Hesitation on Acceleration and Idle Stalling

[scarceller]

You are correct that the hose goes on the other side of the throttle body but it also goes after the AFM so when you open the oil cap you allow false air (un-metered air) to by-pass the AFM and this leans the mixture. This is why you should hear a difference in RPM at idle when you open the oil cap. Some cars even stall. The bottom line is that at idle even though the throttle plate is closed air is still moving through the AFM and when you open the oil cap you allow some of the required air to bypass the AFM and thus lean the mixture. You can also simply try to cap the port at the back of the rubber boot or remove the pipe from the filler neck and plug it with your hand when you do this the mixture will richen then remove hand it will lean. Hope you get the idea. If the oil tank is for some reason not air tight then you have an issue.

Originally Posted by bhoff

Scarcellar, when I look at the large hose coming off of the oil tank to the rubber boot, I see that the boot is above the throttle body so there can be no vacuum. To have a vacuum, it would need to be below the throttle body. The smaller hose that goes to the valve shown in the photo should have a vacuum because it attached below the throttle plate. I do not know what this hose does but it appears to be attached to a PCV valve? When I take off the small control vacuum hose on the top, there is no vacuum on this hose. I do not know if and when this valve is suppose to be open or when there should be a vacuum on the control line. I cannot detect any vacuum when I place my hand over the oil filler and like I explained, I cannot see how there could be a vacuum here virtually being open to outside air pressure by being attached above the throttle body.

As to the DME, I did swap out a known good one last summer for a couple of weeks (prior to changing the O2 and flywheel sensors). I also did a visual inspection of the solder points on the inside recently.

[bhoff]

Thanks Scarceller, I was originally thinking vacuum in the oil tank but now understand opening the tank would allow unmeasured air into the system. I have been playing around with the hose off of the tank plugging it with my hand and uplugging. I cannot detect any change in the engine. Plugging the hose when removed should simulate a sealed oil tank.

I have not clamped off the power assist yet as I think I will need to remove the AFM again to get to the hose. Previously I did tighten the clamp on this hose as it was pretty loose and most likely leaking. No change after tightening.

I still do not understand what the vacuum assisted valve in the photo I have attached is for. As I stated, there is no vacuum at the small control hose on the top. There is vacuum on the larger hose that attaches to the valve. I have not checked the other end that attaches to the tank yet as the clamps are not removable.

[scarceller]

The large hose that you are putting your hand over has a metal funnel restricter in the hose at the end that connects to the oil tank. Check to see if this restricter is in the hose and that it has not shifted. I suggest you remove the entire hose to be sure it is clear and air can get through it. Plugging and un-plugging this hose always effects idle so something is up? You coul also remove the hose from the back of the rubber boot and plug un-plug the connector at the boot to eliminate a hose issue.

As for the vacuum valve you refer to I'm not certain how it works or what gives it vacuum to open the valve but the purpose is to bleed off oil fumes from the tank.

Originally Posted by bhoff

Thanks Scarceller, I was originally thinking vacuum in the oil tank but now understand opening the tank would allow unmeasured air into the system. I have been playing around with the hose off of the tank plugging it with my hand and uplugging. I cannot detect any change in the engine. Plugging the hose when removed should simulate a sealed oil tank.

I have not clamped off the power assist yet as I think I will need to remove the AFM again to get to the hose. Previously I did tighten the clamp on this hose as it was pretty loose and most likely leaking. No change after tightening.

I still do not understand what the vacuum assisted valve in the photo I have attached is for. As I stated, there is no vacuum at the small control hose on the top. There is vacuum on the larger hose that attaches to the valve. I have not checked the other end that attaches to the tank yet as the clamps are not removable.

[david914]

I had similar issues with my '86. Even though everything checked out, I finally took the AFM apart and repositioned the wiper arm (as per instructions in the link that stlrj posted) and the problem went away. The AFM seems to be the weak link in the DME system.

[bhoff]

I adjusted the idle mixture screw last night and saw a substantial improvement in idle. Turned it counterclockwise 1.5 turns to lean it out as it seemed to be very rich with occational black smoke on startup. Cold starting is still an issue. What really throws me is that I can cold start and it will stumble and stall. Turn off key and start again and it will do something completely different like bounce between 1,000-2,000 rpm. Shut down again and it will backfire and blow black smoke. Shut down again and it will run normal. I cannot seem to get any consistancy to the symptoms. I think i really need to have the mixture set correctly by a professional that has the right CO measurement equipment.

I am also certain that I have two separate problems as you guys stated. The stall out is still there. I have narrowed it down to hard acceleration below 2000 rpm. This means if I accelerate around a corner in 2nd gear, the engine bucks hard. Medium to low acceleration is no problem. This will happen on a straight so it has nothing to do with cornering.

As to the AFM, I checked it out and found it to be ok with a very linear voltage change across the entire range with no drops. Opened it up and saw no wear on the wiper path at all, in fact could not even see a path.

[scarceller]

Check and clean the ground screw on cyl #1 intake runner (LHS Rear) you'll see a bunch of wires grounded here, these are all the main grounds for the DME and all sensors. Also check and clean the ground strap at the battery and the other strap at the GearBox. I have seen ground problems cause all sorts of unpredictable issues.

The other thing is your spark plug cables, are they stock cables? Also be sure they are not arcing, check them at night in the dark to look for arcing. The DME does not like any high voltage EFI noise.

[bhoff]

I was able to check the O2 sensor and set it a the correct 0.2-0.6 range (fluctuating). I was set pretty lean after adjusting earlier today without checking voltage. Will see how it runs. Can I correctly assume that I have no major vacuum leaks because I was able to set the CO reading to the correct voltage?

Thanks for the information on the ground points Scarceller. I had previously checked the ground screw on the #1 intake. Been a sail boater for some time and use to tracking down electrical corrosion. Did not know about the gearbox ground strap.

[scarceller]

If you can get AFR or CO correct then you don't have any major air leak, if you did you could never set this correctly.

You can use the existing stock O2 sensor to set mixture in the ballpark.
1 - Warm engine
2 - disconnect O2 sensor but leave O2 Heater connected if possible
3 - read voltage at the O2 sensors disconnected output
4 - turn AFM CO screw clock-wise (richer) till O2 reads >0.7v here you are rich.
5 - now back out screw counter-clockwise (but count turns as you back it out) till the O2 reads <0.3v, now you are lean.
6 - turn screw back in to 1/2 way point then turn another little bit to slightly keep mixture on the rich side. You do this because it's always best to have closed-loop mode lean the mixture than richen it.

The above procedure works well if you have no WBO2 or CO meter.

If the car has a major air leak then you'll find that the CO screw will be turned all the way in and the car is still lean, this is a red flag that you may have an air leak. But it could also be a faulty AFM.

As for the ground strap on the gear box you'll find this right at the back of the tunnel near the front of the transmission.

Originally Posted by bhoff

I was able to check the O2 sensor and set it a the correct 0.2-0.6 range (fluctuating). I was set pretty lean after adjusting earlier today without checking voltage. Will see how it runs. Can I correctly assume that I have no major vacuum leaks because I was able to set the CO reading to the correct voltage?

Thanks for the information on the ground points Scarceller. I had previously checked the ground screw on the #1 intake. Been a sail boater for some time and use to tracking down electrical corrosion. Did not know about the gearbox ground strap.

[bhoff]

Scarceller,
Thanks for the good information on setting the mixture. I was searching this forum for procedures and yours is the easiest to follow. I attempted to use this and am having difficulty finding a point where the voltage is stable enough to locate either the high or low point. I did not have the O2 sensor heater hooked up so this may have been some of the problem. All I could do was find a lower range and higher range. Basically it would fluctuate from about 0.1 to 0.8, which is about the entire range. I could find a point where it would stay more in the lower range and another point where it would stay in the higher range. I started with the temp just past the first mark and towards the end of checking, the engine was getting to the mid point on the oil temp so it may have been too warm. Voltage seemed to be changing as the temp went up as well. I also did not have the idle control valve parked as I think this was only for setting idle. In hindsite, maybe the idle valve should have been parked because that would change the mixture no?

[scarceller]

The narrow band stock O2 sensor is like a switch it goes to >.7v or <.3v you won't be able to get voltages in between these values. I should have mentioned this. Just turn in CO screw till you see the O2 saturate at >.7v then count turns while you turn the screw out till you see the sensor drop to >.3v then just screw back in 1/2 the number of turns, your done.

Of course this is just a cheap way of setting mixture in the ballpark, at some point I suggest a proper mixture setup with a WBO2 or a CO meter.

You do not need to park the idle valve to set mixture, you only do this to set base idle. Also when setting base idle always set the base 20-40RPMs higher than the desired target RPM. Most cars (86-89) idle at 880RPM so when you set the base set it for around 900-920RPM, just a tad higher than the target and let the ICV pull the idle down to 880. This way if the ICV fails the car will idle slightly higher and not stall.

Originally Posted by bhoff

Scarceller,
Thanks for the good information on setting the mixture. I was searching this forum for procedures and yours is the easiest to follow. I attempted to use this and am having difficulty finding a point where the voltage is stable enough to locate either the high or low point. I did not have the O2 sensor heater hooked up so this may have been some of the problem. All I could do was find a lower range and higher range. Basically it would fluctuate from about 0.1 to 0.8, which is about the entire range. I could find a point where it would stay more in the lower range and another point where it would stay in the higher range. I started with the temp just past the first mark and towards the end of checking, the engine was getting to the mid point on the oil temp so it may have been too warm. Voltage seemed to be changing as the temp went up as well. I also did not have the idle control valve parked as I think this was only for setting idle. In hindsite, maybe the idle valve should have been parked because that would change the mixture no?

[AlfonsoR]

Originally Posted by scarceller

The narrow band stock O2 sensor is like a switch it goes to >.7v or <.3v you won't be able to get voltages in between these values. I should have mentioned this. Just turn in CO screw till you see the O2 saturate at >.7v then count turns while you turn the screw out till you see the sensor drop to >.3v then just screw back in 1/2 the number of turns, your done.

Of course this is just a cheap way of setting mixture in the ballpark, at some point I suggest a proper mixture setup with a WBO2 or a CO meter.

You do not need to park the idle valve to set mixture, you only do this to set base idle. Also when setting base idle always set the base 20-40RPMs higher than the desired target RPM. Most cars (86-89) idle at 880RPM so when you set the base set it for around 900-920RPM, just a tad higher than the target and let the ICV pull the idle down to 880. This way if the ICV fails the car will idle slightly higher and not stall.

Question here: from what i have read, the "voltage" method gets close enough, but as typical for a lot of 3.2s, you will have idle hunting as the DME/ ECU try to get to the correct lambda value? Can you explain why the hunting? I don't understand that.

I did read where someone, was able to nail the A/F and CO readings to where the car idled at a rock steady 880 RPM. I would like to do this to my car when I get it....maybe 2-3 more weeks. Thanks.

[scarceller]

Hunting often is related to lean mixture. If you set the the CO screw as I describe you are real close to correct CO but it's always best to be on the rich side than on the lean.

Idle in the 3.2L is maintained at 880RPM via the IdleControlValve and if it is working properly and clean these cars idle perfectly at 880RPM.

These cars have one small issue known as idle dipping when you come back to idle from higher RPMs (like rev to 3000RPM then let off the gas). Often the idle dips below 880 and comes back up. This can be fixed via a chip change, this is something Steve Wong's chips have improved. The issue is because of how the fuel cut on decel is setup in the stock chip.

Originally Posted by AlfonsoR

Question here: from what i have read, the "voltage" method gets close enough, but as typical for a lot of 3.2s, you will have idle hunting as the DME/ ECU try to get to the correct lambda value? Can you explain why the hunting? I don't understand that.

I did read where someone, was able to nail the A/F and CO readings to where the car idled at a rock steady 880 RPM. I would like to do this to my car when I get it....maybe 2-3 more weeks. Thanks.

[Rick968]

Hi.
Am experiencing a lot of what is discussed here.
I just bought a grey market 1984 targa. Did not ship with a Cat.
In fact, Lack of USA emissions suspect overall.
There was an o2 sensor but no wire running to it in the old rusted off premuffler. I replaced that with a cat bypass pipe and plug where sensor should go.

Now Seems to be running rich, with most of the inconsistent symptoms mentioned here. Any advice or do I just live with the low RPM Ambiguities and sweet smell of rich exhaust?

[MConn]

Bhoff - Thanks for this post !!!!! Did you ever fix the problem? What was the solution?

My 87 3.2 has the same symptoms, been fighting it for 2 years. I've had enough and am redo-ing all rubber and gaskets from the intake port on the head to the air box.

Does anyone know what vacuum should read at idle on a 87 3.2 - stock ?

[Tippy]

Mine has had the same symptoms except it doesn't ever really die out.

I have done around 40 checks and 4 parts replacements to no avail.

I am at the point of the computer or the chip. I have 55lb injectors and turbocharged, the reason for the chip.

If the computer or chip is not the problem, the nearest cliff is where mine is going......