View Full Version : Drifting Idle

05-22-2008, 07:50 AM
Idle is set for 750 RPM. When first starting the boat the idle is 1000RPM plus. Allow it to run a few mintues, turn the engine off, restart, the idle drops to 750 RPM. Pull a skier through the course and the idle is high at the other end. Shut down and restart the idle drops again.

1995, Prostar 195, TBI, 250hrs
Perfect Pass Digital Pro

Checked the throttle cable for binding - all okay.
Checked the PP servo cable - all okay.
Raised the engine cover, accellerated to about 1500 RPM, returned to neutral. The throttle returned to position.

Any other ideas?

05-22-2008, 07:52 AM
Throttle Position Sensor? I'm not sure where it's located on the TBI but sounds like it could be going bad.

05-22-2008, 08:10 AM
Can that sensor be tested using an ohm meter?

05-22-2008, 08:36 AM
The idle speed of injected motors is 600 RPM, not 750. If it drifts up, it could be due to a faulty IAC motor or some other air leak. Remove the throttle cable and see if it drops to 600, increase it to some higher point and let the throttle lever go to see how far it drops. If it still doesn't drop to 600, check the MAP sensor tube for cracks and looseness on the base of the throttle body and the sensor itself. If it's cracked and/or loose, repair or replace the tube. Next, inspect the PCV valve and the hose. You should have good vacuum and the valve should move freely, same as on a car.

If the idle returns to 600 at idle just by removing the throttle cable, it may be binding or out of adjustment. It's easy to adjust- loosen the lock nut and rotate the end to the correct point, then tighten the nut.

Besides, if the TPS was bad, it should throw a code and you should see the check engine light go on. A meter will show an open or short but if it's actually glitchy, usually not. That requires an oscilloscope to see short duration drops/peaks. If the ECM sees >2%, the MAP sensor and RPM are what mainly indicate engine load and the TPS really doesn't do much unless the throttle position rapidly changes by +/- 20%, or more.

There is a function called IAC follower, which means that the ECM tells the IAC to open more on hard acceleration and close somewhat on deceleration (it "follows" the opening and closing of the throttle). A sticky IAC or bad connection will cause the IAC to not close quickly or far enough, since IAC open is one pair of wires and IAC close is another. Also, when you shut the engine off, you should hear a faint buzzing at the IAC, which is the ECM closing the IAC completely and re-opening it for start-up. This is called "parking" the IAC and if this works, it can mean that the ECM and IAC are working properly.

05-22-2008, 02:07 PM
Called the local MC Shop. A TPS would cost $90.00. I'll try the tests you described in the first paragraph and see what I get.

05-22-2008, 05:07 PM
OK, now call a Chevy dealer and tell them what it's for. If they say they don't have any info on that motor, tell them it's the kind used on a '90s pickup with the 5.7L motor. AFAIK, the only one that's different is the one used in the LT-1, which shows WOT when this one shows 0% and 0% when this one shows WOT.

05-23-2008, 11:34 AM
The problem is with the cable. I removed the cable and revved the engine. Then I let go of the butterfly. The RPMs dropped steadily all the way to 750. I adjusted the position of the cable and it worked just fine.

The idle is still 750. I don't really know how to adjust the idle. The PP servo cable connects to the throttle cable. I see a set screw that stops the butterfly from rotating any further. I don't see how to adjust that screw.

05-23-2008, 01:00 PM
If you can use a different tach, see if there's a difference. The adjustment for the throttle plate is set at the factory and isn't supposed to be field-adjustable.

If the air temperature is lower, this has a high idle, just like a car. If the idle settles down after it warms up, there's no problem with it. If the cable is binding, ideally, it should be addressed.

Just for grins, loosen the hold-down bolt on the distributor, clean the back of the head, the small plate and the metal at the base of the distributor, but don't move the position of the distributor before re-installing the plate and bolt. If this causes the tach to read differently, it would mean that the corrosion was affecting the tach and over time, any bare metal can corrode, so it's normal.

If this doesn't change anything, and since it hasn't been mentioned previously, have you checked the timing? If you did and set it just by hooking up a timing light, that could be part of the problem. Base timing can only be set by putting the ECM into service mode.

05-23-2008, 03:16 PM
The tach on the PP and the tach are in sync at 750rpms.

This is my first ski boat with fuel injection. How do I put the ECM into service mode? Haven't messed with timing at all.

05-23-2008, 04:41 PM
Do you know about the paper clip method for GM cars and trucks? It works the same way. Then, you need to raise the RPM to 1000 to check and set the advance.

If you don't know about the paper clip method, look for the data link connector (looks like the one in a car) on the rear of the motor and remove the black plastic cover. Bend a paper clip into a tight U and insert one end into the A terminal and the other into the B terminal. Turn the key ON and watch the check engine light flash. It will flash once, then twice with a brief pause. This denotes 12 and it will repeat 12 three times. If you only see 12 flashed, you have no stored codes. Write any other flash combinations down, just in case. At the end of any stored codes, it will flash 12 three times. Indicate any other codes here if you need to know what they are.

Once the paper clip is in place, start the motor and set it to 1000RPM. Check the advance and if it's at 10BTDC, you don't need to change anything, but if it idles at 750, I suspect that it's probably closer to 20BTDC. Back it down to 10, snug the hold-down bolt and re-check. If it hasn't changed, tighten it to about 15 ft-lb.

If the idle speed doesn't change and the timing was correct before, look around for a slight vacuum leak. I already mentioned a couple of the most likely but also look at the throttle body base, where it mates with the intake manifold. You can spray carb cleaner along the gasket edges to check for leaks- if the RPM changes specifically from the spray, you have found the leak.

If you find no leaks or other causes, think about finding a way to hook it up to a diagnostic computer. That will show MAP (vacuum pressure in KPa and also the voltage from teh sensor), TPS, they can re-park the IAC and also function it, raise/lower RPM to almost any speed, see the voltage as seen by the ECM and the sensors' perspective, read codes and also check for knock.

Personally, if I was in your position, I would do a complete survey of the boat, just to have a baseline because it's not a new boat and you aren't the original owner. Having a hard copy of what the engine is doing and its condition is a good thing and I would include data from fuel pressure, compression/cylinder leakdown and vacuum tests. If you end up seeing anything that isn't quite right, you can avoid trouble later, too because you were able to see it without having to deal with a failure on the water. A few hours spent now can save a summer.

BTW- connecting to a diagnostic computer will also show total hours of run time on the motor. If you don't know the complete history of this boat, trusting the hour meter isn't something I would do. That can be swapped for a different one or disconnected very easily and I have seen boats that showed low hours on the meter but higher hours in the ECM. 250 isn't many hours on a 13 year old boat. Averaging less than 20 hours per year seems too low, to me.

05-29-2008, 10:04 AM
Roger, thanks.