Manual balance discharge device?

B

Barncat

1 kW
Joined
Oct 26, 2020
Messages
368
Sorry for asking as this has no doubt already been addressed (can't find) but what would be the best cheap and easy resistive device to put between alligator clips to manually bleed charge off, say, a 5p 18650 cell group? I'd think 5A would certainly be a safe current draw and my Imax B6ac is limited to 2A. Very slow.... Probably a 15 or 20W incandescent bulb?

I'm still testing my no-weld 20s5p design. Intentionally going without BMS. Going great so far and pack is pretty well balanced but need to quickly knock .15V off a couple groups to get whole pack to full charge.

And on a related note, is it pretty common for the most negative and most positive cell groups to tend towards a slightly higher V during charging? I'm using an 84v 3A Yarlan charger btw... probably getting a 5A version soon...
 
Barncat said:
Sorry for asking as this has no doubt already been addressed (can't find) but what would be the best cheap and easy resistive device to put between alligator clips to manually bleed charge off, say, a 5p 18650 cell group? I'd think 5A would certainly be a safe current draw and my Imax B6ac is limited to 2A. Very slow.... Probably a 15 or 20W incandescent bulb?
Click to expand...
What voltage is the bulb intended for? It won't draw it's rated wattage unless it's used at that voltage. ;)

One thread that may help with ideas:
https://endless-sphere.com/forums/viewtopic.php?f=14&t=8320
 
Barncat said:
And on a related note, is it pretty common for the most negative and most positive cell groups to tend towards a slightly higher V during charging?
Click to expand...
THis is sometimes called teh "end cell" effect:
https://endless-sphere.com/forums/search.php?keywords=%22end+cell%22+effect&terms=all&author=&sc=1&sf=all&sr=posts&sk=t&sd=d&st=0&ch=300&t=0&submit=Search
there's other info out there besides those, possibly under other names I can't recall ATM. :oops:
 
Hey amberwolf- thanks again for the prompt assistance. Unlikely I'd have found that thread on my own. So it figures that the matter at hand is not so simple given the nominal 4v supply... will study that info. This procedure would be closely monitored if I try it. In the interim I did see a recommendation elsewhere about using a 12v mini halogen bulb as a discharger for lipo packs. Must read up...

The "end cell effect" makes intuitive sense so I'm not surprised there. Now I know.
 
Oh, and related to the end cell effect I've noticed that the V of those groups very quickly settles by .06-.08 volt or so as soon as it comes off the charger. Should I be governed by the the on-charger V dmm reading or the near immediate off-charger V reading as to avoiding overcharge damage? Or split the difference? I used 10ga stranded copper wire approx 6" ea for the plus and minus leads btw.
 
I don't have any experience with this randomly-come-across product***, but it might help automate your manual process. ;)
https://www.amazon.com/HiLetgo-Capacity-Discharge-Analyzer-1-5V-12V/dp/B0716ZL1SM
Power supply voltage: DC4.5-6V (micro USB interface) ; Working current: less than 70mA. Supported by current: the maximum 3.000A resolution 0.001A.
The maximum measurement error voltage: 1%+0.03V ; The maximum measurement error of current: 2%+-0.010A.
The discharge voltage: 1.00V-15.00V resolution 0.01V ; Termination voltage range: 0.5-11.0V.
Board size: 50mmX34mm ; Product size: 50mmX36mmX17mm (length X width X high maximum position size, containing copper foot height)
Note: this circuit in order to improve the voltage measurement accuracy, specially designed DC bias.when the terminal is what all don't pick will display a small voltage, does not affect the actual measurement.if you are short connected with the input terminal (absolute 0V) will display 0. To understand the principle can consult the superposition theorem in electrical engineering.

Product description
Instructions:
1. The first test should be fully charged battery.
2. Connect the battery to be tested positive to positive input, a negative input to the negative, can not be reversed (with load reversal may damage the circuit)! Load connected to the output of the positive and negative output, to work through the tester micro USB power supply (not available desktop or laptop USB powered), then the battery voltage.
3. Direct start test need only one press of the "OK" button, the tester can automatically develop an appropriate termination voltage of the battery full charge voltage, and will start flashing after the test three times. Need to develop artificial termination voltage only when the battery voltage display state by pressing "+" or "-" buttons to modify the termination voltage display beginning with P, behind the representative voltage resolution 0.1V, set up after pressing the "OK "start testing.
4. After the start of the test the tester will control the load of the electronic switch is turned on, the test data shows that the process will release capacity (Ah), current discharge current (A) and battery voltage (V) between the wheel was. When the battery voltage reaches the set cut-off voltage, load control switches off the tester display data stays in capacity (Ah) and above and the corresponding indicator flashes quickly together, now displays the actual capacity of the battery is discharging capacity, about the press "OK" to terminate flashing allows stable data display, press again "OK" button to return to the power-on state can just replace the battery test the next section ......

Error codes and their meanings:
Err1: battery voltage is above 15V
Err2: battery voltage is lower than the termination voltage
Err3: Battery can not afford to load or discharge current line too much resistance
Err4: overcurrent (current exceeds 3.1A)
Click to expand...

There are passive and active loads you can buy for these types of things, too. Random finds
https://www.amazon.com/3-7V-21V-Electronic-Intelligent-Constant-Discharge/dp/B01LKT0OTQ/
https://www.amazon.com/Electronic-Adjustable-Constant-Current-Capacity/dp/B083V42CTY


Then there's this one found randomly
https://www.amazon.com/PEMENOL-Electronic-Adjustable-Constant-Discharger/dp/B07W3TCRYL
This USB load tester is designed with independent power supply. The product measurement circuit and display circuit do not consume the energy of the test end, making the measurement data more accurate and reliable.
LCD display: can display load input voltage, input current, input power, discharge time, discharge mAh, product temperature and other information.
Intelligent cooling fan (automatically starts when the load discharge power is greater than 10W or the load temperature is greater than 40°C), All aluminum heat sink.
Protection Function: OVP-overvoltage protection; OCP-overcurrent protection; OPP-overpower protection; OTP-overtemperature protection; LVP-undervoltage protection; automatic data storage when power is off.
Application: can be applied to various power (charger, power Bank) aging or functional tests, test battery capacity discharge characteristics, as a low power load. ❤️❤️❤️ For detailed instructions, click the link: https://m.media-amazon.com/images/I/A1Gx5eT1LpL.pdf
Click to expand...

***I have bought a few things from the seller, and they showed up as described and quickly, and they're cheap. Have only used a couple of the things bought so far (switch and USB serial) and both worked. Figured I could start simple and work my way up. ;)
 
You can also insert pins into the lipo cells BALANCE LEADS and use the discharge feature of the rc charger.
If you have a pack of #S lipo, you can insert pins into the balance charge/discharge leads.
This could also speed up overall balancing time if your cells are to far apart in voltage.
ES member Icecube57 did a video on that on his you tube channel.
 
markz said:
You can also insert pins into the lipo cells BALANCE LEADS and use the discharge feature of the rc charger.
If you have a pack of #S lipo, you can insert pins into the balance charge/discharge leads.
This could also speed up overall balancing time if your cells are to far apart in voltage.
ES member Icecube57 did a video on that on his you tube channel.
Click to expand...

I don't see how that applies to the OP's "5p 18650 cell group" that he's working with? (especially since he is not using an RC charger, just a common type of bulk charger)
 
Barncat said:
Oh, and related to the end cell effect I've noticed that the V of those groups very quickly settles by half a volt or so as soon as it comes off the charger. Should I be governed by the the on-charger V dmm reading or the near immediate off-charger V reading as to avoiding overcharge damage?
Click to expand...

Really need to clarify your goal here.

Charging regulation

with an unbalanced pack, or at least one not top-balanced

requires measuring per-cell "live" voltage, you stop based on the first cell hitting your selected setpoint to prevent damage to that cell.

The overall pack stop-charge setpoint can be much lower, should be in fact for everyday cycling, for good longevity.

So at a high C-rate,

20S could be set to stop when the pack reaches 81V

with the per-cell setpoint at 4.25

while a low C-rate you could use 83V and 4.15Vpc.

But I thought this thread was about discharging?

For the purpose of balancing? would imply you wanted a bottom-balanced or mid-point balanced result?

At-rest recovery voltages are measured after at least an hour, cells isolated from any circuitry.

To estimate SoC, or see the benchmarked result of your charging or discharging protocol.
 
This thread is about manual balancing. Firstly how to bleed a couple high groups. Second, which voltage I should trust on end cell groups- on charge or just off charger.

I'm working on it. Need a clamp meter.
 
Use the live charging readout to prevent damage to cells out of balance at the top.

Trust the resting voltage to see if your balance delta is "low enough", whether using top-balancing, bottom- or midpoint balancing

some say 30mV, others go all the way to 10 or even only 5mV.

If the method used is inconveniently long or stressful to the cells, then allowing a higher delta may be called for.

That however may require performing it more frequently

 
Thanks John. My instinct was to respect the higher on-charge readout. I'm monitoring this battery very closely as it's a work in progress. And it's 100 cells so not something to trifle with.
 
And amberwolf- thanks for the research on that $20 load tester with the fan. It's limited to 5A draw down which may actually be pretty slow but I'll maybe order one anyway. Unless I'm badly mistaken my 3000mah Panasonic cells at 5p could be safely discharged @ 15A -20A as occasionally needed.
 
You can install balance leads on a 18650 pack if your building your own pack, or make it easier to disconnect the BMS leads.

There are also other threads on alternative methods of discharging, some have used heaters for whole packs, others oven elements, power resistors with cooling, light bulbs as mentioned.
 
Barncat said:
And amberwolf- thanks for the research on that $20 load tester with the fan. It's limited to 5A draw down which may actually be pretty slow but I'll maybe order one anyway. Unless I'm badly mistaken my 3000mah Panasonic cells at 5p could be safely discharged @ 15A -20A as occasionally needed.
Click to expand...

There are much better ones out there if you have the time to poke around, and the money to spend on them. Some are passive loads and some are active--the active ones have much higher load potential, as they are probably PWMing (or even linearly activating) some FETs in a way that puts it in it's resistive mode (rather than it's switching mode), which generates a lot of heat and wastes a lot of power, and the FETs are probably directly placed across the battery output (via a current measuring device of one kind or another).
 
We'll see. That little device and an ac/dc clamp meter will be here tomorrow. Will experiment. I'm no EE but picking up the basics fast.

I'm actually surprised that 16 of the 20 cell groups are self balanced to within .01v after 5 discharge cycles. I've not run them to below 3.3v/group, though specs say it's safe (not advisable) to bottom out at 2.75v for the LG's and 2.5v for Panasonics...

Will be disassembling the entire pack again soon for examination. Very easy to do with my design- nobody else can say that.
 
Barncat said:
I'm still testing my no-weld 20s5p design. Intentionally going without BMS. Going great so far and pack is pretty well balanced but need to quickly knock .15V off a couple groups to get whole pack to full charge.
Click to expand...
Why do you need FULL charge especially when using two different name brand salvaged cells of questionable quality (Panasonic NCR18650A 3100mAh 6.2A and LG INR18650D2 2950mAh, 5.9A)? Your no-weld 20s5p pack design should provide you more cycle life charged to 4.11V instead to 4.2V (FULL).

Isn't the advantage of your no-weld pack design such that you can rearrange/exchange a few of the cells in your p-groups so the voltage variance (low/high) is as low as 0.07volts (e.g 4.06V to 4.13V) or isn't that possible? Even though the best arrangement of the twenty p-groups is a resting voltage variance of only 0.07V BE AWARE! Five cells of questionable quality in just one p-group could have a resting variance of 0.30V (e.g. 3.75V to 3.45V); especially when using salvaged cells of two different cell brands ... even if both have similar ratings.

Bottom balancing the 5 p-groups before bulk charging and again top balancing (if necessary) is time consuming. Necessary with a pack of used cells with two different cell ratings and chemistry variation (Panasonic NCR18650A (3100mAh 6.2A) and LG INR18650D2 (2950mAh, 5.9A).

What was your method for determining what cells (brand intermixing) to put in each p-group so as to balance the twenty p-groups as best as possible? With p-groups variance of 0.15V (used salvaged cells) isn't a BMS a must if for no other reason then protection (BPS) to shut the pack down should there be an internal short in one of the p-groups :bolt:
 
It was not possible to get most of the pack to over 4.0v because a few groups were at 4.2. So I was leaving some power on the table as it were. Still working on it. A 50W power resistor only pulls 2.3A clipped on a group. Too slow. Still waiting for other testing parts.

I'd have preferred to use identical brand cells... 75 Panasonics and 25 LG'S. Not intermixed in groups. Will get a few more Panasonics soon to make pack uniform.

Will likely inspect cells individually tomorrow as part of my monitoring process. If a cell shorts you're screwed anyway. I'm not convinced that a bms is entirely necessary, yet.
 
Barncat said:
A 50W power resistor only pulls 2.3A clipped on a group.
Click to expand...
At what resistance?

Assuming start of 4.2v and no significant voltage sag (shouldn't be at that current draw with a parallel group of cells), and a current draw of 2.3A, that's only about 9w its' actually outputting, and presumably is too large a resistance for your needs at about 2ohms.

You can parallel two of them to double the current and power. Three to triple it.

2.3A means it will remove 2.3Ah from a group in one hour, assuming it remains constant. How large a capacity is the group in question? I'd guess it's at least 10-20Ah if it's really going to take that long to bring 4.2v down to 4v.
 
i use a 12v bulb on 4p of 25r, its pretty quick

YxxChuZ.jpg
 
Uniform cells are brand new, all from the same production run.

Mixing cells of different wear levels is just as uneven as different brands / models / chemistries.

Would take regular capacity testing of cells individually, and a strong high-current balancer,

IOW a lot more ongoing maintenance headaches

to make such a setup perform nearly as well as a proper pack of matching new cells.

 
Right. Failed to specify resistance. 50W2RJ which apparently is just 2ohm. Across 5p. I could try them in parallel. I have that 5A load tester but the 5v low A power adapter hasn't arrived.

I'll try out the light bulb goatman.

I was under the impression that the overwhelming majority of DIY battery builders use used cells. New are 4x the price on average. It may be false economy however given the enormous hassles. Can new cells from a reputable supplier trusted as perfect?

You'd like to think playing around with ebikes would be a cheap hobby but it ain't. And they are just barely legal toys that can only operate and compete with low speed urban traffic. Still fun, but I've rebuilt actual cafe racers for less $$....
 
i use salvaged cells, no bms
you could bottom balance your cells to 3.0v then charge to what would be 4.10v/p-group

13s@4.1v would be 53.3v

when the pack is fully charged at 53.3v the top end voltages might be slightly different like a 0.15v spread

write the voltages down and use it for a week and recheck the voltages, they will probably be the same

the only time i manually balance my pack is if i notice that the fully charged resting voltage changes

if after 4 months the voltage drops from 53.3v to 53v fully charged at rest or im seeing a bit more volt sag than usual ill check my pack and burn off some volts, its usually the pack end groups,

heres me checking 3 of my packs after 4? months you can see i need to burn off some voltage on the ends
(3p pack wasnt fully charged, its my range extender/back pack battery, it was sitting on the shelf not fully charged so i was just checking it)

gFR9txV.jpg


if i charged to 4.2v
the pack would probably go out of balance in 2 weeks
 
You must log in or register to reply here.

Similar threads

nutnspecial
Parts/build list
Replies
6
Views
2,503
nutnspecial
nutnspecial
Back
Top