get the "42A" suffix, for adjustability.
 
john61ct said:
If you want LFP to have optimal lifespan, stop charging at 3.45Vpc, a little higher like 3.5V if high C-rate.

No need to hold CV stage, just "charge to" if a PSU can use HVC to automate.

3.6Vpc is "do not approach" maximum too stressful shortens life, does not add much range/capacity anyway.

So, 41-42V at pack level. Fine if the source is set to put out a little higher, just stop when the circuit hits 41V.

But really, you should be monitoring cell level voltage like a BT BMS

stop when the **first cell** hits 3.5V

The Meanwell HLG series is a good PSU to use, be sure to get the "42A" suffix, for adjustability.

do you know what the low voltage cut off for A123 cells are...
what is the lowest safe / good voltage i should draw these cells down to befote recharging
At 3.3 volts, what is the SOC of each cell... is 3.3 volts like 80% capacity...
 
With psu's you got to make sure the ground is isolated. Someone will chime in on this. I have two hp600 12.6v and two meanwell 24v turned up to 28v for 42v at 10amp for 12s lifepo4 A123 times two for 24s no BMS just me. I also got a boost charger of to 90volts I think 600watts you need a power supply I use a meanwell clone 12v. I have only used while watching.
I think with your 12 volt or 12. 6 volt psu you must check voltage.
I charge to 3.5v per cell .
 
bike4life said:
do you know what the low voltage cut off for A123 cells are...
what is the lowest safe / good voltage i should draw these cells down to befote recharging
That will vary a lot depending on loaded discharge current rate.

3.00V **at rest** (after isolating a few hours) will prevent "damage", but 3.10V or even 3.05V will drastically improve longevity - will lots of other interrelated factors.

> At 3.3 volts, what is the SOC of each cell... is 3.3 volts like 80% capacity...

No, at rest 3.33-3.35V is 100% Full.

~3.2V is 50% but only compared to say 2.5V as defining 0%, very life-shortening do not go near.

Get a coulomb-counting wattmeter and set up a constant current load at 0.1C. Start out at 100%, drawn down 10% of rated capacity or stop at 1hr whichever comes first, let rest isolated a few hours, measure voltage.

That's 90%

Rinse, repeat.

Soon as you hit 3.00V, recharge, and observe whether actual Ah was higher or lower than rated.

 
It's kind of hard to tell under load but not lower then 2.8v But mine is 20ah cells but really never got more than 18 ah . 7.5yrs old 32,000 miles same batteries.
Quality matters. Don't over discharge your battery how big of a controller (amps ) are you going to use ?
Look at these https://batteryhookup.com/products/100-brand-new-lifep04-32650-3-2v-6000mah-cells
I don't know anything about them but they look interesting. 18amp 6,000 mAh 3.99ea. USA.
 
yes excellent deal

may not last as long as those from the top makers, but brand new at a fraction of the price

doubtful wouldn't make it to 2000 cycles with reasonable care, 3-4000 with coddling
 
bike4life said:
Ok so i have purchased 4, 12 volt server power supplys at 10 dollars each....

It sounds like you know enough not only to set your battery on fire, but also your charging "rig". :shock:
 
Some psu's have isolated grounds, others have not. Not all PSU's can be used as a charger because you need CC and CV stages which ramps up the current and ramps up the voltage rather then just shotgunning the juice and :lowbatt: :kff:

Icecube57 did a lot of work and research and talks about the Dell server PSU's (NOT FOR CHARGING!) for powering RC chargers. He talks about how to determine if one is isolated gnd or not. How to reduce fan noise, also posted in Hobbyking forums.

Some Meanwell PSU's can be used for charging, model HRP are isolated and I used them in series which worked well.
Other models can be used, like LED psu models which tend to be lower amp ratings not sure which ones are isolated or not.

I use Dell laptop chargers for a split pack, 5S+5S which brings each 5S to 20.50V out of a posibble 21V (4.20V).
 
Isolated ground?? huh...

When I wired all them up I wired them to a plug without a ground... So its isolated i think cause I wired them all up with a 2 prong plug... I did not solder anything onto the ground...

Regarding chasis ground, I actually put wood shims on top of each psu and taped them all together so the case wont touch.
 
markz said:
Not all PSU's can be used as a charger because you need CC and CV stages which ramps up the current and ramps up the voltage
Actually only special more-expensive PSUs (like CC-only)

All standard normally priced PSUs and DC-DC converters deliver CC/CV

the key differentiating feature, critical for LI batteries, is effective Current Limiting (not just current protection)

Otherwise, the only job of their charge regulation is holding the max voltage (CV) setpoint.

The timing and SoC% level where the CC-CV transition takes place, are not determined by the source circuits, more by the battery chemistry, CAR spec, the "negotiated" current rate etc

And unless doing precise benchmarking or getting every last spec of Wh range crammed in

there really is no need to hold CV (Absorb stage) just "charge to xVolts and stop" CC-only, can get to 96-98% of max capacity anyway, and with better treatment for pack longevity.



 
Back
Top