I have an '87 GTi with CIS-e fuel injection, RD engine code. About a week ago it stared idling at about 1000 RPM on warm engine, up from the usual 850. There is also a slight oscillation in the idle speed. When I turn on the A/C the idle jumps to about 1500, much more than normal.
There is no problem starting the engine, cold or warm, and it runs fine aside from the high idle.
If I disconnect the electrical connector from the idle air control valve (or whatever they called it that year) it idles at about 1100 RPM and turning on the A/C drops the RPM somewhat, as expected.
If I remove the intake air hose to the idle air control valve and plug both the hose and the intake to the valve, the car idles at about 500 RPM. If I open up and let air into the valve the engine stalls as the mixture gets too lean.
With the engine off but ignition on I can hear the valve humming as expected. The idle air control valve is the type with three terminals. There is continuity between the center terminal and each of the to other terminals (about 12 ohms). I measured the average voltage at the test connector for idle valve duty cycle and found 10.0V at idle. System voltage is 14.0V. A very crude estimate of the duty cycle would be (14.0V - 10.0V) / 14.0V = 28%, so it is in the right ballpark. I don't have a dwell meter, so I'll bring out my scope tomorrow.
On a whim I got a replacement valve at a local junk yard ($ 4.00) and it initially seemed to fix the problem. Idle was normal, and turning on the A/C bumped up the idle a tad as expected, not up to 1500 RPM as before. However after a short drive (~4 miles) the symptoms reappeared.
I also squirted a fine mist of water around the intake system to see if I could find any air leaks. No dice.
First order of business tomorrow is to test the throttle position switches.
Now if anyone has made it this far in this long post, I have a couple of questions:
- With no power applied, is the three-terminal idle air control valve supposed to be open (i.e. allow air to pass)? My understanding is that un-powered it stays in an intermediate position and that one coil closes it and the other coil opens it fully.
- If I have an air leak from atmosphere to the intake manifold, this air would not be metered by the air flow sensor and the CIS-e system would not add fuel to get proper mixture. Such an air leak would thus stall the engine rather than increase the RPM. Is this a reasonable theory, or will the oxygen sensor loop sense the lean mixture and call for more fuel?
- Any other ideas about what I should look for?
Any and all help much appreciated.