Driving with a degraded hybrid battery (P0A80/P3017 code).

Our 2013 Prius (190k miles) threw up the "check hybrid system; stop the vehicle in a safe place" message and the dashboard lit up with the triangle of death, traction control, engine check light, and ((!)). The hybrid battery fan is also running at top speed. The fault codes are P0A80 and P3017; the battery evidently needs some attention.

Unfortunately, my wife needs the car for work tomorrow, and we can't get an alternative car before she leaves in the morning. Is feasible to drive the vehicle in its current condition or would we risk ruining the rest of the battery or even have safety concerns? The trip is 60 miles each way.

Some context in case it's helpful:
- we haven't felt any sluggishness but don't drive the car very aggressively (but don't know how it was driven before we got it in September).
- some cells were replaced by the second-hand dealer we got the car from.
- fuel efficiency seemed reasonable in the past weeks (40 to 50 mpg).
- it had an oil change about a week ago, but I imagine they didn't check the battery.

 

Thank you both! Shame that it looks like a new battery is required.

Having said that, I've been doing some more digging to get a better understanding of how the car works. As part of the journey, I'm now somewhat hopeful that replacing one or two bad cells might keep it going for a little while longer. Details below, would love your input.

I appreciate this is likely a game of whack-a-mole. But we may have to move without being able to take the car in the next year, making whack-a-mole somewhat more appealing than buying a new battery.

Anyway, here are some more details. We took the car for a short drive with reasonable acceleration and deceleration but nothing extreme. It was reasonably cold at 25 F in case that makes a difference. I collected the voltage of each module continuously using Car Scanner ELM OBD2 and the Veepeak BLE+ adapter recommended by Dr Prius, downsampled the data to one measurement per second, and then calculated the median voltage across all modules for each second. Module 7 identified by code P3017 is an obvious outlier, and the rest of the modules seem to cluster relatively tightly in the figure below.


I threw out module 7 for any further investigations because it's obviously dodgy. To get a better understanding of how the whole pack behaves under load and charging, I calculated the deviance between each module voltage and the median voltage in the pack to see how much of a misfit each one is. The figure below shows a scatter plot of the deviance against voltage of each module; each point represents a one-second interval.



Even though this doesn't tell me whether the modules are shot, it at least looks like they're somewhat similarly shot because the deviances are relatively small. For about 80% of the drive we had the absolute voltage difference less than 0.15V, and it never exceeded 0.2V (ignoring module 7, of course). Modules 1 and 6 look a bit over- and undercharged, respectively. Modules 3, 5, and 14 maybe need a bit of care because their voltage is below the pack at low voltages and above at high voltages, suggesting that their capacity may be degraded.

My current plan is to replace one or both cells in module 7, cycle 3, 5, and 14 to hopefully get a bit of capacity back, and make sure that voltages are similar for each module (double checking 1 and 6) before reassembling. That plan may of course change after I've opened the can of worms/batteries.

Any thoughts/suggestions/warnings you may have would be much appreciated.

 

Thanks for correcting the terminology; makes it a lot easier to search the forums.

The battery is now out, and the bad apple is easily identified: open circuit voltage of 6.88 for module 13 and between 8.09 and 8.18 for the other ones (details below). I'll try and get a better understanding of capacity and voltage under load.

It looks like the mechanic who previously replaced modules was a bit of a cowboy. A number of clips had been broken off, and cables were not secured. Some of the screws securing the battery in the car were also missing. So maybe a blessing in disguise that we can secure it properly when we put it back.

Good question. We'll definitely have it for the next year but may be moving without being able to bring the car thereafter, depending on jobs.

Here are the open circuit voltages.

Code:

#    Open circuit voltage
 1    8.17
 2    8.18
 3    8.12
 4    8.15
 5    8.09
 6    8.10
 7    8.14
 8    8.11
 9    8.10
10    8.15
11    8.11
12    8.10
13    6.88
14    8.15
15    8.15
16    8.15
17    8.14
18    8.14
19    8.13
20    8.14
21    8.13
22    8.11
23    8.14
24    8.15
25    8.13
26    8.14
27    8.16
28    8.17

 

A bit of a combination. I got an IMAX B6AC v2 charger/discharger to get a rough understanding of what's going on and four second-hand modules arrived from 2ndlifebattery today. Because I don't know the mileage on individual modules in the pack due to previous replacement, I bough the "generic" ones Joe sells on eBay.

The B6AC only allows up to 5W discharge so I couldn't test under heavy load (well, I'm probably a factor 50 to 100 out from what the battery does in the car). Other than the bad one, each cell dropped between 0.08 and 0.11 V after 2 minutes at 0.6A discharge current. The four replacement modules I received dropped by 0.10 to 0.11V under the same conditions (although starting from a lower open circuit voltage).

That's a very useful thread. It seems that the pack is a cheerful mixture of two separate packs, both much older (2006) than the car (2013). There are also a few blocks with serials 19700A... which I couldn't match with the pattern in the thread you linked.

I've put all the results into an online spreadsheet below.

 

Thank you for all the input. The battery is now back in the car and it seems to be running fine--for now. Fingers crossed we won't have to do it again any time soon, but I'm not holding my breath. Here's the voltage profile over a short period for the different blocks.