Testing a 10S 36 Volt BMS

[newb123]

A month ago, I carelessly plugged my charger into the load port, and I used the wrong charger. Battery was destroyed, and would surely have escalated into one of those youtube fires you see when they overcharge a cell. It was still pretty hot though, and I had to toss it in a bucket of water.

I use a small battery insulated clip now to hook up to C- for charging so it forces me not to cross connections with the P-

curious, why would a cross connect to P- blow up the cells?

I see how the li ion charging cycle works now, he explains it well @ 14.00

https://www.youtube.com/watch?v=A6mKd5_-abk

 

[docw009]

A month ago, I carelessly plugged my charger into the load port, and I used the wrong charger. Battery was destroyed, and would surely have escalated into one of those youtube fires you see when they overcharge a cell. It was still pretty hot though, and I had to toss it in a bucket of water.

curious, why would a cross connect to P- blow up the cells?

I grabbed a 48V charger instead of the 36V charger. No over protection (or balance) with P-. Glad to hear you will using C-. Picture suggested the solder tab had never been used,

 

[newb123]

Yes I charged with the boost converter and started at .5 amps at battery level 35 volts, went for about 4 hours to 42 volts 0 amps, seemed to work great

 

[newb123]

for the low end of the charge cycle...I see where the pack should not be left charging overnight, approx. 4 hours is max. time. when the CC gets to a percentage of the charging current that was initially set

If the CC goes to zero...does this damage the cells?

 

[john61ct]

First off, understand that the transition from Bulk/CC to Absorb/CV is not determined by any charge source control circuitry,

all that does is hold current to the max Amps setpoint (if current limited) and then Voltage to your max Absorb setpoint.

The battery CAR/chemistry determines the timing, and the Bulk current rate,

higher the amps, lower the SoC at the transition point.

Got that clear in your head?

OK, so at some point, as SoC climbs toward 100% Full, Absorb V is reached.

If the current rate was a low C-rate, that is "full enough" for normal cycling, Just Stop there, CC-only charge profile, hardly any usable additional Ah can be added, and longevity is harmed by pushing upward in CV territory.

Unless the balancing circuitry design requires holding above a certain voltage for some length of time.

Poor design IMO.

But, if the charge rate was high, say over 0.3C, so you only got to say 90% SoC at the CC-CV transition point and you want full range, or

suppose you want to go to "vendor spec 100% max Full", say a maintenance protocol, or benchmarking capacity SoH via CC load discharge testing.

So you hold Absorb stage, the charge source keeps Volts from climbing, and amps are declining as SoC and battery resistance climb, CAR falls.

The endAmps spec where the maker defines the stop charge point might be 0.005C

Personally I use a gentler 0.05 - 0.02C setpoint, but some even hold all the way to Zero current flowing.

So compare:

Charge to 4.05V and hold until current falls to 0.05C

vs 4.2V (4.35V for HV cells), hold to 0.005C or zero

The difference is not immediate "damage", so much as lost longevity cycles off the back end.

Too many other relevant variables to quantify, but for many chemistries, not at all insignificant.

 

[newb123]

Ok I will go with the 4.2V at .0005C

 

[john61ct]

As long as you understand that is the choice that reduces battery longevity by pushing to (too much) higher SoC.

My point was that that is a mistake, without a known-certain and compelling reason to do so.

But, your rig your choice

 

[newb123]

on the panasonic CGR26650A (im using B) spec sheet it has the CC Max at 8.80 Amps. the capacity current @ 2650 mAh. So when charging using the CC/CV boost converter, I should set the CC at 1/2 the value or less of 2650 mAH? about 1 amp?

CAPACITY Capacity, max. 2650.00 mAh
CURRENT Constant charge current, max. 8.80 A
Max. continuous discharge current 40.00 A
Peak discharge theoretic, 4 sec 70 A
C-rate (charge, max.) 3.32 C
C-rate (discharge, max.) 15.09 C
POWER Watts (charge, max.) 31.68 W
Watts (discharge, max.) 144 W
ENERGY Energy, max. 9.54 Wh
Density volumetric theoretic 265 Wh/L
Density gravimetric theoretic 106.00 Wh/kg
E-rate (charge, max.) 3.32 E
E-rate (discharge, max.) 15.09

 

[newb123]

I noticed that after several hours, and the pack charged to 42 V 0 amps, the BMS would be very warm (its heat sink) is this normal and if so what is it doing?

 

[dustNbone]

That's basically what will happen when it's balancing. Worst case scenario is when you have one low cell (relative to the others), the BMS will be bleeding power off from the 9 fully charged cells to prevent them overcharging, while allowing the remaining one to continue charging.

That excess energy is dissipated by bleed resistors, one (at least) for each cell group, and they convert it into heat, which is transferred to the air via the heat sink.

In the opposite situation of one high (fully charged) cell and the rest still charging, you'll only have one bleed resistor active and therefore far less heat being dissipated.

So yes, the warm heatsink is normal. There's an upper limit to how much heat it will need to dissipate (based on ohms law, volts, resistance and all that), and the system should be designed to handle that.

 

[john61ct]

Yes, the ability to balance at a higher current rate adds to the cost, so cheap ones can be 300mA or less, very very slow to correct higher imbalances.

on the panasonic CGR26650A (im using B) spec sheet it has the CC Max at 8.80 Amps. the capacity current @ 2650 mAh. So when charging using the CC/CV boost converter, I should set the CC at 1/2 the value or less of 2650 mAH? about 1 amp?

No. First off, "capacity current" is not a thing.

And you should not keep overusing CC for current in general, only when you are clarifying Bulk stage, as opposed to Absorb (CV).

Also every boost converter is CC/CV, that label is redundant as well.

So, the cell capacity is rated 2650mAh.

A conservative rate for longevity would be 0.2-3C, or 500-800mA. At 0.5C 1100-1300mA.

8800mA being the marketing do-not-exceed mfg spec, not at all a good idea unless you're in a real hurry and have overtemp protection.

 

[newb123]

Thanks for that

separate topic but...

I see different opinions and debates on soldering wires to 18650 cells. If you use flux, wide tip iron and wet the iron and wire tips good, is there any damage to the cell? I started with a flux smear on the cell, then wet the iron with a blob, touch for about 1 sec or so to the cell, move to the next cell, same thing, then back to the blobbed cell, build up a bigger spot, then wire to that

it doesnt seem to get that warm, dont know though

 

[john61ct]

Change of topic that radical, best to start a new thread to keep this one more useful as a reference for others.

Or resurrect the discussion after reading through the **many** existing threads where this topic's already been very well covered

https://www.google.com/search?q=site%3Aendless-sphere.com+soldering+18650+cells

 

[newb123]

ok, will do, but from what I gather from views...its not conclusive as to if the intermittent heat damages or not, coming from the iron

I got some 18650 cells in better shape, all at 4 volts at the same voltage, so I soldered 10 pairs (20) in parrallel - series

will charge with the BMS and boost, report results. the range of the older green Panasonic ones was just too crappy and some of the cells were at like 3 volts when others were at 3.5 after riding so they were shot I guess. They would charge and balance at 4.1 but a string of 10 in series at 41 volts would only last 20 min or so.

 

[newb123]

I finished soldering the pack in parallel series config, no more BMS cut outs when riding because after I drained the batteries from 41.5 to 37 volts, all cells were at 3.6 volts, (my DMM only reads out to 3 places) so they seem to be in better condition that the others, but the charging is very slow; Im guessing its because there are 20 cells now in series parallel.

Is there a way to guestimate range based on motor wattage, amp draw, voltage, cell amp hour etc?