Ah... best to work with a consistent wait time. 7-9 days is fine. Try to estimate the wait time for each module. Read voltage immediately prior to the post-sit discharge test.
Yeah. I feel your pain on the safety triggers. My hardware halts the cycle if a safety is hit.
Since you already have the initial data (I'll check the spreadsheet shortly), I would:
set sensitivity to the lowest value available, 5mV or below.
Increase charge current to 4A
If possible, set a 20 minute wait between cycles
Devise a means of circulating air through the pack. Please refer to the following:
If you can suck or blow into either fan opening while blocking off the other, you can get air to flow through the pack. Anything that results in perceptible flow through the pack is acceptable - even if horribly crude. If you have a carpet dryer, you could force it into the black opening, block off the circular opening and force air up through the pack.
Alternatively, if you can suspend a box fan over the pack blowing into it while blocking off the circular fan opening, you can probably get some cooling.
I've made some notes and comments in the sheet/chat. Populate to CHG6 and then evaluate.
I spent some time rigging up cooling. Learned something about the system.
The GM fan for the battery only pulls air from the top of the pack and exhausts out the rear ducting. Blowing into the fan ducting does nothing. Learned this once I removed the assembly.
But blowing into the air inlet on the bottom/side of the battery nicely pushes a ton of air between all the modules. You can feel it coming out with your hand.
I had a bathroom exhaust fan from a reno that I rigged in as a low cost/fairly efficient fan that even has a 110 plug on it. I was fairly pleased.
Good work. It's been a long time since I had one of those cases open, and I see my mistake.
Please confirm if you had a carpet dryer like this:
you could just aim it into the vent at the end?
Sorry for steering you wrong.
No apology needed, you haven't! I was thinking about adding cooling before so your suggestion just encouraged me to follow through. I could just tell the fan ducting wasn't the spot so no worries there.
A carpet air mover would definitely move more air, but the pack is close to a workspace I don't want a bunch of noise at. The bathroom fan is definitely forcing lots of air through and I think it's plenty.
The 12V amazon battery tester arrives any minute now. Do you have any instructions on testing/data recording with that?
I have a spare, Q200 charger.
Would this help him to condition the cells?
The Q200 can charge 4 at a time at 5 amps, but only discharge at max 2 amps per cell unfortunately.
I used two of them to work on 8 cells at a time when I reconditioned my battery pack.
Here is a link to my spreadsheet. https://docs.google.com/spreadsheets...t?usp=drivesdk
I had a spare battery, Battery A. Battery B is from my 2013 Silverado.
The "final" tab shows what cells made up my final pack which I balanced (connect all pos and all neg terminals together and letting it sit for 48 hours) after all were given a last 5 amp charge.
I will run the test and upload it in a few days.
(A temp probe should have been used on each cell to ensure charging was stopped when the cell temp increased too much after they did not automatically terminate when full)
The most important piece of information you've gathered thus far is the initial discharge value. Those appear to all be pretty close around 2000Ah. After a 1-2 year sit, you can be reasonably confident that all modules have 2X THAT measured value in total capacity (about 4000mAh), which is great for this car and this battery.
On that basis alone, I would conclude your pack is serviceable AS IS with no additional processing; however, you've started something, and now you need to finish.
Comments:
1) 3A charge is too low. For the way NiMH charging terminates, it's important the charge current is near 1C or 1 X capacity. Capacity is 6.5Ah, so 1C is 6.5A. Given their age and presumed reduced capacity, 4-5A would be preferred. If you can adjust the sensitivity of the chargers, that would be good. Normal sensitivity is 5-6mV/cell. You want to be around 2-3. Good news is 3A is a relatively low current. Little chance of damage. The fact that you're pumping 7500 (I'm assuming a set cut-off) into most of them, but only getting 4000-4500 out indicates the charge termination criteria isn't sensitive enough. For 4500mAh output, one would expect to only put in 6000-6500mAh typically. For charges that terminate below 7500, those are either actual properly terminated charges for false triggers due to a bad connection. LASTLY, for consistency sake, do not make any changes to the process.
1A) your leads suck. Nothing personal. Nobody tells you about this. Recommend you use a zip tie to "pinch" the banana plugs together where they exit the charger. You don't need a lot of force, but you want to put pressure on them to insure they have good and consistent contact. Cut the alligator clips off and use crimped ring terminals - even the cheap ones are fine. Secure to terminal with the nuts that came with them. Torque to 48 in-lb. If you don't have a 48 in-lb capable torque wrench, get one. You'll need it for the assembly. Proper torque is critical.
2) 1A discharge is too low. I understand that you're limited by the chargers. The downside is that your results may not be in any way indicative of their actual rated capacity. #3 will help correct that.
3) buy this: https://www.amazon.com/gp/product/B07Q6NBXQG If you already have a digital tester like this that can measure milliohms (mΩ often reported as "mR"), you don't need it, but we'll need to figure out if it will work.
4) DO NOT CHARGE THE NEW MODULE UNLESS IT'S COMPLETELY RESTRAINED. IT WILL INFLATE LIKE A BALOON AND DESTROY IT. DO NOT BE SURPRISED IF IT'S A COMPLETE PIECE OF ****. SO MANY UNSCRUPULOUS/MISLEADING SELLERS.
5) Your spacing between chargers is good. Keep doing this. You need the adjacent untested modules acting as a heat sink. Even at low current, these things can generate heat. If any exceed 40°C, terminate charge. If you can conveniently blow air down through the modules for cooling, that would help.
6) The loss of capacity after sitting for a week is normal. They bleed down fairly rapidly from full charge and then they settle to a very slow discharge.
7) Recommend you conduct the first discharge and 1 charge on all remaining modules before proceeding with cycling. Again, leave your current settings intact to ensure you get consistent results, but apply the zip ties and ring terminals.
The tester in #3 can measure internal resistance very reliably with the proper settings. It may not be absolutely accurate, but it will be very useful for checking for outlier behavior. It will help to confirm that your 1A capacity will translate to usable capacity in the car.
The "final" tab shows what cells made up my final pack which I balanced (connect all pos and all neg terminals together and letting it sit for 48 hours) after all were given a last 5 amp charge.
This is a completely fruitless exercise, and I try to discourage it any time it crops up. To give you an example, I put a completely empty module in parallel with a module that I specifically put 4500mAh into and left it in parallel for a week. At the end, 1500 was in the empty one and 2200 was in the charged module. This indicates a > 10% variation in state of charge, which is not good.
In addition to that test, here are some fundamentals:
Resting voltage of the modules doesn't matter. You can have them all perfectly the same voltage, but the moment the car hits it with accel or regen current, they're going to respond according to their charge level, not voltage, and they will all diverge.
Current flows based on voltage difference divided by the resistance. Modules in parallel almost instantly stabilize in voltage. If there's no voltage between the modules, there will be no current flowing.
Alternatives:
Simply charge to full and then discharge to an elevated voltage cut off. Best to be determined experimentally. See what the voltage is when you've discharged to ~60-70% capacity and then discharge all other modules to that voltage.
With known capacities, charge to a set input - 60-70% SoC for each module
If all modules are within 10% capacity, charge to a simple cut off equal to 60-70% SoC of the average capacity.
If you insist on paralleling all modules, CHARGE them this way to a given voltage. This can take a very long time due to the 260Ah rated capacity of a 7.2V configuration.
After a weekend away with kids sports, back to record some more data.
My main problem right now is not knowing when most of the cells had their last charge. So populating that data will be difficult unless I start cycling them all over and record their date of charge.
Anyway, I was able to use the Ancel BST200 tester and the results are below. The modules showing higher voltage were definitely charged more recently. The resistance on my bad module (#27) clearly shows a huge difference!
Shall I switch charging technique? What is the best way to move forward ?
Being fully charged any time in the last week-ish is fine.
CA vs. MCA was the intent, but it's fine.
All are over 500MCA and all are high 6/low 7 mΩ.
You're good. Those number support the conclusion that the low current discharges are representative of high current performance.
I have left instructions in the sheet chat.
On the tester itself, I did select CA for the test. I believe MCA was just "measured cranking amps". Well that's what I concluded anyway, since I made sure I was on the CA test setting.
When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
04-18-2023, 02:15 PM
Re: 2009 Silverado Hybrid battery conditioning
