Just Another HV Battery Thread and Experiments

Time for another quick update. Battery "P" has been on the bench for a few days and had seven discharge/charge cycles performed. The modules were discharged at ~1.3 amps to 6.0v and then charged at 2 amps until hitting a limit of 8000mah. None of them stopped at the delta peak setpoint, which would have occurred if they actually achieved full 100% charge. The way my work schedule is, I do two full cycles per day. I start a cycle before I leave for work and then start another after I get back home. I figured 3.5 days would be comparable to the time it will take to do the complete 3 time Prolong cycle on the "A" battery. The "P" battery has been resting on the work bench for about 2 days and should be installed in the car this evening to start testing. This may be delayed, as a friend is trying to get a card game going tonight, and I have to keep my priorities ). The "A" battery is still installed in the car and I did a final load test on it this morning after arriving at work. After it gets swapped with the "P" battery, I'll put the "A" battery on the bench and put it through the full Prolong triple cycle.

These are the screen captures of the load test performed this morning. Remember, this is the "A" battery that had the Prolong "topping charge" performed and then put back in the car. I think it's been 6 days since it was "topped" and probably about 400 miles. I had an incident yesterday shortly after leaving work where the battery dropped rapidly to 2 bars and then 1 mile later it was back at 5 bars. Module 6 is back to being a problem again. You may notice that the delta v is slightly lower than before, but that is most likely due to the current draw only being 4 amps instead of 8-10 amps. The AC compressor just doesn't draw as much current in cooler weather. The estimated capacity is back to 1.3 mAh.

 

Here's a bit of information about the "P" battery modules. There were 7 cycles performed.

From the final cycle:
The 28 modules averaged out to show 6201 mah capacity.

The module I used to replace the arc damaged module performed the best showing 6370 mah capacity.

There were 8 modules that showed >6300
There were 8 modules 6200-6299
There were 5 modules 6100-6199
There were 4 modules 6000-6099
There were 3 modules that showed <6000 mah, they were 5924, 5948 and 5997 mah

If I were to rebuild this pack for sale, I would replace the "<6000" modules with three that were in the 6250 range.

There was no other work done to the battery. All modules (except the replacement) are positioned exactly as they were when the pack was originally removed from the "P" car.

I'll try to get some testing performed on it this weekend and get those results posted for comparison to the earlier "P" battery load test results.

 

Battery "P" has about 100 miles on it now after the hobby charger cycling. This is the first load test. Unfortunately, the weather here has been the coldest of the season so far (35-50F). Because of this, it's on;y loading between 4 and 6 amps. The bounce in the graphs are due to the current load changing between 6 and 4. Even so, the average delta is ~0.1 volts over the entire 14 minute load test. All 14 module voltage curves are grouped fairly tightly with no outliers. Estimated capacity has bumped up to >5Ah. I'll get more load tests over the next several days to give us something for comparisons. Hopefully the weather will warm up a bit so I can get into the 8 -10 amp range.

 

MTL,

I understand exactly what you're talking about. I was trying several things to try to get the battery load higher, including putting the car in reverse. I typically run the AC on full and that USUALLY gives me 8 - 10 amp load. Today, because of the low temperature, it was bouncing very frequently between 4 and 6 amps, along with a couple transients at the 9 and 14 amp level. I believe that's the 'spiking' we're seeing. When I enlarge the graph, it shows those spikes aren't a single module, they're actually multiple modules all moving together. It's supposed to be in the mid-60s tomorrow, so I'll try the load test again around 2-4 pm.

 

Here is the closest close-up I can get of the largest spike.

 

Here's the graph of today's load test. I wasn't able to get it done during the temperature peak today, but I did get it done this evening while it was still ~60F. Not ideal for loading the AC compressor. I initially was running only the AC but it was bouncing between 4 and 6 amps again. Then I put it in reverse with some brake and throttle and it seemed to lock in at 6 amps. Then I turned on the highbeams and it flickered between 6 and 8 amps a few times and then locked in at 8 amps for the entire remainder of the test.
I was a bit uneasy performing the reverse/brake/throttle maneuver. I've never been a fan of forcing current through windings when they're not actually performing their job. I kept feeling like I could detect an intermittent surge in the car. It's hard to describe. Kind of like a seat of the pants feel that the car just had a slight jerk, but small enough that you're not sure if you really felt it. Like if you're taking a hearing test and you're not quite sure if you heard that super low decibel tone or not.

Anyway, total load test duration just 4 seconds short of 10 minutes. I think it ended up being around 8-9 minutes at 8 amps. Estimated capacity about 5.7Ah, average delta V was ~0.12

Look at the Delta Thresholds on the final screen capture compared to the ones posted earlier in the thread.

Not bad.....

 

Also wanted to provide a minor update on the "A" battery. It's currently on the third discharge using the Prolong System. A couple hours ago, it was at 202 volts and 1.489 amps, on it's way to 17 volts. I'm probably not going to perform more than the three usual cycles (as this is probably the most likely scenario for a typical user). I'd like to keep the "P" battery in the car for at least a week or so to get a load test on day 6, since day 6 was when the "top charge" seemed to falter.

 

You aren't kidding. My norm is 6.0v at the battery terminals, not 0.6, lol. But, I have faith in the theory. If it wrecks my modules, I'll just send a bill to Jeff. After all, I'm the first person in the world to ever do this, right?

(this was added 11/14 @ 9am) OK, I may be interpreting some later posts incorrectly, but I just wanted to clarify that my statement that starts with "If it wrecks" and ends with "right?" was meant in jest and with sarcasm intended about being the first. I have no doubt the modules will be just fine and there have probably been multiple thousands of previous 17v cycles.

 

This is the part that I'm also very curious about. Hopefully, if I'm able to get a decent load test performed everyday, we'll be able to see if that's in our future. I like the estimated capacity numbers. It can vary a bit from test to test, but maybe we'll see a trend developing after a week or so.

8-9 minutes at 8 amps and a 5.7 Ah capacity estimate is pretty respectable for a rebuild that has 23 modules date coded 4/18/07, 3 modules date coded 6/19/05, 1 date coded 3/24/05 (this is the replacement module) and 1 date coded 8/22/06. It will be interesting to see how it holds up.

 

I've had Battery A installed in the car for several days and just finished a 450 mile round trip with it yesterday afternoon. I performed a load test on it this evening when I got home from work. Outside temperature was around 68 degrees. I had the AC on full blast with headlights on. Current draw was 8-10 amps. The HV battery started around 65% SOC and I stopped the load test around 43%. The car engine started just as I pressed the stop button for the load test, so it actually registered the load current for cranking the engine. Thats why it shows a max current of 70 something and kind of threw the minimum voltage readings out of whack. That big current draw transient caused a big voltage dip. I included a close up screen capture of the last second of the load test so you can see the load transient when the engine cranked. There's starting to be some separation in the module load curves, as a couple of the "better" modules/blocks are showing themselves. The remaining modules/blocks are grouped fairly tightly. It's still only 0.3 volts from highest to lowest at the end of the discharge. Remember, battery A had the full recommended cycles performed using the Prolong System. Estimated capacity is doing relatively well.

 

To help anyone who wanted to look at the history of Battery "A":

10/28/2017: Battery "A" was removed from the owners car. I installed it into my car and drove home 200+ miles. Performed a load test shortly after arriving home and documented in Post 66.

10/31/2017: Performed another load test and documented in Post 67.

11/1/2017: performed load test and documented in Post 68.

Performed a topping charge using the Prolong System, installed in car and performed a load test documented in Post 84.

Six days after topping charge, started having issues again. Performed load test and documented in Post 101.

Removed it and started the full 3 cycle Prolong charge/discharge routine.

After the routine, installed in car for several days and several hundred miles. Load test performed 12/1/2017 and documented in Post 118.

 

It's been 10 days and probably 500 miles since the load test on December 1. Here's a new test performed today. The current draw was a bit higher today, but at a glance, the battery seems pretty steady and consistent with the previous load test. this is Battery "A", which had NO work done to it, just the Prolong cycles. Although, the estimated capacity has dropped a bit. Block 1 and 2 are consistently outperforming the others. Block 6 and 7 are the laggers. As expected I suppose. The outer blocks are outperforming the inner blocks. This seems like a good candidate to move blocks 1-7 to the 8-14 position and the 8-14 blocks to the 1-7 position. It also probably wouldn't be very difficult to match a few modules to move those out-performers closer to the average, which would minimize the deltaV for the short term.