Power-Idle Hot-Cold performance puzzle

[daegrigg]

Continuing saga on my 94Q with 78K miles. It is not in my head - the car definately runs much better cold than when warmed up. Down shifting happens only with lots of accelerator travel.

So far I have:

Done all the MAF, cleaning, B44 stuffReplaced four injectors that ohmed too highNew NGKsPlugs look great except 2 and 4 that show too rich No engine codes (55) although CEL goes on in heavy traffic sometimes

What to check next that would fit the hot/cold differential profile?

Temp sensor - what whereO2 sensor - no codes, still possiblefuel pressureTiming

Thanks,

Don


Modified by daegrigg at 7:22 AM 5/19/2006

[goody90q45]

I have the exact issues with my Q. She runs great and goes like a bullet until at operating temp.. At that point I lose some throttle response ,some acceleration, and have a very mild miss at idle. I've done the same maintenance you've described trying to solve the issues. I also have one cylinder (#6) that is running a little rich. I'm getting the 32 and 34 code (EGR) after about 20 minutes in traffic but the EGR and passageway has been cleaned and checks out OK.

I wish she ran as good at operating temp as she does cold. I'll be following this thread to see what I can do to solve this problem. Good luck

[elwesso]

Best to take it to a dealer and have the consult done and have them print things out so we can see whats going on.

[96Qowner]

Huh, I was just thinking about this yesterday - mine, too. It runs noticeably stronger when it's cold. I've just been assuming I need a new thermostat, given Dennis' repeated statements about its narrow range of suitable operating temps.

Love to get a definitive explanation. IIRC, it's not all that unusual.

[elwesso]

i can tell you its normal because mine does not run that way, runs a lot stronger after its warmed up

[Q45tech]

Realistically at WOT the only differences come from reduced air density and ignition advance vs coolant temperature.

Really important to grasp how the air temperature changes from cold to extremely warm inside plenum and how 10% lower density [110F] can lower the power by at least 10%.

Important that coolant doesn't go above 195F-200F and that the ecu coolant sensor is ACCURATE.

"An increase in ambient air temperature of 5.6C [9-10F] increases the octane requirement of an engine by 0.44 - 0.54 MON [27,38]. When the combined effects of air temperature and humidity are considered, it is often possible to use one octane grade in summer, and use a lower octane rating in winter.

The Motor octane rating has a higher charge temperature, and increasing charge temperature increases the tendency to knock, so fuels with low Sensitivity ( the difference between RON and MON numbers ) are less affected by air temperature."

EVERY 10F warmer air needs 0.5 points of octane increase!

[duck3986]

This is exactly what was happening with my '98 at 100k miles. I replaced plugs and MAF, etc.

There were now trouble codes

Turned out it was my knock sensors. I replaced both, including the wire harness and it made a world of difference in performance.

I think when they start to fail, they don't neccessarily trip a code, but they must retard timimg as if it's protecting the engine from detonation.

[daegrigg]

Mine is not throwing any codes as generated from the ECU. Would the dealer power balance test be able to detect bad KSs?

[elwesso]

if the ECU doesnt throw a code then generalyl one can say theyre OK.. i guess the onyl other way would be to look at the ignition advance on the consult and see what you come up with.....

[goody90q45]

Have you ohmed the KS at the connector? It's not exact but it will give you an idea if they are working. The harness is next to the FI harness.

We're both chasing the same warm engine/reduced performance problem. I noticed a couple of months ago that I was missing the ducting off the bottom of the air box that feeds cool air from the intake path behind the headlight assembly. This means that I'm sucking in air from the engine compartment- Very hot air once the engine warms up.

I didn't realize this was a problem until I read Q45 Tech's post above. I've got the missing piece coming from PopPop. If you haven't already, you might want to check that your air intake ducting is intact from in front of the radiator to the air box. I'll post up if there's any improvement when I get the part installed. Good luck.

[Q45tech]

Even with entire oem air intake intact, you should measure the plenum air temperature [or at least on front of TB after the MAF].

Due to conduction thru all the plastic it will be 10F warmer than in front of air horn which will be 8-20F warmer than the radio reports [the in shade above ground temperature].

The air passing thru the intake valve [even after the vaporized gasoline spraying thru it] will be 30-40F warmer than the radio reports.

That could be 140F vs 40-50F in winter.

[duck3986]

I can remember lots of posts over the last few years regarding knock sensors failing between 75-125k miles.

They are mounted to the block in a very harsh environment and they provide the computer with the data to adjust timing to temperature and fuel conditions.

I'm a believer that they "degrade" not just fail.

If you're keeping your Q for more than a few months, I really believe it's money well spent, and I also think it will solve your performance problems.

Don

[daegrigg]

So what is the correct resistance for the knock sensor and at what temperature, hot cold, does it matter?

Don

[elwesso]

550k is the correct resistance but id say you probably have +- 100k to play around with to be acceptable...

Like I said if the KS arent throwing an ECU code theyre probably OK.

[daegrigg]

I checked plug number2 1 and 3 that had been running rich. Changed injectors and ran for about 2500 miles. No change in performance and the plugs look too rich as before. So much for leaky injector theory.

Rough idles and uneven performance hot/cold continues.

Today I decided to read the codes although when I checked a month ago, there were no codes, that is it was the 55 code.

Tonight however, the ECU threw code 34, the dreaded knock sensor code. Bingo, I said, even though the KSs are a pain to change. But then after turning the key off and resetting the ECU to normal, and setting it to read error codes again, the 34 code went away and back came the 55. Tried running the car for a few minutes at idle thinking that I had erased the code. The 34 code did not come back.

What did I do wrong. Also, can someone point out the connector harness for the KS. I want to clean the contacts.

Should I drive the car longer to generate the 34 code again?

Thanks,

Don