High delta SOC and other symptoms

Apologies for the cross-posting, this is following from my comments on someone else's thread but that conversation seems to have stalled. I'm hoping someone can share some diagnosis suggestions for my '05 Prius (101K mi) that has been exhibiting abnormal HV battery behavior.

• With city driving, the HV battery is consistently being run down to one purple bar on the SOC meter. Have been experiencing this for at least the last week+.
• When this happens, acceleration becomes very sluggish and the ICE runs almost constantly as it tries to charge the HV battery.
• I checked voltages with a mini VCI/Techstream, but there don't appear to be any outlier blocks.
• The max/min voltage seem to track okay (within 0.1–0.3V).
• Delta SOC% is very high—consistently running at ~80%.
• HV battery SOC appears to drop at 0.5% every 1-3 seconds even with the car stopped and/or in P.
• No error codes are being generated (yet).

I saw on the Gen II troubleshooting sticky thread that ECU or the invertor might be culprits. Can I check these with the mini VCI? Are there other possible maintenance issues that I should consider? What might the high delta SOC indicate?

 

Thanks for your help yet again!

1. Yours seems to just go down and stay there: There have been some instances of fluctuation, but yes, mostly it just drops low and stays there.
   
2. definition of the delta SOC parameter: "The normal, low to high SOC deviation is 20%. If the Delta SOC exceeds 20%, this means that the HV battery ECU cannot correct or maintain the SOC difference within the acceptable range." (source)
   
3. what happens on force charge?: I haven't checked this, will report back.
4. instantaneous MPG at constant speed: I'll have to check this explicitly under those conditions (tough to find highway flat stretches near Seattle though), but average highway mileage has been only a bit lower than usual (~43-45 mpg). Not sure about instantaneous since it fluctuates (few flat stretches) between ICE on/off.
   
5. amps coming from MG1 and MG2: Unfortunately, I only recorded the HV battery data. I'll take a look at the engine data when I attempt the flat road test.

Some other info that may help:

• See the attached "screenshot 1". I recorded a drive around my neighborhood just to see what happens with the different HV battery measures. Notice how the SOC drops when at a stop (in D), going from 75% to 40% without pressing the accelerator (starts around 280s mark on the x-axis timeline). Is this normal? If not, what might be the causes?
• Similarly, "screenshot 2" shows a 10% drop SOC within 2 minutes while powered on but with the ignition off (sorry, not sure what is the official name of this power state, with start button pressed twice without foot on brake). Is this normal? If not, what might be the causes?

 

Some more updates:

• Did another brief in-city drive, and got some more weird symptoms. Let the car idle briefly upon startup which charged the HV (see screenshot). Then drove around briefly, after a couple minutes of which the SOC began to plummet all the way to 0%. I didn't think it was possible for the HV to discharge to that point??!!
• I repeated this sequence two more times, but another drive this evening failed to get the initial charge. SOC sits at 0% (!!!).
• Still no DTCs, but with SOC at 0%, I do get a warning about putting the car into N that the main battery will not charge.

I'm not sure how to monitor amperage from MG1/2 going to the HV within Techstream. Do you happen to know?

 

Just bumping my own thread in case there is anyone out there who has seen this strange behavior. Techstream reporting 0% SOC, how is this even possible (while still running ICE, and without any DTCs or dash lights)??

 

I think you are on to something with the high battery temps. At the start of this drive (but not during this particular recording), I did hear for the first time the high HV battery fan noise. Total drive time was ~45 min. but this included several stops with the engine off. Ambient air temp, as you note from the intake, was relatively high (90F). A quick look at this PC thread indicates that, indeed, 126F is considerably higher than expected though I'm not sure about the poster's driving conditions (if that could even account for such a large deviation).

Attached is another set of screenshots from the same drive showing SOC %, one of the battery temp readings, and battery current. Tomorrow morning I'll try to get a recording from a cold start.

I still can't believe that there haven't been any DTCs thrown—would make this diagnostic process a lot easier. I'm taking the car to the dealer tomorrow for a "free" battery test so maybe that will provide more info. I'm still absolutely puzzled by the 0% SOC reading—how is that even possible?

One more question: do you think it's worth checking the battery cooling fan? I don't have pets and am generally pretty tidy with the car, but I'm less certain about the previous owner. Just trying to consider other causes of the high battery temps, and wishfully thinking it's not the actual HV battery.

Thank you again.

 

Thanks for chiming in.

Either something is wrong with my VCI/Techstream setup, or there is some computer/sensor error on the car itself. The ICE definitely starts while reporting the 0% SOC which, even with my limited understanding, just isn't possible.

Is there any chance that the car would fail to prevent starting the ICE when SOC < 16%? Given that the car still starts, I think that the reported SOC must be inaccurate. But I also want to be sure that I don't damage the battery (if that hasn't already happened) by continuing to drive it.

 

Actually, the MFD does seem to track the MVCI readings. At 0% SOC on the MVCI, it shows a single purple (pink?) bar.

 

Thanks everyone for your troubleshooting, very much appreciated. The added pressure for me is a planned road trip to Yellowstone in 1.5 weeks. Not figuring this out beforehand would leave me very uneasy about driving my Prius.

slimothy:

• Looking back at my screenshots posted earlier, it looks I was getting ~40A out of the battery with in-city driving (if I am reading that correctly, might have +/- reversed and/or might be looking at the wrong data).
  
• Modules look pretty balanced to me—difference between max/min is between 0.1-0.2V including under discharge (in R with foot on brake).

usnavystgc:

• What sensor(s) or ECUs do you think might be the culprit? Hybrid Control ECU? Battery ECU? Other components? Other than swapping parts, would there be definitive ways of testing?

nh7o:

• Thanks for the suggestion, looks like I might have my Sunday morning activity planned for me. I saw some videos online, doesn't look too complicated as you indicate so I might have a go at checking the fan/ducts and battery.

 

Well, that's reassuring. But there is still the bizarre behavior with the 0% SOC—if I didn't multiple recordings of it on Techstream, I'd think I was going crazy.

I'm currently neck deep in revealing the HV battery and, so far, nothing there either. Drove the car this morning and it exhibited totally normal behavior with the HV system. The inconsistency of the 0% SOC is both reassuring and incredibly frustrating. And with no DTCs, I doubt the dealer is going to uncover anything at all in their $114 diagnostics test.

One more question: do any of you know what data set I should be tracking in Techstream to monitor the HV output current?

 

Fair point, my only thought is that this is an ECU issue (since 0% SOC is not possible), but I would expect a defective ECU to trigger a DTC as well. It's probably quite obvious that I'm fumbling at the limits of my comprehension, just trying to figure out some way to understand or explain the weird SOC behavior.

I exposed the battery this morning, visually inspected the terminals (some corrosion on 8 terminals but looked pretty minimal to me), and manually checked each cell with my multimeter. I also inspected the battery fan which only had minimal dust on the fins. End result: nothing of note.

So this leads back to the sensor suggestion. Any advice on further troubleshooting that?

Thanks again.