Strange cold start

On my 97 Formula with the auto tranny at 37000 miles, the first cold start of the day, I've been noticing an RPM stumble immediately after it fires up (just one time). It goes from 1300RPM right after starting up but rapidly drops to 800RPM for 1/2 second, then goes back to 1200RPM and stays there until warm where it drops to 700RPM. Warm starts are always perfect and the idle speed is steady both cold and warm. Car generally runs awesome otherwise, never got an SES light, and just recently passed emissions/OBD2 inspection.

This started happening 5 years ago and can't seem to figure this one out. I had replaced the alternator at that time and am not sure if that is the culprit (reason I even suspect the alternator is because it started happening right around that time, and the volt gauge is higher at cold start than it was prior...on that line after 13 where it previously stuck between 13 and the next line). However, when I check with a digital volt meter it reads 14.65 cold idle, 14.1-14.2 warm/hot idle which I hear is perfectly normal.

Had the fuel pump replaced 5 years ago to take care of another issue (the cold start issue happened before as well, and I thought the pump replacement would've taken care of it but it didn't). Also had the fuel injectors cleaned, throttle body serviced, and the fuel filter replaced at the dealer (all done when the fuel pump was replaced).

I have the following mods: Hypertech PP+ and 160* stat, Moroso CAI, Airfoil, Spintech muffler. I tried removing hypertech and airfoil and nothing changed so I put them back in.

Any idea what might be happening? Right now the car still runs great except for the cold start (which after 1 second it runs/idles normal).

Possibly a problem with the coolant temp sensor in the water pump. The PCM uses that to set cold start A/F enrichment and elevated idle speed. Could possibly be indicating the engine is warmer than it actually is. Off, but not enough to set a code and turn on the SES light.

http://shbox.com/1/4th_gen_tech2.html#ect

Thanks, I think this was suggested in another forum as well but still unsure how to check this. I just don't want to replace the part and wind up not resolving the issue.

Quote:

---

Originally Posted by formula97

Thanks, I think this was suggested in another forum as well but still unsure how to check this. I just don't want to replace the part and wind up not resolving the issue.

---

Did you read the link I posted? Explains in detail how to test the sensor and the circuit.

I read through it, and was actually wondering if a scanner can do this or do i need to use a volt meter.

Also -- is this the same sensor that operates the cooling fans?

One last question -- is this also an "open loop" vs "closed loop" scenario? Reason I ask is because I've always had better throttle response in closed loop (warm engine) than open loop (cold engine).

A scanner would tell you what the sensor is reading. You would have to judge if the reading shown on the scanner is accurate or not. Compare to dash gauge, but those are not all that accurate. If it's a large error you might be able to judge it is inaccurate. If you determine it is inaccurate, then you have to follow the procedure to determine if the error is due to the sensor or the wiring, using a multi-meter.

Again, the answer to your question about the fans is in the write up. The 2-wire sensor in the water pump is the only one that connects to the PCM. The PCM controls the fans based on the sensor reading. The other 1-wire sensor in the driver side head only connects to the dash gauge.

Three things have to happen for the PCM to enter closed loop. 1) the pre-cat O2 sensors have to get hot enough (~600°F) to work, setting the left and right O2 sensor ready flags (observable with a scanner); 2) the coolant temp has to reach a programmed temp that is typically in the 120-140°F range; and 3) the timer from cold start, typically 206 seconds, has to time out. A faulty CLT sensor, reading lower than actual temp, could delay the transition to closed loop.

Thanks, I'll have to check the coolant temp sensor one of these days (today was a hot day in NJ at 90 degrees, and I still had that strange cold start).

If that coolant sensor checks out, I'm wondering if the MAF sensor can cause this to happen. Again, I only have this happen when first starting up for the day (it idles fine, runs strong, etc).

I also noticed a mechanical sound today during the cold start, sounding like it came from the brake master cylinder.

I was having a similar issue with my car, would crank and crank and finally sputter to life then idled like shit for minute or two then clear up and run fine. it only happened when it was cold, I figured it was sensor related as it had adequate fuel and spark. after seeing this thread I checked my temp sensor and the wires had some nicks in them near the connector. so I spliced in a new connector and replaced the temp sensor for good measure and the car fires right up first try every time.

One thing that came to mind about the ECT -- the fans kick in right when they are supposed to (I'm aware this is the ECT in the pump that controls the PCM, and not the one in the head that controls the gauge).

Glad you're up and running again.....my car always starts right up on the first crank, but it seems to have that brief drop to 800rpm for about 1/2 second right after the cold start, then rises to 1200rpm (idles smoothly the entire time during warm-up, and runs perfect once warm). Warm starts are always perfect.

Although this past Sunday my "cold" start was actually pretty close to perfect, so it seems to come and go. Engine is always more responsive once warm / closed loop.