Mike Frye

History: Last year I did an intake refresh and in the process, chose not to replace my TPS. Bad idea. Since then I've had a problem with a surging idle upon startup (hot or cold). I was getting some funky readings (ok, no reading at all) from the aftermarket O2 sensor, so I replaced that: No change.

Then I metered the TPS and found that it wasn't giving a perfectly closed circuit when it was at idle (about 40 ohms or so when it should have been zero) and it wasn't resetting every time I let off on the throttle, even if I popped the throttle linkage off to isolate it from the gas pedal and cable.

Yesterday and today with the help of Adam G. I replaced the TPS and with the help of my new Porken Blink'r (Thanks Ken!) got the CO dialed in and the idle corrected, but it has a new 'feature'. It seems that it will start and idle fine cold or hot at almost exactly 700rpms unless I give it a little gas. Once I give it gas, it will only idle at about 1000rpms no matter how many times I blip it.

We also metered the TPS idle switch back to the LH and here's what I got:

Between 25 and 3 on the plug (idle contact circuit)
0.0 ohms at idle, 1.2k ohms just off idle

Between 25 and 12 on the plug (WOT switch circuit)
1.2k until about 2/3 throttle, then 0.0 ohms

When we disconnect the TPS harness plug at the front of the engine, there is still 1.2k ohms on each circuit (when read at the lh), but if we meter from the harness to the switch, we get a dead short or infinite (open circuit) as it's supposed to be.

Can someone tell me if there is something else in parallel with those circuits that is sending 1.2k ohms back to the harness and if it's supposed to be doing that?

I wasn't able to adjust the idle down on my car because it is up against the stop. When I first start it, it's fine. I can't seem to get it to return to that setting after blipping the throttle (again, removed the throttle linkage to isolate the gas pedal/cable as a cause).

Also, vacuum test shows exactly 20 (inches?) of Hg, rock steady, when I blip it, it goes down to 15, then back up to 20 and sits perfectly steady there.

So here are my issues: After startup, it will idle nice and steady (yay!) at 680-700 for as long as I let it, but once I tap the gas, it will only return to about 1000rpms.

The solver of this little brainteaser will be entitled to enjoy one very cold Guinness at my expense the next time I see you.

Anyone?

PorKen

You can check with a timing light if the idle switch is working. ~10 advance at idle, >20 off idle.

How old is the idle control valve? If you can't adjust your idle down, you have a vacuum leak or the ICV is stuck?

The throttle switch ground goes to the back of the engine (MPVIII). Check pin 25 to another ground.


Check at the EZF, too.
Pin 4 = idle
Pin 12 = ground (MPVIII)
Pin 17 = WOT


Also check at the diagnostic port (where the Blink'r plugs in).
_C_
B_A

* | LH |EZF| ?
A | 22 | --- | oxygen sensor pulses
B | 12 | 17 | high load throttle switch
C | 25 | 12 | ground (under airbox)

Mike Frye

Hey Ken,

Thanks for chiming in. I just figured out part of my problem, or rather lack of a problem.

The 1.2k ohms I was reading were coming from the EZF. I only had the LH unplugged and if you read from 25 to terminal 3 of the LH with the EZF plug in, you apparently read through that circuit board. Funny what happens when you read the directions .
Idle and WOT part of TPS are working perfectly back to the LH.

New ISV when I did the top end refresh and I tested it using a 12 vdc cordless drill battery, it seems to work fine. All temp sensors are new as of last fall as well.

I don't think I have a vacuum leak anywhere because the vacuum gauge is reading a rock steady 20 (and if I pull a vacuum line off, it idles up higher).

And by the way: WOW. The car is running great since the new TPS and dialing the CO in correctly using your Blink'r tool. Very cool! Throttle response seems noticeably better too. Don't know if the old TPS was causing resistance or the CO adjustment just makes it run more smoothly.

First we dialed in the CO on the MAF using a meter, then when the car warmed up and we used your blink'r it started flashing just perfectly. I also used it to test the WOT.

(I think you have a typo in your pdf though, you have terminals 4 and 6 on the MAF should be 382 ohms and my test plan says 3 and 6).

PorKen

Maybe the idle switch adjustment is borderline?

Is there a strong vacuum at the EZF hose at idle?

That's cool that a CO meter showed the same. I used a big shop unit when I was testing out the first Blink'r. What did the MAF adjustment end up at?

(Pins 3 and 4 are both ground - I'll change the text, though. )

Mike Frye

I'll go out and check the EZF hose, I don't have the vacuum gauge anymore (Adam takes all his toys when he goes home) but I'll see what it feels like.

So I guess your directions would give the same result. That's what I get for second guessing you.

I don't know what you mean by the CO meter, I only used your blink'r to dial it in. I think we got it to 380.2 before we put it in, and then we didn't have to adjust it at all after the car warmed up, we just flipped the switch, the light went out for about 4 seconds and then it started blinking away.

Mike Frye

Ken!

I think we're onto something here. I started it up without hitting the gas (idle at 680+/-)
Then I unplug the vac line on the EZF (I hear the vacuum pretty loudly)
If I blip the throttle with the vac line disconnected from the EZF, it returns to idle nice and smoothly. Then if I plug the vac line back in, the car idles up to 1000 immediately. If I unplug it again, it idles down to 680.

Also idle switch is perfect, dead short every time at the LH (when the EZF is unplugged!) and wide open when I push the pedal, returns to closed circuit every time without any coaxing.

OK, cool.


What the heck does that mean?

PorKen

I misread. You set the CO resistance using an ohmmeter to the basic setting of around 380, and that happened to work. I thought you dialed in the CO using an external CO gas analyser, then verified it with the Blink'r. Nevermind.

The vacuum line to the EZF goes to the 7-way splitter, which also goes to one of the carbon cannister (gas vapor vent) valves. Maybe you have a bad valve and/or bad hose, broken y-connector? (I don't know how this system fails, because I took the whole system off within a few days of buying the car.)

Adam Geist

Mike,

The carbon cannister control valve is the one on the passenger side fenderwell where we teed in the vacuum gauge. I have a Mity-Vac if you need to check the control valve for a vacuum leak. The "Y" connector that Ken mentioned is under the inner fenderwell at the cannister. The same connector was broken on my gold '86.5.

Now that I think about it, remember how the vacuum lines were misrouted on the gold car? IIRC the vacuum to the carbon cannister control is temp controlled through a temp switch under the air cleaner housing.
Adam

Mike Frye

We verified that there was no vacuum leak by testing for vacuum with it idling. The line that we teed off of came straight off the 7-way and was a steady 20 with no fluctuation. Doesn't this confirm that it's not a vacuum leak?
Adam, I don't see how that can affect the idle, but if you think it might be a factor I'll dig in there. I've had that fender liner out and physically inspected the y-connector, knowing it's plastic and a common failure point, it wasn't noticeably broken, but it might be leaking vacuum. I thought if it cracked you could smell fuel in that area as well. I didn't smell any when I was in that area.

Does anyone have any ideas about the clue we got last night?

When I remove the vacuum line from the EZF, it idles right down to where it belongs (actually a little lower, which means I could probably adjust the idle up a bit and not be at the end of the range on the throttle idle adjustment screw). So apparently something in the EZF is reacting to the amount of vacuum and advancing the timing or increasing the idle (so the timing advances). When I remove the vacuum, it drops down, replace the vacuum and it jumps back up.

toofast928

Worn throttle plate bushings? Or is the throttle bore out of round and the throttle plate hanging up on the bore not allowing it to close?

Engine and/or computer grounds clean to bare metal? Ohm the engine ground cable.

Measure the coolant sensor resistance at the sensor. Is it in spec? Then move to the computer and back probe the coolant sensor pins at the connector. Same resistance?

Tony

worf928

Caveat: my S3 advice is usually wrong

Just to be completely totally sure, check the at-the-fuse-panel vacuum behavior of the line at the elbow that would otherwise be plugged into the EZF. Do you see idle vacuum at idle? What about off idle? Return to idle? Just because you checked vacuum at another location doesn't mean that that's what's happening at the business end next to the EZF. I made that mistake once with an S4's flappy. Wasn't the solenoid. Wasn't the actuator under the intake. It was a collapsed vac line.

Does the line have a splice in it between the EZF and the 7-way? I've seen creative splice jobs on that line wherein the rubber connector used was too big and allowing a leak.

Have you triple-checked the vacuum line routing against the WSM/hood sticker? You probably have, but it never hurts to check the hood sticker and make sure.

PorKen

No leak at the EZF line, but what I'm saying is if the idle goes down when you introduce a leak into the 7-way, maybe something else which is controlled by vacuum from the 7-way is to blame. EG. the vapor recovery valve is opening at idle. It should only be at part throttle cruise because it is controlled by ported vacuum.

It could be a bad EZF? I can't think of how it could be, but it's easy enough to plug in a known good one, and/or checking the amount of advance with a timing light.

WallyP

Are you certain that the EZF knows that the idle switch is closed?

e.g., Is the EZF advancing the timing because it is seeing high vacuum and thinks that you are at cruise?

JP Rodkey

Did you check your throttle linkage/adjustment. It should always be tended to when the intake is removed.

Mike Frye

Wow, thanks everybody. I just got in. Let's see:

I'll have to check for the exact amount of vacuum, but there was definitely suction when I pulled it off at the EZK when at idle.
Double checked it, but I'll check again. In 5-speed there are only the 5 legs used plus a plugged one and they're all pretty clearly marked, but It couldn't hurt, thanks.

If I plug that vapor recovery leg from the 7-way would that isolate it?
I did check the advance, it's about 10 deg. when I first start up, then it's about 22 after I blip it and it settles down to 1000 rpms. But the idle switch is resetting every time.

I metered wires number 25 and 3 on the LH and it tests as prescribed. I need to test the ones on the EZF independently, but I know the contact itself is wired right and adjusted properly because I've tested it thoroughly at the plug in the front with the circuit isolated.

I even tried popping the ball of the linkage off at the console to isolate the pedal/cable/cruise control/console. The throttle plate is closing (or appears to be closing) completely and the idle contact is resetting completely.