I've finally gotten around to running the HV Battery test via the Hybrid Assistant app and am quite shocked at the results I'm seeing. Can I really trust these block readings instead of pulling the battery and testing each module for ~7.2v? I expected the maximum delta for observed readings to be far higher than it was. Should I really expect years more out of this battery with no stress? Our family makes an annual 5k road trip from New Mexico to New York to Georgia and back and a failed battery pack in the middle of the road trip would be a ginormous bummer. Screenshot_20180715-095632 by hearsay posted Jul 15, 2018 at 10:03 AM Global stats Minimum observed battery voltage 196 V Maximum observed battery voltage 226 V Minimum observed block voltage 13.87 V Maximum observed block voltage 16.19 V Maximum Delta 0.32 V Average Delta 0.20 V Minimum observed current -18.00 A Maximum observed current 6.00 A Minimum observed SOC 47.8% Maximum observed SOC 68.6% Delta SOC 20.8% Energy 366mAh Estimated Capacity 1.87Ah Screenshot_20180715-095714 by hearsay posted Jul 15, 2018 at 10:03 AM
expect, or hope? gen2 have been known to go many miles without much trouble, and the battery prefers to be used. still, hard call on the road trip. unlikely to have a failure, but always possible.
Yes, you can really trust these readings. Interpreting the readings can be difficult. Pulling a battery apart to measure static voltages tells you next to nothing. I don't think so. Strike one is being in the heat of NM. Your battery has lived a hard life. There are two concerning numbers in the Hybrid Reporter data. The first is block six being the minimum 47.3% of the time. That's not normal. It suggests that block six will be the one to fail. Strike three is the estimated capacity of 1.87Ah. That's really low. Typical would be in the 4 to 6 Ah range. Could you show more of the Hybrid Reporter results? How does the Consecutive Samples at Delta .2V look? Driving a ten year old car across the country is not without risks. Here's a short list of things that could be replaced or serviced before such a trip. It could add up to so much money that it does not make sense. 1. Install NEW HV battery, 12 volt battery, tires, both water pumps, serpentine belt. 2. Change engine oil, tranny fluid, coolant.
Great thoughts above Strawbrad, and below are the results from the Consecutive Sample chart. All looks good on it from what I can gather and the blocks are consistently within .2v of each other. Delta Threshhold/Consecutive Samples 0.2/167 .45/0 .7/0 .95/0 1.2/0 Your point about the one block reading minimum voltage nearly half of the time concerned me too. It seemed like an obvious weakness. I was in contact with the developer, and he claimed that one block reading low for a bit was no big deal. His emphasis was that the greatest indicator of battery failure was the voltage delta, which seems to check out fine in my case I do not know how to interpret the low capacity of aH. I'm guessing that swapping out that one weak cell would not help the overall capacity all that much? Is there a way to determine what is causing the low overall capacity?
I do not know how to interpret the low capacity of aH. I'm guessing that swapping out that one weak cell would not help the overall capacity all that much? Is there a way to determine what is causing the low overall capacity? Your graph shows that the battery test continued after the engine started charging. My experience with HA is that you MUST stop the test before the engine kicks on and starts charging the battery. That's why I stop my tests at ~43-45% SOC. If the engine kicks on, it totally jacks up the capacity calculation. I've also found that you must completely back out of the HR app before you look at another test (saved file). Something in the data doesn't clear and a previous report will totally jack up the Ah capacity calculation of the next one, UNLESS you totally close the app and then reopen it.
Just for comparison I ran a HA test on my 2008 with 150K on the original untouched battery. This car has been in MN since new. Our generally frozen climate is very easy on batteries. I wonder how HA calculates % Time @ Min V. All the blocks should not be %. Maybe they need consecutive samples to count. Hersay, It would be interesting to see how your block 6 responds to a forced charge. Weak modules will swing to the highest voltage on charge.
Awesome insight, I appreciate all of the thoughts here. I've run another test making sure that the HV battery didn't start to recharge and the numbers are a bit more encouraging. I also ensured that I fully backed out of HA before opening the report to confirm that the battery capacity data wasn't screwed up, which I had done the first time as well. The numbers are a bit more encouraging, but overall capacity is a bit low at 3.56aH. Would replacing cells from the suspect blocks and rebalancing bring this number up? Block 6 still looks a bit weak as well, but not nearly as dramatic as before. Any additional insights gathered from the data below? Screenshot_20180717-081717 by hearsay posted Jul 17, 2018 at 8:39 AM Screenshot_20180717-081725 by hearsay posted Jul 17, 2018 at 8:39 AM
Your HV battery capacity is determined by the weakest cell (a module has 6 cells in series) in the HV string (28 modules) so rebalancing the weak modules or replacing them with modules that match the other modules will increase the overall capacity. JeffD
Block 6 is definitely your weak link. I would suggest going even lower on the discharge test, take it down to 14v/block and see the difference between the highest and the lowest. If you're looking to extend the life of your battery then reconditioning/reshuffling is the way to go. There's a possibility that after reconditioning block 6 may not recover capacity and may need a module or two replaced.
On the voltage plot, the voltage distribution appears to stay consistent both during discharging and charging. This could indicate just an imbalance, and not necessarily a weak cell. A simple top balancing with a grid charger might significantly reduce the delta V. A subsequent deep discharge reconditioning cycle would probably also improve capacity somewhat. I don't see any indication of an impending failure. I agree, that would give a higher level of confidence to see the response under strong charging and high SOC. You can charge it up to 80% SOC with force charging. Btw: My HV battery is a little older and still going strong. Last year I began annual top balancing with a grid charger, purely as preventative maintenance.
Ok, I figure it's time for another update. I recently ordered the OBDLinkX adapter recommended by Hybrid Assistant after losing faith in my cheaper OBD adapter I was using for the first round of testing. Suddenly, the test results are quite a bit different, and even more encouraging. I'd appreciate any comments here. Not much has changed with the car other than a 1,000 mi round trip camping trip to Timp Point on the North Rim of the Grand Canyon. First up, we have my new results on the battery discharge test. There is no longer a single block that is showing up as holding a very low charge like Block 6 was doing before. Could the OBD adapter really be reporting this that differently? I suppose switching back to my old adapter may be the only way to truly know, and i may do that at some point for confidence. As far as I can see, there is no big max bias to stress over, and moreover, the overall capacity (is this accurate) is not bad either. Next up, below are the results of a heavy charge test. Can anyone gather any good data from this? As always, the input of this forum is always greatly appreciated. I'm focused on getting to 300k in this prius, and keeping the Hybrid battery is clearly a huge step in the goal. I am certainly considering using a Prolong charger to keep the battery balanced and operating at max capacity. The only problem I really see with the battery is that my mpg is a bit lower than I would expect. Even using hypermiling techniques and watching Toruque/HA, I can rarely top 50 mpg on a trip. Perhaps it's because we live in a steeply hilly town, but I still would like to get better than a 45 mpg average.
I'd like to see you add the Prolong setup to your battery, do a grid charge then test the battery again after around a week of driving. The weakest blocks (esp if you have one failing) will usually fall to the lowest voltage when under load and then jump to the highest voltage when charging. This is something best seen in Techstream while someone else is driving as it shows what the voltages are on the blocks in real time. What you want to see is a random variation of blocks at min voltage under load (accelerating) and max voltage under charging (braking) conditions. You can see the plot of the SOC for all blocks that Brad posted above is very tight (this is what you want to see while in use). I'd like to see a new one of those for your vehicle as well but I also strongly suggest at least installing that Prolong system and preferably you only use it in cooler months but make sure to crank up the AC if you need to use it during the hot months down in NM. It will certainly help you get the most out of your battery pack.
Keep in mind that the capacity is not a direct measurement here. It is an extrapolation of an estimate based on constantly varying measurements. Voltage, current, and temperature are used to estimate SOC based on an algorithm which occasionally recalibrates itself. The SOC estimate is in turn extrapolated from the capacity. Different conditions and self-recalibrations can produce different results. So no, the calculated capacity is not always very accurate, but considered in context of the other data, it gives you a good idea of the true capacity, and yours looks good, considering the age of the battery. The voltage distribution under heavier and fuller charging is still consistent. It helps me to visualize with a graph: HV Bat Avg Bias Chg Dischg by Fred_H posted Aug 11, 2018 at 12:25 PM The blocks are behaving consistently. There are no especially weak blocks behaving much differently than the rest. I see just the normal, typical minor imbalance, which shouldn't drastically affect the performance of the battery. I wouldn't necessarily attribute that to the battery, unless you have noticed unusual behavior of the SOC and increased running of the ICE.
