"Idle speed cycling' problem

[johnbendik]

I've had a problem with my idle speed ever since a recent valve cover gasket replacement.  I originally thought it had to do with my SolVac (http://forums.amceaglesden.com/index.php?topic=45028.0, but now I'm not so sure.

When I start my engine, let it settle down, then blip the throttle, it drops into a state where the idle speed cycles (about every 2 seconds), starting from a 'normal' speed, then dropping much lower, then recovering back up to 'normal'.  The cycle takes about 2 seconds and runs continuously while I'm not on the throttle.  Sometimes the behavior stops for several minutes or so and then comes back.

If I turn on the lights or AC, the problem worsens to the point where it sometimes kills the engine (which is why I originally thought it was the SolVac), but I recently realized that it was cycling all the time, regardless of whether the electrical system was loaded or not.  Hence this new post...

I've seen other old (i.e., pre-computer) cars do this before, but it's never been one of mine, so I never had to figure it out.  Anybody got any ideas?

Thanks in advance,
John Bendiksen

[JayRamb]

Take GUMOUT carb cleaner when the engine is at normal temp. Raise hood, spray GUMOUT/carb cleaner around the air cleaner, back of engine, side / base of carb, front base of carb around your vacuum lines....ALL while the air breather is on. If you have a SURGE...higher idle, you have a vacuum leak. Start with vacuum.

[carnuck]

If the float level is slightly high, it will "bob", causing idle up/down. Or if the EGR is hooked to manifold vacuum, it will do it too.

[johnbendik]

JayRamb, that's just the kind of old-school tip I was looking for.  Unfortunately, I performed the test and didn't get any surges, so I guess I don't have a vacuum leak, at least in the area of the carb.

I'm going to get out my shop manual and check out Carnuck's suggestion about the EGR being ported to manifold vacuum.  It sounds like the kind of mistake I could have made while unhooking/hooking everything back up  during the valve cover removal.

Thanks, guys.

JB

[macdude443]

What is your curb idle set at?  According to the TSM, if the RPM drops below a predetermined RPM, the solv-vac will kick in and bump up the idle to prevent a stall.  While making adjustments, I triggered this a few times accidentally.  It seems to happen just below 500 RPM.

The AC/Headlights/Rear Defrost will trigger the Sol-Vac to bump the idle 200-300 RPM.  This is done by the "diode trio"/"diode pack" and will function without the computer intact.  Only the computer will sense the low RPM condition and itself bump up the RPM with the sol-vac.  You could try back out the sol-vac adjustment screw to take it out of the equation.  You may also have a an intermittent pass. fender vac switch which could be telling the computer false info about the vacuum signals and making it bump the timing around.  This would affect engine RPM.  The can be tested with a multimeter and a vac pump.  They are both normally open.  One close at 4" and the other at 10" nominal.

[johnbendik]

macdude443,

I was unable to follow some of your suggestions because I don't have a vacuum pump or a working tach (although I have two non-working ones installed!).  I did figure out some other stuff, based on your tips.

(1) My SolVac is being engaged continuously, whether the AC/lights are on or not.  I determined this by hooking my voltmeter to the hot lead of the SolVac. I had assumed it wasn't working because I never saw it move when I turned on the AC, but that's because it was already on ... The SolVac adjustment screw is now screwed in as far as it would go

(2) I was able to help the problem a little bit by screwing the 'curb' idle screw all the way in (assuming the 'curb' idle screw is the one closest to the fender, and not the inboard one that contacts the little white plastic cam).  This prevented the 'low' part of the cycle from dropping enough revs to stall the engine, at least until I try to use the lights or AC.

(3)  Also, when the idle speed is 'cycling', there is no movement of the throttle linkage or SolVac.  This suggests to me that the cycling is caused by the engine computer bumping the timing to prevent the stall, as you suggested.

(4) So it seems that something external to the throttle linkage is trying to stall the engine, and the cycling effect is just the engine trying to keep itself running.  Presumably this is also why the SolVac is engaged all the time.  Problem is, I have no idea what that 'external' thing could be... a vacuum leak away from the carb, maybe?  Is that the normal effect of a vacuum leak, to drop the engine revs?

(5) The other baffling thing is that the problem usually seems to cure itself after about 15 or 20 minutes of run time.  I just have to keep blipping the throttle at stop signs for those first 15 minutes.

Please feel free to pick apart my logic.  I'm getting a little discouraged here.

Thanks,
JB

P.S.  I just read the post from OregonCowGirl about stalling, and it talks about cleaning the 'idle tubes' in the carb.  I think my next step will be to fire it up with the air cleaner off and feed it some carb cleaner directly in the butterfly valve.

[macdude443]

If you need to clean the idle tubes, you'll need to remove the venturi cluster to get to them.  Remove the choke butterfly carefully, then remove the shaft.  Below (in the throat) you'll see two flat head screws with holes in the center.  Remove these.  The cluster will come up and out.  Careful with the gaskets and the pump discharge check ball.  The idle tubes are two long, skinny brass tubes with their holes facing downward.  These are the tubes that need to be cleaned and may in fact be causing your problem.  AMC recommended opening up the orifices to .032" with a #67 drill bit to prevent clogging.  When the car is running, clogged idle tube(s) will cause fuel to drip out of the upper nozzles.  You can easily see this if you look past the choke butterfly while the problem is ocurring.  There should be no dripping of fuel at idle.  You may also have an intake manifold leak that seals up once the car reaches operating temp and the metal parts expand.  Have you torqued the manifold bolts?