My '85 MGM, "Maisa"

[87gtVIC]

Ive always grasped them with bolth hands..tucked fingers under as much as possible and pull it off. Strong metal spring clips holding them in place. If you keep your hands a bit apart you will remove them without issue.

[gadget73]

if there is an "oh shit" handle, remove that first. I don't remember if GM's got those or not. Screws will be under the end caps for the handle.

[Arquemann]

Got the car inspected today, passed without issues! (kinda)

An old issue comes back to haunt me: running rich. On idle (park in this case) it runs very rich, and as such the CO2 values are super high and it wouldn't pass emissions. But again, it's only on idle, inspector said that as soon as you touch the throttle, the CO2 comes down. And so, I passed the emissions. (also no tach so the idle can't be "confirmed", so it's a bit more lenient)

Last years inspection was totally fine, and the only thing that has changed from last time is the IDLE RPM, that I meddled with recently, since it was quite high, and has been for a long time (including last year's inspection)

I still remember the mechanic from 2019 saying that the car runs really rich on idle, and at that point the car still had the running issue, which also caused the idle to be quite low.


After coming home, I readjusted the idle again, since it was left a little bit too low last time. Today I even had me an RPM gauge. It was about 450-500 rpm in gear from last time, I set it to about 600 rpm now, it bounces around probably ±50rpm. It's really shitty to adjust the dashpot, even though the "sled" moves smoothly, but the adjustments go all over the place. Took me three tries to get it somewhat better and actually stay there. It seems the dashpot get vacuum sometimes and sometimes not at all, regardless of in park or gear. All adjustments were done with the ac off.



Those Holley EFI units really looking tasty...

[Arquemann]

Seriously, how fucken' dumb is the CFI ECM, if the O2 sensor says rich, and the ECM doesn't shit about it?

Can I somehow get a readout from the O2 sensor? Or does some other sensor reading take priority over the O2? MAP sensor maybe?

[matth825]

My '85 MGM, "Maisa"

It’s a pretty dumb ECU. I believe it switches to open loop at idle since it’s an unheated sensor. Not enough exhaust flow at idle to sufficiently warm it to operate properly.

You’d have to use an oscilloscope to view it’s output. When in closed loop the ECU will mess with the fuel trim to get the sensor switching back and forth between around .2-.8VDC. If it’s flatlined low the engine is running lean and flatlined high the engine is running rich. You’d need a wideband O2 to know your exact AFR. A narrowband only tells you when you are close to 14.7:1 as indicated by the switching.


Sent from my iPhone using Tapatalk

[Arquemann]

You reckon the O2 sensor just doesn't get enough airflow at the proper idle speed and is designed to go open loop? And the overly high idle RPM I've had has meant it's been operating in closed loop on idle? Buuuuut, do these CFI cars go to open loop when decelerating?

Since this car bucks on decel, might that have something open loop related issue aswell?
O2 sensors are new, and so are all engine sensors except the MAP. And MAP is one other sensors used in most open loop modes IIRC.

We might be on the verge of something great here, or on the verge of a certain CFI unit flying into a bin...

EDIT: I do remember testing the MAP sensor some time ago with a multimeter and the readings were correct...

[gadget73]

you can read it with a digital voltmeter but its just going to bounce back and forth if the mix is right. If it sticks high or low, that means lean or rich.

Switching to a heated O2 is also not real hard. The stock one is 1 wire, heated are 3 or 4. 3 wire are a signal out where your single goes, a ground and a switched hot for the heater. 4 wire have the signal out, signal ground, and the two for the heater. The 4 wire is really the better mousetrap since it doesn't rely on the exhaust pipe for the ground, you can tie it directly to the ECM sensor ground wire.

Later cars tap the heater off the same +12v that feeds the injectors and ECM and whatnot. Whether or not that would actually fix anything I have no idea. Its a case of I doubt it would make it worse, but thats about the most I can say.

[matth825]

My '85 MGM, "Maisa"

Correct—-at a normal curb idle the O2 cools off due to lack of exhaust flow. The engine would need to run around 1000-1500 rpm to sufficiently heat it. It needs to be approximately 600 degrees F in order to produce a signal. You’ll know when it does open loop if you have the hood open because you’ll hear the thermactor air bypass valve start dumping smog pump air to the atmosphere. I would also imagine it goes to open loop under decel for the same reason…throttle is closed so minimal exhaust flow.

What the exact AFR the computer targets at a hot idle I’m not sure but I’m almost certain I read it was on the rich side. Ford published a CFI training manual that explains the convoluted mess that is the CFI ECU but I’ve never been able to get my hands on the correct one.

Gadget’s got a great idea substituting in a heated O2. That would make a big difference.

How does the emissions testing in Finland work? Are they checking it to modern standards or 1985 US EPA standards? I remember emissions test way back when on my Dad’s 1984 Caprice with the feedback carb. It was never checked at idle speed. The tech would use an inductive tach to run the engine up to about 1500 rpm for a minute or so before he looked at the results of the tailpipe sniffer.

One last thought…how high was your idle speed before you corrected it? The ECU messes with the ignition timing to try to adjust it. I would think the ECU would pull timing to slow it down which could possibly result in rich exhaust since the fuel/air charge has less time to burn in the cylinder.

Also, anytime idle speed is adjusted via moving the dash pot sled a procedure needs to be followed to “teach” the computer the new idle speed target. I believe it’s on the emissions sticker on the fan shroud.

Sent from my iPhone using Tapatalk

[Arquemann]

When the system goes open loop, does the O2 sensor reading just go blank? Since that would make it pretty hard to check if its rich or lean at idle...
Would splicing in a heated O2 prevent the system from going open loop, if the O2 sensor keeps giving a signal? Or is the loop mode change most likely a combination of sensor outputs, such as TPS and MAP?

Since this is an '85, it has to pass the emissions requirements for '79-'86. The testing is done at idle in park/neutral (supposed to be under 1000rpm) and the necessary values are CO% (max 4,5) and HC ppm (max 1000). Then there are CO2 and O2 measurements, but those aren't needed here.
I made an error on an earlier post, it isn't the CO2 value that was problematic, it is the CO%. Inspector said the percentage was around 7-8% at idle, but dropped significantly with very light throttle.
HC ppm passes by miles and the CO% drops very low with throttle.

Last year on the left, yesterday on the right.

I cannot guarantee exactly accurate readings on the paper, since the inspector wanted this to pass and generally there's some bullshitting involved with getting an old car inspected with a familiar inspector...

I'd reckon the idle speed before me faffing with it was about 700rpm in gear, now it bounces around 600. After my first attempt to adjust it, it was too low, about 500 in gear. Even now with the rpm in gear being a bit better, the idle in park goes a bit over 1000rpm.

The CFI idle set procedure doesn't seem to include anything about "teaching" the new speed. Gadget linked the Mustang idle adjustment workshop manual last page, #278. The emissions sticker on the car tells pretty much the same story.
Only slightly worrysome symptom of the too low idle was OD coming on a bit later than usual. I have no clue of the AOD rod adjustments, and I haven't touched it myself.

What comes to the ECM trying to lower idle by adjusting timing: When I unplug the SPOUT, the engine tone changes, and I'm pretty sure the RPM decreases.

[gadget73]

closed loop is when it reads the sensor.

The O2 is always going to output something, but when the ECM is in open loop it just doesn't pay any attention to it.

The heater wouldn't affect the ECM going closed or open. The computer doesn't even know its there. The only thing the heater does is make the sensor read accurately more quickly. It heats the thing up so it does its job better. What turns it to closed loop operation is the coolant temp sensor hitting some value. I think its 160F, but it may be some other number.

[Arquemann]

Yeah, I got that right, I think.

So the ECM goes open loop (no O2) on idle, since the O2 doesn't get hot enough. Also the same for decel? And from gadget said, on warmup aswell.
But what triggers the open loop mode on idle or decel? MAP, TPS, so on?

Basically the (heated or not) O2 won't do jack shit about my rich idle. But what would do something to it? It's got new injectors, pressure reg, new TPS, pretty much every sensor is new, except for the MAP, which did check good last year... It's even got a new ECM, which is supposed to be correct aswell. What could possibly cause the ECM to go rich? Or is this shit really just how it's supposed to work?



Other news, I got the tires rebalanced today, which was a trainwreck in and of itself. The car had a bit of a vibration at 55-65 mph, which has worsened slowly from last summer. Honestly it wasn't as bad as the tires looked.
Anyways, front tires have awful inside wear, and one had visible run-out, most likely the cause of the vibration.
They are better now, but they recommended to get atleast one new tire, the tires are quite low miles, but 5 years old.

[DerekTheGreat]

I remember adjusting idle on my '85 well. Yes, you will need to readjust the TV linkage pretty much anytime you mess with the idle. THere are two adjustments for this, a fine tuning one near the throttlebody and a gross adjustment down at the trans. Depending on who was there before you, you might be able to get away with the fine adjustment by backing off the screw. If you're like me, you'll have to adjust down at the trans and then up at the fine adjustment.

[Arquemann]

I remember adjusting idle on my '85 well. Yes, you will need to readjust the TV linkage pretty much anytime you mess with the idle. THere are two adjustments for this, a fine tuning one near the throttlebody and a gross adjustment down at the trans. Depending on who was there before you, you might be able to get away with the fine adjustment by backing off the screw. If you're like me, you'll have to adjust down at the trans and then up at the fine adjustment.

I have absolutely no idea on what basis to adjust it, nor do I have a suitable pressure gauge even do it. The adjustment screw on top is pretty much in the middle of its travel.
After yesterdays idle adjusting, the trans shifts smooth and seemingly in time without jerkiness and the OD seems to act just like before. Since it feels the same it has felt for the last two years and I'm rather inclined to leave it the fuck alone...

Buuuut, if you have some smart parameters for adjusting it and it not being complicated as hell, sure.

[matth825]

Glad they are using the appropriate standards to gauge the emissions. If the HC’s are low at idle then I would think the cats are working properly which is good. It might just be CFI funkiness that’s causing it because I don’t know what else to say to check at this point. At least they passed you though so that’s a positive.

I’ll have to look at my emissions sticker tonight but i believe part of the idle speed set procedure is to run the engine at 2000rpm for one minute. I believe that is what triggers the idle speed learning mode in the ECM. Whatever is sees for idle speed after that become the idle speed target.


Sent from my iPhone using Tapatalk

[gadget73]

IF the rich idle is caused by low O2 sensor temperature, then a heated O2 may help with that. I kinda have my doubts if the engine is well warmed up though.

It should be in closed loop at idle though. If there is some problem causing the mixture to be rich beyond what the ECM can correct its just going to be rich. I'd expect a code about it stored but maybe not. The other possibility is a false reading on the O2 sensor from the smog pump injecting air through the back of the heads. The sensor will read that as a lean condition and add fuel to compensate.