Ah. got you confused with AlexK and his 2010. Sorry.

SoC is not voltage-based. It's voltage, current and temperature based. NiMH is kinda nutty. I've seen cells at 60% SoC with voltage ranges from 1.25 to 1.33. This equates to 312.5V to 332.5V, i.e., 20V difference.

 

The voltage probably sags when the car pulls decent current and then rebounds when the current draw stops. That's how it can not move much in the voltage department. I don't have much NiMH anymore besides AA cells but I see this with all my lithium powered devices and lead acids.

 

All battery voltages sag when a current is applied and rebound when it is removed due to Ohm's Law. Batteries have an internal resistance, so they change in accordance with V = I * R.

The only battery cells that have a very strong correlation between RESTING voltage and SoC is Lithium NCA, NCM and LMO chemistries.

Lead-acid, particularly AGM, is pretty good too.

NiMH, NiCd, LFP and LTO chemistries are all pretty much junk for resting voltage vs. SoC.

 

Just to add some info on the Scangauge, if HV is disconnected at the battery, The SG will not turn itself ON when the key is turned. It requires a manual button push to activate the display. That's apparently why the SG not turning ON has been a reliable predictor of low SoC, more than just 12v must be required for it to sense a turn-on signal.

 

UPDATE: I followed the procedure above, but I had to allow additional time as I was away from home on a trip.
The FEH was stored with SoC = 50.0% (TBV = 336, BTM = 77.0). I returned to find S0C = 0.0% (TBV = 330, BTM = 75.2).
I was able to start the FEH after 3x jump start cycles.

Since the HV battery was completely disconnected from the vehicle, this seems to prove conclusively that the battery is self-discharging not being drained through a path in the vehicle circuitry.
I am puzzled why TBV changed so little (336 to 330) for a big change in SoC (50% to 0%).

 

I finally got the chance to run the RMode Battery Balance function today. It performed the "Service Procedure" and ended normally, and the ICE turned off afterward. FORScan seems to work fine with the new laptop & OBDLINK-EX adaptor, it found all the modules. It looks like I have a single DTC in the ABS module, but not serious (no light set). I'll look at it again tomorrow. Otherwise zilch, which is good because I have to get the car inspected this week.

It only took about 23-24 minutes to run the BB function? Is that good, bad or indifferent? The car was cool/cold when I ran it, outside temp. about 76 degrees and overcast and humid, and I don't know why but I decided to open the rear hatch throughout the procedure (electric AC and all accessories OFF). Took it for a very short ride afterward with the A/C on to cool everything + me off (humid!) and in 5.3 miles I got 46.2 MPG. I know that is an anomalously high number, the ride was very mild in terms of speed and hills. The battery fans are working, I could hear them after the test drive. I let it sit in "IDLE" for about 2 minutes with the fans running before I turned it off.

I don't know whether it did anything positive in terms of long-mileage stints but it doesn't appear to have harmed anything, EV mode was working fine and seemed as strong as it should be, even with A/C on.

 

I'm trying to think if something reset the SoC to 0%. Was 12V disconnected at any point?

Did you attempt to start the vehicle prior to HV jump? If 330V is an accurate number, the car should have started even if 0% SoC was indicated.

Have you verified that numbers reported via scangauge are accurate?

15-25 minutes are typical run times for me provided the battery is already in the 70-80% range.

 

Just found out that SoC defaults to 0.0 anytime the service disconnect is open, so I didn't really read SoC at the end of the procedure which invalidates my test. However, the HV battery voltage is still readable with service disconnect open.

After my 3x HV jumps, the SoC read about 48%. Since each HV battery jump adds a calculated 2.7% to HV battery SoC, the SoC after sitting for two weeks was about 40%. In other words the self-discharge only amounted to a 10% drop. But after reconnecting the battery, there was a real drop in SoC to 29% within a few hours. I'm back to the theory that the HV battery is being primarily drained by the vehicle not by self-discharge. Don't know how to proceed from here.

 

Now I'm intrigued. I'm wondering if there is a way to measure current drain through the connectors with a clamp meter like a Fluke 378FC (which I don't have.)

Please excuse my novice-level thinking on this but what happens if the inverter/converter (or something in the battery) is drawing and dissipating a small but continuous amount of current? The vehicle hasn't really lost HV isolation so no code would be thrown but it would drain the battery? Maybe I'm all wet.

https://www.fluke.com/en-us/product/...-meters/378-fc

 

That's an interesting theory but my feeling is that the current draw isn't continuous. I have seen a 20% SoC drop occur within six hours, but it sometimes takes several days before coming off of 52%. Seems like something is triggering a current draw at random times and for unknown reasons.

I have scheduled an appointment in early Aug at a local shop that specializes in hybrid vehicles. At this point I'm willing to spend a couple hours of shop time to get other ideas on the problem.
If this car wan't my daily driver I could be more aggressive in troubleshooting, but I have to keep it drivable. Thanks for the ideas!