Traction Battery Monitor results and information interpreted.

Initially posted in "what did I do to my Prius today" but this is probably the better place for it with some editing, research, more detail and links added.

My new to me 2010 Prius has 111,000 mi. and I've gone through every maintenance item possible.

The big remaining unknown is if the high mileage HV battery pack is healthy or needs replacement or possibly individual cell replacement. These batteries seem to be built in a similar manner to aircraft batteries with individual cells connected to each other in series to up the final output voltage.

I saw the Traction Battery Monitor data logger mentioned here on priuschat and got one:

Traction Battery Health Monitor for Toyota Prius - OpenElectrons.com

NOTE: You can trigger the triangle of death and check hybrid system message if you disturb the data logger while the car is in ready mode or running. Don't do this.
It was not possible to clear the message with my OBD II scanner. However, if you don't panic the message will go away after a few start cycles.

The documentation and user's guide is very sparse so I'll try to give a better understanding of the data you are provided from use of the monitor/logger.

In correspondence with the maker, he says very little data needs to be drawn to get a view of the battery health. That's true, but the longer sample I took has lead to more useful temperature information and value extremes for particular components.

The following battery test results are from my commute to work. 1994 seconds (33 minutes). Mostly flat, slow moving traffic but not much stop and go.

Battery SOC (State Of Charge)
From what I've read, even when the battery indicator indicates fully charged or 'empty', that is rarely the case. The system keeps the battery from getting to either extreme in order to protect it's heath and recharge capacity. (I am planning on doing an experiment on a long regen downhill to see where it truly does top out)




The I(A) chart is amps going in and out of the traction battery / motor-generator. Showing the positive spikes of acceleration and negative dips of regenerative power going back in.
(The long regen downhill experiment will show some interesting results here too.)




HV. High Voltage Battery total charge. Seems to be mostly between 220-240V with only a few excursions. You will see this is loosely tied to the state of charge graph.




A strong indication of battery condition comes from comparing cell voltages to each other. There are 14 individual cells in this HV battery. If one cell starts deviating from the others it can lead to problems in other cells, but if it is detected early enough it can be replaced.
The data obtained below is in milliVolts (thousandths of a volt) so allows a very fin resolution which is helpful when looking for differences. 14000mV-18000mV is the typical range (14V-18V)
In this chart the individual cells are graphed one on top of each other.
They all coincide pretty closely so that's a good sign. (NOTE: As expected this graph is identical to the HV graph above as this is just a breakdown of the individual cell voltages. (Multiply these volts by 14 to get the HV voltage at any point along the graph))
*error in the horizontal legend. Should read '14000-18000 allowable'*




In the previous chart there is visually some slight deviation but it's hard to tell how much, so I did an experiment to see what just a 2% deviation would look like. I took one cell's data and skewed all it's values down two percent and re-graphed it.



2% turns out to be really noticeable so seeing a cell going bad would be easy with this device.



The strongest indication of cell health is it's internal resistance. If a cell starts going bad, it's resistance will increase in all conditions and start to deviate from the others. This is caused by 'sulfates' / deposits / breakdown of the cell's inner structure. You would probably start to notice a spike in temp in that region of the battery pack too. This is what can lead to other cells going bad.
The scale here is in thousandths of an Ohm (mOhm). A very small unit of measurement, so the tolerances are pretty tight and the changes would be immediately noticed. The acceptable range is 15-30 mOhms.
Here are the 14 individual HV cell resistance readings all together. These are well within the allowable range. You can see the values start going up and topping out as the battery warms up to normal operating temperature. You'll see that the battery temp chart is similar.





And the battery temp graphs. The blue line is the temperature of the cooling air coming into the battery. It went up a couple of degrees, probably due to the latter part of the drive that went into the valley where the outside temp is significantly warmer. The three battery temps are pretty close to each other, 1 and 3 being identical, my guess is that they are on either end of the pack. The number 2 temp is probably the temp in the middle of the battery pack so understandably warmer.





Here's the Motor/Generator temps. Until now I thought there was only one traction motor...I learned otherwise:

Toyota Prius - Power Split Device

That tells a LOT about how our drive system works. Play with the sliders, very interesting. It also shows how robust the 'Continuously Variable Transmission' is. It's not actually a CVT like in a snowmobile but rather a Power Split Device. There are no components to wear out and slip like a belt or clutches. The oil just bathes the planetary gears and possibly some of the electronic components.
MG1 is the smaller of the Motor Generators and is also the starter for the gas engine. It's still a big motor and one of the reasons starters aren't really an issue on these cars. Interesting to note that at 76mph, both electric motors and the ICE are at the same rpm, the PSD components are stationary to each other and the least amount of wear and heat are produced in the PSD.

On to the motor temperatures...
They looked to be leveling off, but it would have been interesting to see where they would top out on a longer drive. This data could be good to compare from one historical reading to the next and see if the motor windings are breaking down, rising in resistance and heating up over time.





Next are the temps of the inverters that supply the motor/generators. Pretty spiky, this looks to be load dependent in an immediate way.
Inverter 1 (blue) and it's Motor as responsible for starting the ICE so that may explain the bigger spikes.
Inverter 2 (red) and it's Motor are the only thing directly driving the wheels.
The electric motors spin both forwards and backwards in normal operation and at various speeds so making much more sense of the data could be difficult.
What could be useful is to see mass increase in temp over a span of time indicating a breakdown of the electronic components and greater inefficiency. This doesn't seem to be a common failure in these vehicles though.



That's it. This little device has led to me understanding a lot more about how the vehicle works and it's condition. I was a little wary of the high mileage but am pretty confident in it now!

 

Anyone in Los Angeles that wants their HV battery/traction motor information read and graphed, I would be glad to help out.