03 Prius intermittent engine stumble as if it's trying to start

Here are some screen shots showing what the miniVCI covers for the engine ECU:

Sorry for the terrible picture. I'll try to do a better one later.

I've attached the data file:

• Stage 4
• Stopped in parking spot: parking set, front wheel against curb, "D"
• Set cabin temp min.; A/C and MAX A/C -> engine started
• After ~10 sec, turn A/C off -> engine stopped
• Set cabin temp max.; turned on heater -> nothing
• Set "Defrost" -> engine started
• Turned off "Defrost" -> engine stopped

Let me know what you think. I won't be able to look at the data until this evening.

Bob Wilson

 

Sorry about the screen shot, I'll do better later tonight. But I just took a peek at the data. Next time, I'm only going to record a smaller set to increase the poll-cycle rate:

• IGN Advance
• CALC Load
• MAF
• engine rpm

I don't see the others as adding value. But even with the coarse, nearly 3 sec/cycle, my eyes see:

• ~6s - getting the engine spun and started - this explains the delay between setting AC and the fan coming on
• Coolant Temp, Intake Air, Throttle POS, Short FT and Long FT - they just are not changing enough to be useful
• Injector timing and MAF overlap - does not make sense to record both, using MAF

But going back to the original problem, I can see the compressor load as being something that a 'weak' or otherwise 'not quite there' engine could have a problem. I'm also worried about the low, user reported MPG.

Most 'weak engine' symptoms that I've experienced have come with distinct, associated symptoms. I'm going to expand the ambiguity group to include possible back pressure issues with the exhaust. The "30 MPG" suggests something bad (TM) is going on and in my experience, only exhaust flow problems can cause an engine-wide, reduction in power without other overt symptoms. The other option is too terrible to consider:

• the rings/cylinders are too worn - this would mean engine rebuild, too terrible to contemplate but requires some cylinder pressure metrics

Bob Wilson

 

Comments inserted
* * *
Hello Bob, Thank you for uploading the data. One thing I noticed when comparing data is that my long term fuel trim is positive while yours is in the negative.

This may also explain your poor mileage. It is as if the car is running super rich.

Upon inspection of my spark plugs I found that 2 of the plugs electrodes were slightly damp with fuel. The coils inspected ok. The plugs had no defects and were gapped within spec .042". I adjusted their gaps to .039". I carefully removed the fuel injectors and rail while the system was pressurized. None of the injectors were leaking. The valve cover design wrapping over the fuel rail, limits how far the fuel rail can be lifted up on the driver's side. This makes a leak inspection of the last 2 fuel injectors difficult to see, even with a mechanics mirror. I swapped the 2 suspect injectors locations with the presumed good ones. I'm going to pull the plugs after more driving to see if the there is fuel on the 2 plugs that were bone dry.

I'm thinking back to other vehicle that had excessive rich mixtures but they didn't have a catalytic converter 'soot trap.' Hummmm. Aircraft engines were perfectly happy to run super rich as the coked up the plugs.

I removed the crankshaft position sensor connector to inspect the contacts. I found engine oil soaked both inside of the connector and the sensor. Several drops of used oil came out of the sensor. There was no external evidence of an oil leak as it was all contained inside of the sealed connector.

A failed or failing sensor should keep the car from running at all. Everything suggest ultra rich mixture.

I found the recall for the 2001-2002 model year Prius for the CKP sensor. After reading through the recall it sounds like my Prius has the same or very similar faulty sensor issue. I dried out the connector and sensor with compressed air like it states in the recall instructions. I inspected the connector for oil contamination after a short test drive and it was fairly dry.

I checked on Toyota's website and there are no open recalls or service campaigns for my Prius. Do you know if there is an improved CKP sensor part # for the 2003 Prius? Would it be the same part # that's listed in the recall for the 2001-2002? I'm going to replace the CKP sensor. I'm looking for one that has been improved and not prone to failure/leak oil. Do you think the engine oil soaked CKP sensor pins created enough resistance to cause this issue?

I think of the CKP sensor as more like the distributor points of the old style breakers. Their usual failure mode was it would not run at all until you took out the emery board and dressed the points. I'm defer to any other opinions as you're in an area I don't have hands on experience.

I'm working to resolve these new found issues before looking into the potential AC issue, and performing the fuel economy testing and benchmarking.

No problem! I fully agree with your approach.
* * *

When you do your next run, see if the exhaust smells of gas, the other tell for excessive rich mixture. I wonder what the O{2} sensor voltages look like.

Bob Wilson

 

This is a puzzle:

The normal Lambda inputs indicate everything is working correctly, closed loop. My experience is:

• 0-> 45/55 seconds - open loop, works off of stored trim parameters and does everything it can to use MG2 while keeping the engine at idle. Continues until O{2} sensors begin operating.
• 55 -> ~300 seconds with coolant < 70C - closed loop, engine operates all the time with a rare windows when shifting into "R" (warm external temperature and coolant <<60C) can auto-stop engine. Otherwise, it runs until it can go into stage-4.
• coolant >= 70C - closed loop, auto-stop depending upon load, proper operation, goes into stage-4 and full hybrid laws apply.

Everything sounds right as far as the engine is concerned. Just double check the HC converter linkage, make sure it is not frozen. A frozen HC converter linkage would semi-block the exhaust and reduce engine efficiency. We normally expect an engine ECU code but . . . there are no guarantees. Also, nothing rules out a blocked catalytic converter or exhaust but since the O{2} sensor say it is working, I believe them.

Before doing a mileage test, check the rolling gear (not to be insulting, this is what I do with my own car) :

• check tire pressure, maximum sidewall (usually needs topping every two months)
• at same time, check and feel both the outer and inner tread feeling for abnormal wear
• not ready to check the brakes because it means pulling the wheels until after mileage testing (looking for heat)

Whether using a loop or two passes on a straight line, level is pretty important and of course, dry and low wind. If possible, use 38 mph on cruise control, at least 5 miles AFTER warm-up and in stage 4. The other optimum speed is 60-65 mph as this is just above the knee of the high-speed curve. After the last run, park and feel the hubs to see if they feel excessively hot, looking for a sticky or hot brake. In a 38 mph test, make sure the engine cycles ON/OFF.

Getting information from the transaxle and running gear is where a Prius aware scanner would really help. We would want to see ICE coolant, MG1 and MG2 temperatures as these are the primary determinants of how well the transaxle is working. We would also want the four traction battery temperatures and 19 module pair voltages, again, making sure they are all working OK.

I'm running out of idea, perhaps someone else has a thought?

Bob Wilson

 

Considering the 175K mile odometer reading and your finding that the engine is running rich, I suggest replacing the B1S1 oxygen sensor now. Although the sensor is working "more or less" based upon your tests and the lack of a DTC, it may not be switching fast enough to provide optimal fuel economy.

If the iridium spark plugs have logged more than 60K miles, I suggest replacing them. Note that it is not recommended to adjust the gaps on used iridium spark plugs as the concern is that the ground electrode may weaken and fall off.

 

It sounds like a minor issue with the hybrid vehicle ECU. Since your dealer found no faults, the only other item I can suggest is that you ask that the firmware in the engine and hybrid vehicle ECUs be updated to the current level. Then, if the problem persists, accept it as normal behavior.

 

Sounds like everything is normal. What MPG did you get at 38 mph?

Everything sounds good. Unless there is some previously unidentified, parasitic load, frozen A/C compressor clutch (?) or failing water pump (?), this is a real puzzler. Heat in the wrong place would be the way I would go. Say battery fan running or the MG1 and MG2 temperatures.

Roll-down testing is a difficult but fairly definitive diagnostic but you'll need some basis. The reason it is so hard is even very small grades significantly impact the results. So multiple, roll-down tests are performed and averaged. It really is a pain-in-the-*ss to do roll-down tests right. This is why I prefer a 10 mile, cruise control managed, benchmarks . . . it takes the variability of slight grade changes out and tests the whole system.

Bob Wilson

 

Thanks. I will look into that.

 

AFAIK the only way to update Gen1 ECU's is to physically swap them out with newer ones. Expen$ive at the dealer.
I can arrange loaner ECU's for anyone trying to solve a problem, I'll just need a deposit before I ship, and to know what version you have presently.

 

True regarding Classic. WeeDrgn has a 2007 Prius. We've got two different discussions going on in the same thread.

 

As a general rule, I don't loan tools without a deposit but in this case:

USPS tracking: 03120860000031337072​

I'm not doing anything with my miniVCI and I'm curious about what the OP's scanner can not see. With luck, he'll be able to report the 19 traction battery modules, four battery temperatures, MG1 and MG2 temperatures. He can also check the variable valve control operation and record some metrics. This will pretty much exhaust what we can learn about his NHW11.

Bob Wilson

 

Interesting thread. For what it's worth I've noticed this behaviour a few times in my car, a 2001 w/174,000 km. Fully warmed up with the car and ICE stopped, there will ocasionally be a slight lurch, which feels to me as though the motor is being rotated through just one or two compresssion cycles. You have to be paying attention to notice it. I read something here about this last year, but can't find the thread. In any case there are no other issues with the car that I'm aware of, fuel mileage is good, and it seemed plausible to me that the ECU might do this deliberately, so I've noted it and moved on. I first noticed it last summer shortly after I got the car, and the behaviour doesn't seem to have changed in the year/35,000 km I've driven it.

I defer to the wise old hands here, but I'd be surprised if there is any connection between this issue and changes in your fuel mileage.

 

Long time lurker (really long time, I've literally read every post here) finally checking in:

I observe the same symptoms with my '03 occasionally. In fact it's been going on for at least 6-8 years. I have a completely unrelated theory about the cause: the brake pedal.

Whenever I feel this shuddering, I can immediately stop it by firmly depressing the brake pedal. Sometimes when at a stoplight I don't depress the brake pedal hard enough (mostly out of laziness), and I suspect that I am holding the brake pedal at the exact threshold where the ECU activates its "simulated creeping". I think the car feels like it is shuddering because the ECU is rapidly powering and unpowering MG2.

Next time you observe this behavior, try firmly depressing the center of the brake pedal and let me know if that stops the issue.

 

I like this theory! It seems like the best explanation yet. Thanks.

 

Well, I finally noticed the 'Alerts' box in the upper right and am really surprised at this post because my wife thought it was the brake pedal not being pressed firmly enough and the car was actually creeping forward, but hearing how the car senses the brake pedal pressure and it can cause the engine to 'activate' makes some sense. But it isn't a complete solution to my issue. the issue I am concerned with is during the very first ICE run after pressing the start button. Yes, it can happen every time, even after turning it off for only 15-20min. And mine happens with it in a gear or in Park. It is fairly gentle - does not feel like it is causing any damage, it's just a minor oddity that I tried to have fixed but after three times to the Toyota dealer - nothing! Everything checks out perfectly... So, I am living with it. One of these days I my make a minor change and it will go away, hopefully I will know what I did.

Later, all.