Car starts fine
Idle drops after normal cold start
Then the idle will go back and forth from 1200 to 1500 rpm for a couple of
minutes and then stabilize.
Some days the idle is at 900 rpm and others 1400 rpm
Timing is dead on
TPS is green light go.
Grounds added.
Fresh motor, 1200 miles, the above same happened with old motor.
Suggestions?

 

It might be the thermowax pellet. I have the same problem with the idle pulsing when cold, but idle is perfect when warm. Haven't gotten around to it yet though.

 

I have come to the same conclusion. After the weather warms up i will install a modified throttle body and see if it goes away.

 

Thermo wax will never pulse... it is a simple spring.

Pulsing is simply the TPS or a vac leak... and I wouldn't trust using the light method of setting the TPS

 

believe it or not... airfilter.

 

I vote for a vac leak.

 

I've checked my TPS with a DMM and the twin light method and both are perfect. I have no vac leaks at all. Both those were confirmed at a local rotary shop. From visual inspection my air filter is pretty damn clean, if it wasn't i would be losing power. I'm really curious as to what this is.

 

If you unplug the TPS, does the problem go way???

If so, it is the TPS. Like it or not

 

Sorry to resurrect this oldie, but I am struggling with the old oscillating cold idle problem and even after taking your advice Mark, the problem remains. [That is, I uplugged the TPS while the idle was pulsing and the idle smoothed out so I purchased a new OEM TPS ($$$), installed it, adjusted it, but no joy.] Car is stock with modest intake and exhaust mods, S5 NA, '90 GTU. Compression tests excellent.

This problem only occurs when the cold idle drops to the 1.5K rpm level and then it oscillates +/- 200-300 rpm either side of 1.5K rpm until the engine warms up and the idle drops to a baseline of ~800 rpm. After that, the engine idles smoothly and runs strongly up to the red line in the lower gears. Additional information below. Would appreciate any other troubleshooting suggestions. TIA
-------------------------
Work completed prior to experiencing problem with oscillating cold idle:

Replaced primary fuel rail due to concern about potential PD failure (new OEM part)
Replaced fuel injectors with refurbished & tested units. All new rubber & pintle caps. (460cc injectors-OEM)
Replaced HP fuel hoses associated with fuel rails
Replaced MOP oil lines (all new “gaskets”)
Replaced most vacuum lines with silicone hoses
Replaced coolant hoses including heater hoses
All new gaskets used on BAC valve, UIM and dynamic chamber

------------------------------
Components tested, adjusted or replaced after oscillating-idle problem experienced:

Replaced throttle position sensor with a new, OEM unit ($$$)
Adjusted TPS according to both methods in shop manual
Adjusted idle to spec (initial set connector grounded, adjust screw at top of dynamic chamber)
Tested BAC valve
Tested air intake thermosensor
Checked for vacuum leaks or unconnected lines--visually and using starting fluid spray

 

...as far as I can tell, the 1500rpm is the point at which a fuel cut is issued. Ground (or short) the initial set connector, and set the idle speed. Too high and the 1500 fuel cut comes into play. If this doesn't fix the problem, then it's most likely a vacuum leak somewhere.

 

Thanks for the suggestion. I did set the idle using that technique. If there was a vacuum leak, wouldn't the idle be rough even after the engine was at operating temperature and idling at the baseline? Like I said, once the engine warms up, the idle is smooth at about 800 rpm. In any case, I did some checking for a vacuum leak using starting fluid spray in strategic locations.

I find it a very confusing as to what component/sensor is significant at what rpm point in the cold idle phase of startup, particularly during the mid part of the sequence of events leading to the baseline idle.

I'll look into the fuel cut aspect. Thanks.

 

i would think the reason small vac leaks smooth out at normal operating temp but are pronounced at cold start would be as a result of the expansion of various components when they heat up, swelling to eliminate the leak.

or not.. just seems logical

 

At that rpm the timing also advances itself. Just a hint.