slowly dies at idle

[azruss]

80 FI. computronix ignition, new gas tank, fuel pump, rebuilt injectors, new filter, rubber lines. blew out steel lines, rebuilt head with new seals and guides been running flawless for a year +. a couple of weeks ago, it would die when i dropped the clutch at a light. Thot i was just getting sloppy with the clutch. problem has gotten worse. If i sit at a light, it will eventually die on its own. Always restarts right away. It also wants to kill the minute i give it a little gas, like its getting air, but not fuel. Once going, it runs flawlessly. clean the plugs, they looked fine. valve lash was in tolerance. I will check timing and run some injector cleaner this weekend. Any ideas or suggestions.

[RRoller123]

80 FI too. Clogged charcoal canister (or clogged line from canister to tank) will do exactly this. If I put my finger over the air intake hole on top of the canister, the idle slowly drops and car would eventually die. No makeup-air getting into the system.

[azruss]

update: found the fuel pump was turning on and off at idle. in turn, this made the idle lower and lower making the fuel pump spend more time off than on. intermittent dual relay chatter but doesnt seem to correlate with the on/off. pulled the hose off the charcoal cannister without any change, AFM door is working perfectly. Turned the idle screw up a skosh taking my RPMS from 800 to 950-1050 depending on how long i've been idling. This seems to stop the chatter and the on/off. I still get random firing glitch at idle, so idle is never smooth or consistent. I can watch the light on the computronix and it is very consistent, so that tells me ignition is firing fine.

[FiatMac]

Turned the idle screw up a skosh taking my RPMS from 800 to 950-1050 depending on how long i've been idling. This seems to stop the chatter and the on/off.

Check to make sure the throttle position switch is adjusted properly.

[azruss]

This problem reared its ugly head last night for the 1st time since kicking up the idle. It seems to occur after lots of sitting at stop lights. This time, very little warning that it would happen. At this point I suspect the TPS. I checked it and it was showing a proper reading, and showed signs of being adjusted several times in the past. The screws were too worn to loosen, so i left it alone. I'm not sure how to test it as it appears to be an interminent problem. Dont particularly want to start swapping parts, but not happy with the car dying at night in the middle of the most accident prone intersection in town. Will pull the top of the AFM and look at the condition of the contacts.

[RRoller123]

Could be the AFM door spring tension setting was fooled with in the past or has tightened somehow, keeping the flap near the "engine off" position and allowing the arm to close the circuit and cut out the FP at low rpms. You could loosen the setting very slightly, just a couple of clicks, AFTER MARKING the present position, and see if it makes a difference. It can always be easily put back where it is presently.

[bradartigue]

Clicking TPS is always an adjustment issue. If it gives bad readings when adjusted then you replace the thing. But even a car with a clicking TPS (or no TPS) will typically idle, they just run like crap off idle.

If you crack the top of the AFM you should see the fuel pump contact engaged all the time when air is flowing. If not, don't adjust it - yet - go by the book and adjust both the idle speed and the throttle position screws. This is important because if you don't have it right then the air passing through the AFM isn't adequate to keep the pump running. Oftentimes people mess with the screw outside of the AFM because they think its a mixture screw - it isn't - it is an air bypass (as in, air bypassing the AFM plate), and if it has been adjusted too much then too much air gets by the door.

Don't forget if you adjust the idle screws you have to adjust the TPS, because the TPS rides on the throttle shaft.

[azruss]

thanks for the suggestions, guys. What i am struggling with is the car has been running and idling fine. The problem seems to be when i idle repeatedly at lights with less than a couple of minutes run time between. Only seems to get critical after several of these idle, short drive, idle cycles. I havent messed with the AFM door when the car is hot. I supposed that could cause sticking. When its cold, the door motion is smooth without any issues. The spring seems very light requiring very little effort to open the door. I did adjust the idle without adjusting the TPS. Is there anythink servicable in the TPS. Time to go though your FI tuning instructions with more detail, Brad.

[micbrody]

Did you ever figure out problem?
I am having similar issue. Car starts fine; isles fine. When I take for a drive, it drives fine, but when I come to a stop sign/light, the RPMs drop way lower than idle (below 300-500); it doesn't happen all the time.
My first thought was that I had some sort of minor leak in brake booster in that when I was stopping, I would be actuating brake booster, and engine would be exposed to unmetered air; but after many experiments driving, its unrelated to pressing brake pedal.
My next hypothesis was that oxygen sensor was sending signals that were not "healthy" for engine performance. I had just changed oxygen sensor from the one that came with car when I purchased a few years ago. A PO had definitely "screwed" around with AFM bypass screw (no cover). With original oxygen sensor I did adjust the screw to give me an average of 0.45ish on voltmeter.........however I realized I did this adjustment at idle. With new oxygen sensor, my readings were all over the place......but related o flaky wire from oxygen sensor (voltage changed with slight movement of wire). So I just installed third oxygen sensor. Voltage stable with wire tug.
I then proceeded to try to adjust bypass screw: I wedged some clump of aluminum foil at throttle screw to keep RPMs around 2000. I started screwing and unscrewing bypass. Seems to be within 0-1 volt, but lots of range, but I think it was close to averaging around 0.5 ish. I then removed aluminum foil so car would be at idle. The voltage dropped to almost zero. (0.05 ish). I ran out of time to screw around more; however my brain has been thinking about it (and I read a few more posts).
I have questions for experts on the board:
1) when measuring voltage, I disconnected oxygen sensor from ECU line and directly measured voltage relative to ground. Is this correct? Or should I have tapped into connection, leaving ECU to still see signal?
2) I started thinking more about throttle position switch. Is purpose of switch to tell ECU to "pay attention" to oxygen sensor during throttle up only? But when switch at rest (idle), don't look at oxygen sensor (meaning engine doesn't care if it is running lean at idle); and at full throttle, ignore oxygen sensor and let engine run rich? All this can be summarized: is it normal for oxygen sensor to show lean conditions at idle, but 0.45ish only at throttle up? Or should it always be 0.45ish?
3) assuming I do get bypass screw "dialed in" correctly , will everything change if I reset idle with throttle body idle screw?

[RRoller123]

T off the O2 sensor, leaving it connected, and go through the readjustment process for the TPS and that may help, based upon what you have said.