DIY Load Testing Assembled Cell Packs

[Rawkkrawler]

Requesting some DIY advice from the experts. I asked a similar question already, and another member had some great responses. But I'm now looking to dig a bit deeper.

Purchased to identical scooters, both using 52 Volt 14S9P batteries. Both only charged to 55ish volts. A few packs were out of balance of course.
I was successfully able to get one battery to balance, and repeatably hold a balance across all 14 packs. Unfortunately the range is terrible, and i drop equally across all packs down to 55 volts. From the other member here, he suggested that i have a capacity issue. Battery is 2 years old, with 100 miles on the scooter itself.

So the question to you all, how would a DIY a load test on the assembled battery across each pack without removing the nickel strips? Using some electrical knowledge with ohms law, and 7 parallel batteries in a pack, at 2500ma each, it appears a 1 ohm resistor could "safely" drop the voltage so I can plot the decline of the pack. I guess what I'm looking for is a comparative chart across all 14 packs to find the smoking gun. Also, is it possible that one bad pack, or a cell within that pack, be effectively dragging the overall battery voltage (14 cell packs) down evenly? Or the obvious question, how does a pack that stays balanced within .02 volts drain so fast.

Anyway, that's where I'm at. Appreciate any assistance you provide.

Mike

 

[docw009]

You should have referenced your other thread, Mike. There wasn't enough info in this stand alone post.

52 Volt Battery Issue (14S9P)

Hope this lands in the correct sub thread... A little background. I've balanced several E-scooter batteries with success, but this latest issue is puzzling me. This battery comes from an Apollo Phantom V2 E-scooter. I recently purchased two as a package knowing the batteries weren't great...

endless-sphere.com

Still don't have enough info. What was the pack voltage when the scooter shut off? It should have been down in the low 40's. Are you saying it's 55V? Sounds wrong,

You could use a high wattage 1 ohm resistor to drain each group, It;s going to pull 4 amps at the start, which is 16 watts (I squared R) . Needs to be a 25 watt resistor, , If it is 9P, that's 18-27 AH, so it will take up to a day to drain the cell and you have to be there watching it with a stopwatch when it reaches cut-off at 3V. Otherwise, all you know is that it's drained, This is too tedious for me.

I would suggest you try to get a better number on the battery voltage when it cuts off, Can you get your Fluke on the battery outputs immediately? Maybe hang a voltage display on it while riding. If you know what that voltage is, then you can go further,

EDIT: Unless you come up against something mechanical like a broken spot weld that lifts off during riding, it's generally not worth fixing. Even if the cells could talk and tell you there's a bad one, can you replace it?. And if you did, there's probably another one waiting to fail.

 

[Rawkkrawler]

You should have referenced your other thread, Mike. There wasn't enough info in this stand alone post.

52 Volt Battery Issue (14S9P)

Hope this lands in the correct sub thread... A little background. I've balanced several E-scooter batteries with success, but this latest issue is puzzling me. This battery comes from an Apollo Phantom V2 E-scooter. I recently purchased two as a package knowing the batteries weren't great...

endless-sphere.com

Still don't have enough info. What was the pack voltage when the scooter shut off? It should have been down in the low 40's. Are you saying it's 55V? Sounds wrong,

You could use a high wattage 1 ohm resistor to drain each group, It;s going to pull 4 amps at the start, which is 16 watts (I squared R) . Needs to be a 25 watt resistor, , If it is 9P, that's 18-27 AH, so it will take up to a day to drain the cell and you have to be there watching it with a stopwatch when it reaches cut-off at 3V. Otherwise, all you know is that it's drained, This is too tedious for me.

I would suggest you try to get a better number on the battery voltage when it cuts off, Can you get your Fluke on the battery outputs immediately? Maybe hang a voltage display on it while riding. If you know what that voltage is, then you can go further,

EDIT: Unless you come up against something mechanical like a broken spot weld that lifts off during riding, it's generally not worth fixing. Even if the cells could talk and tell you there's a bad one, can you replace it?. And if you did, there's probably another one waiting to fail.

Thanks for the reply, and here's the additional info...

The two scooters are Apollo Phantom V2's, each with a 52 volt (14S9P) battery
When i purchased then second hand, they would only charge to 55ish volts
I was successfully able to balance one battery so far to reach the 58.8 volts
After a short ride, the scooter shut off. The measured voltage with the Fluke on the physical battery was 54.3 volts
Each cell pack was at 3.86 +/- .02 volts
I charged it again to 58.8 volts, another short ride, same result, scooter shut off within 6-8 miles
Pulled battery once again, 54.3 volts and balanced. Topped off to 58.5 volts 2 days ago, still holding at 58.5 volts

I am currently dealing with a scooter specific issue that i also can't answer, and may explain why the scooter shuts off at 54 volts. The onboard voltmeter reads exactly one volt (57.8) at the display when fully charged to 58.8 volts. The scooter itself is designed to shut off at 44 volts. The display reads 48 volts when it shuts off completely. Not sure what's happening there.

I ordered a MakerHawk capacity tester to automate a pack specific test. I would rather not remove the Nickel strips if i can avoid it. Happy to rebuild the pack if needed as i do have the spot welder and other supplies.

 

[docw009]

The Makerhawk load testers are made by Atorch and sell for a bit less on aliexpress. If you have the version with the dial for setting current, don't change the value when it's on load.Adjust when the load is off. Mine glitched and blew up, I replaced with the model with color display and that's been reliable, except when I shorted out the exposed traces on the back.

I use the Atorch to load test my packs and it's very reproducible. You could test each individual group at 5 amps. putting alligator clips on the nickel. One of them will probably cut out around 3.9V,.

 

[Gruesome]

I think the first step is to actually measure the charge or discharge capacity. Charge might be easier than discharge, since you don't need a high power discharge load, just a charger. (If you don't have a high power resistor, you could use a sufficiently large light bulb, e.g. a 150W 120V bulb will yield about 0.7A at 50V for discharge; but you definitely have a charger, so that's easier.) You do need a current meter. clamp-on style current meters are pretty convenient; all you need to do is separate the + and - charging leads enough so that you can get the clamp around one of them. You don't need to track the whole curve of voltage vs charge (charge is time times current, so 2A for 3 hours is 6Ah, etc.) , but you do need a few points in time. From the slope of that charge curve you can compute the total capacity of your battery pack. (E.g. say you track the charging of a 14s battery; if it takes 5 hours at 2.0A = 10Ah to get from 45V to 55V, and the full charge and empty voltages are 4.2V and 3.0V per cell, then the total capacity would be 14*(4.2-3.0)/(55-45)*10Ah=16.8 Ah in this example.
Correction: At the highest and lowest voltages the capacity is not a linear function of the voltage, so this procedure gives an overestimate of the battery charge capacity.)

If you keep track of the individual parallel section voltages (you call them 'packs', it seems), you can compute the capacity for each section/pack/group, and find the bad one. You don't have to repeat the charging exercise for each section separately. Just write down all the section voltages every hour or two.

Sorry if this is too basic.

 

[Rawkkrawler]

I think the first step is to actually measure the charge or discharge capacity. Charge might be easier than discharge, since you don't need a high power discharge load, just a charger. (If you don't have a high power resistor, you could use a sufficiently large light bulb, e.g. a 150W 120V bulb will yield about 0.7A at 50V for discharge; but you definitely have a charger, so that's easier.) You do need a current meter. clamp-on style current meters are pretty convenient; all you need to do is separate the + and - charging leads enough so that you can get the clamp around one of them. You don't need to track the whole curve of voltage vs charge (charge is time times current, so 2A for 3 hours is 6Ah, etc.) , but you do need a few points in time. From the slope of that charge curve you can compute the total capacity of your battery pack. (E.g. say you track the charging of a 14s battery; if it takes 5 hours at 2.0A = 10Ah to get from 45V to 55V, and the full charge and empty voltages are 4.2V and 3.0V per cell, then the total capacity would be 14*(4.2-3.0)/(55-45)*10Ah=16.8 Ah in this example.)

If you keep track of the individual parallel section voltages (you call them 'packs', it seems), you can compute the capacity for each section/pack/group, and find the bad one. You don't have to repeat the charging exercise for each section separately. Just write down all the section voltages every hour or two.

Sorry if this is too basic.

Basic is good! I started balancing batteries last year on a different scooter battery, and learned quite a bit. However I'm now realizing that I've only learned half of what I really need to know. Voltage is actually easy when it comes to balancing across the 14 packs/sections. Capacity is the other important part. This is evident when i test ride a "fully charged, balanced" battery pack and it dies after 1/3 of it's advertised range. If the batteries are degraded, range doesn't exist.

This is what i purchased to do the testing, i believe it'll help..

https://www.amazon.com/dp/B0BK8YB2F4?ref=ppx_yo2ov_dt_b_product_details&th=1

Adjustable Constant Current Electronic Load - 150W USB Electronic Load Battery Capacity Tester USB Resistor Load Multimeter Voltage Current Tester Voltmeter Meter Amperage Reader 0.2-25A 2-200V

I anticipate that I'm going to find several, if not all cells to be severely degraded.

The plus side, I have two 14S9P batteris as well as a 14S7P battery. All from Apollo Scooters, all the same 18560 2500 mAh batteries. If i break down the packs, i'll have 350 cells. Maybe if i test each i'll end up with enough to build a reliable pack. Or, i just spend the money on new Samsung 25R's and build new.

While i'm in this rabbit hole, I still would love to which BMS i should be using. On this site i've heard about Daly and ANT, but i'd like to just keep it simple and find one that works good enough. Love to hear anyones thougths on that as well. And preferably one with a harness. For some reason it's been impossible to find a 15 pin mini harness. I tried Digikey, hard to navigate the site though.

Thanks as always!

 

[Gruesome]

That's fancy. It does all the current integration bookkeeping for you. You'd still need to track the individual parallel section voltages by hand, but that should not be a problem, unless the discharge happens way too fast. But your packs are about 1kWh or larger, so at 150W maximum load power you have several hours to occasionally go through the parallel sections and write down the individual voltages. That should show whether there are outliers.