Warning for those who bulk charge using bench power supply

pullin-gs

1 kW
Joined
Jul 31, 2009
Messages
395
I have been bulk charging lithium type packs using a lab switching 5A/60V PS for going on 6 months now....for 12 years before that I used a linear 3A/30V lab PS without issue.
However yesterday my 5A/60V bench power supply shot up to maximum pre-regulated DC voltage output @maximum amps without warning.
Fortunately I had just connected up my battery and noticed voltage was climbing fast due higher charge current....as soon as I disconnected the pack, the PS voltage reading shot up to over 80V (and climbing). Within seconds an output capacitor exploded before I could pull the plug.
Repairs involved failure of final-stage MOSFET (gate/source/drain short) and capacitor...the pair of primary transformer winding switching MOSFETS were fine.
Power supply is working again....will not use it anymore though. Will invest in better quality one (suggestions that wont break bank?).
Anyway, if failure had occurred while I was not there I would probably be talking to my insurance agent regarding fire right now.
PS: Pack being charged at time did have BMS, but my other pack does not have BMS.....this could have been very bad outcome.

v/r,

P
 
I'd say a proper charger or at minimum a LED driver that does CC/CV

Meanwells are great for that task. I like the potted units as they are very robust and can travel. Some are good enough for bike mounting. One that would put out a comparable wattage to your 5A/60V setup should run under a hundred.

These are good up to about 14S (the 54V unit will reach 58V):

https://www.meanwell.com/webapp/product/search.aspx?prod=HLG-320H
 
Any hardware can and will fail.

Unlikely to actually cause a fire, but. . .

Whether or not you have "a BMS" a high voltage cutoff of some sort will give better piece of mind.

But especially if your charge sources aren't completely automated, keeping an eye on things is a good idea
 
That kind of failure could've happened with any PSU or charger, depending on it's output stage design (if it has any built-in "fuse" parts, for instance, that will blow and break the output circuit away from the innards).
 
thats the worstcase scenario I worry about.

I use these cheap overvoltage relays(5 dollars) on all my lithiums. The one I use are 0-40 volts but you can buy 0-99 volts (about 9 dollars). They have small relays built-in(10 amps), I use those relays to trigger a larger 30 amp automotive relay.

Install the larger relay between charger and battery, if the battery voltage climbs above a certain limit it pulls the relay. This will protect against bms or charger failure. These are very solid protection that you can leave connected all the time, I've had one running 24/7 almost 2 years. These overvoltage relay will work with any charge source, power supply,solar etc.

This monitors overall battery voltage, it won't protect if your battery goes severely out of balance. They do sell overvoltage alarms, but you would have to be in the area to hear it go off.

This is like a deadman switch in case everything fails.

a overvoltage relay.jpg
 
jonyjoe your solution is a good one, and something I was going to mention. In the aerospace sector (where I spent my career) we always did FMECA (Failure Modes Effects and Critically Analysis) and used concepts of single failure tolerant and two failure tolerant to protect life threatening critical faults.

In the most basic mode, taken at the box level, the power supply could fail high or off. If current can't feed back from batteries into a dead power supply, that would just be a "failure to charge" no biggie. Failing high would be a catastrophic failure with potential loss of facility or life due to fire.

We would then make the charging system one or two failure tolerant. Your over voltage cutout circuit would make things single failure tolerant. Add another such circuit in series and you get two failure tolerant. We would make the second over voltage cutout circuit a different design, likely from a different vendor, so that there were no common failure points/parts in the two over voltage detection circuits.

BTW Sorensen makes good power supplies. I have filled my lab with old, commercial quality stuff off eBay. My 3 Sorensen power supplies all have a separate, built in, over voltage protection circuit that is user settable. Therefore a Sorensen, properly set is single failure tolerant (sort of) in and of itself.

Thanks for sharing your near miss. Much can be learned!
 
A good BMS would prevent overcharging to a point of fire but at that high a voltage anything could fail. Chargers can fail which is why you should not leave a charging battery unattended unless it's in a fireproof area and you are nearby. The problem is that charging often takes a while so most of us do leave it unattended.
 
jonyjoe303 said:
thats the worstcase scenario I worry about.

I use these cheap overvoltage relays(5 dollars) on all my lithiums.
Excellent.....I found a 60V one for $8.
 
Volt_Ampere said:
A good BMS would prevent overcharging to a point of fire but at that high a voltage anything could fail.

This is a good point. Any protection device needs to be capable of handling the maximum fault voltage the charger could output, or at least fail safe. Bear in mind MOSFETs often fail closed. Max voltage probably equivalent to the peak rectified mains input.
 
Apparently these are very robust, no external contactor required, can be programmed and monitored from a PC.

https://www.ebay.com/sch/i.html?sid=sbs.solutions.sales.usa4&isRefine=true&_sop=16

Using the cheap Chinese versions, I'd probably use two at a time for failsafe backup.
 
Back
Top