Zulu

'85 GSL-SE 6 port RE-EGI 13B

Hey, this has happened to me twice, once I stopped at a chevron and checked everything (made sure all the electrics were propperly connected) and started it again and the problem was gone, today, however:

I'm driving to work and about half way thru my drive I notice that if I'm only part on the throttle it sounds like it's either firing on one rotor or on only the primary or trailing plugs.... wtf? ... whatever, give it more gas and it seems normal. I come to a stop sign, then I notice the wonky idle again. By wonkey idle I mean it'll drop down to normal idle, then blip up to about 2k, then drop down to normal idle and blip up to 2k again immediately... give it some gas and it drives alright but sounds funny. I stopped, turned it off and checked all the electrics, everything looks physically connected alright. Started it again, ran at 3.2krpm for a bit, then went back into the normal wonkey idle. heh...

So, ignition coils, ignition timing, fuel pump, fuel filter, injectors, TPS - things I'm wondering about... Anybody had a similar experience? Ideas? input?

Zulu

http://mrmazda.members.atlantic.net/gslseidle.html

gump

I have had and continue to still have this same exact problem you are describing. I've gone through that "solving idle problem" sheet numerous times, removed the throttle body and cleaned it, replaced plugs/wires/cap/rotor, tinkered with the fast idle, throttle sensor, air adjust screw, everything!

It's become less frequent over time, cleaning out the throttle body and making sure the butterflies open and close freely did help the most out of everything I've done, but it contiunes to happen occassionally. I can pretty much keep it from happening as long as I don't do a lot of part throttle driving (foot lightly on the accelerator) because then it starts making that wierd noise you decrible (like running on 1 rotor) and then if I come to a stop after doing that or put it in neutral, the idle blips up and down. Then If I shut it off, and turn it back on, it goes straight to 3k rpm until I drive for a few seconds, then it comes back down. I wish I knew a definate way of solving this. It's come up on the forum once or twice before, at least for the time I've been a member. Please let me know if you solve it for good!

Zulu

Yours sounds like it happens more often, it's been about a month since it last happened. I'm going to change my fuel filter, check my timing, take off my cruise control and cleanup anything else and see how long it's gone, or if it stays gone for good.

Zulu

still happening. today it was funny during accel too, waited for it to warm up fully and then gave it some **** and everything went back to normal. wtf....

Manntis

low compression on one rotor?

*just a shot in the dark*

Zulu

while I haven't done a compression test to say 100% no, I'm still going to say no


it's happens about once a month. if I give it some **** (hit redline a few times, engine breaking, do a few hard runs), it goes back to normal and will be fine for about a month.

damon

sounds like the injectors may be worn and getting stuck open or closed.

peace

Zulu

Hrm. That sounds very probable... gah injectors aren't cheap tho :p

FJ

Injectors are not likely, as they usually just quit altogether, or cannot flow enough which would be apparent all the time. The typical problem when they get dirty is slight leakage when the car sits, causing flooding upon startup.

I don't have the answer, but a little something you may have overlooked...
After you had adjusted the TPS and cleaned around the throttle plates (whether by removing the throttle body and cleaning the assembly or just removing the intake duct and cleaning the plates and bores) did you clean and lubricate the throttle linkage? Even if they open and close properly as you watch them, it takes very little for the secondary throttle plates to stick open intermittently, and it will give you a bouncing idle. The extra air is measured by the AFM but the TPS is telling the computer something different.
I haven't run into it, but if they were open enough (or sticking partially closed when they should be open), I don't see why it wouldn't cause part throttle problems also.
A small problem, but with results similar to your idle.

"it's happens about once a month. if I give it some **** (hit redline a few times, engine breaking, do a few hard runs), it goes back to normal and will be fine for about a month."
It could be that by playing with the throttle the stuck linkage comes free, but it sure doesn't sound like timing, ignition or fuel to me. That's what makes me think of the possible "overlooked" items; a few well placed drops of oil may do the trick.

If it is not the linkage and the "running on one rotor" symptom happens consistently at the same throttle position it might be a bad TPS, but I'd expect the idle to be fine once you released the throttle.

The fact that the idle goes to 3k for both of you after restarting it, with a warm engine, is curious. I have never read anything about which sensors trigger the 3k rpm startup, but it must have to do with the thermosensor or the water temp switch. If either was intermittent, it might cause the computer to give too much fuel, thinking the engine is cold. But that will be hard to prove if it happens only once a month. It might be worth cleaning the connections on those parts, though.

When it happens, do you have any black smoke at all? If the flap in the AFM is sticking, it will cause the engine to run rich. If it is stuck open enough, it will definitely sound like it's running on one rotor, but I would expect to see smoke. Remove the air filter and clean out any residue from around the flap, then spray the flap pivot side with WD-40 to rinse out any dirt. If you pull over when it starts to act up, leaving it running, and tap on the air cleaner it should be enough to dislodge the flap, if it was sticking.

A little long and not much help, really, but a couple of thoughts. Wish I knew what it was, but they're strange symptoms.

-John.

Zulu

awesome input, next time it happens I'll go check the AFM. The throttle linkage itself all seems fine, and when I was examining the cold-start secondary throttle plates (contemplating removing them like the common 2nd gen mod, but decided against it as winters coming up ) I confirmed that everything inside is super smooth too.

Definately more stuff to think about/check out, thanks for the ideas.

gump

Wow, that really is a great idea, thanks! I'll have to try it today. I have the problem more often then Nitjsu. I drove 200 miles yesterday and it happened to me at least 5 times, but If I drive it hard, or do a long engine brake from like 4000rpm down to 1000rpm, then put the clutch pedal in, it goes away. I can sorta control if and when it happens too, if I'm crusing and i'm barely on the throttle, that seems to trigger it for me.

Lately I've been thinking it was related to the BAC assembly, because when it happens and I go under the hood I notice lots of air is going through that rubber elbow that comes off the side of the dynamic chamber and feeds into the BAC. If I pinch that tube so air cannot flow through it, the idle completely returns to normal, but as soon as I let go, the idle begins to blip again. Can you confirm that Nitsuj next time it happens? I'm debating blocking off the BAC assembly from the dynamic chamber just to see if that stops it.

One more thing. This might not help anything but....you know how we both (Nitsuj and myself) mentioned that if the car is turned off while this is occuring, and you turn it back on after a few minutes or a few seconds, the rpm shoots right up to 3000 and holds steady? Well while messing around the other day, I was able to make that happen by myself while the problem was not occuring. I did it by pulling the small vacuum hose to the lower left of the BAC assembly. As soon as I reattached it things went back to normal.

Anyways, I'll go out in a little bit and try to make it happen, shouldn't be too difficult. I'll bring some WD with me. I'm very eager to fix this one for good, I've been battling it ever since I got the car 16 months ago. State inspection is coming up soon too, and the last thing I need is for the person who inspects it have this problem happen! I'll keep you posted. See ya.

Oh yeah, one more thing, no smoke from the exhaust ever.

Zulu

If you do go the BAC route and it works, take lots of photos

gump

I played with the AFM while it was happening today. I sprayed the door mechanism with WD and tapped it a bit. All it really seemed to do was drop the idle as i pushed it open further with my hand. Didn't mess with the BAC at all, maybe tomorrow.

FJ

Sounds like a clue to me.

I thought of the BAC for the bouncing idle but didn't see how it would cause the 3k idle on it's own at the same time, without a sensor going bad. But it could well be the case.

The BAC should have no effect on idle, unless you are loading the engine. In case you are not aware, it's function is to keep the idle stable when the lights are on or other load is applied at idle. Pinching the elbow should not make a difference at idle at all, as long as the accessories are off.

So if it's the BAC, the question becomes whether it's the valve or the signal going to it, as a result of a faulty sensor or a bad ECU. (I don't think bad ECUs are common as far as it relates to a BAC valve, although that's not uncommon on the 2nd Gen; apparently, just a transistor that burns out in the ECU. Easy to replace.)
It is not a DC voltage but a PWM signal controlling the Vent and Vacuum solenoids, which push and pull on the diaphragm in the BAC. A series of pulses that vary the "on" time of the valve, in the same manner as the injectors. But you can still measure the DC equivalent: I just ran mine to check the voltages and when warm had 8.4V on the Black wire and 11.7V on the Red wire (battery voltage is 13.6), at the Vent and Vacuum solenoid connector. (If you are not familiar, the Vent and Vacuum solenoids are the two round solenoids just below the BAC, and have a common connector; 3 pins.)

Or, if you can catch it when it's acting up, disconnect the BAC connector and see if it changes the idle. Not as accurate, but any info is good.

If your voltages appear similar to mine, (if you can catch it while it's acting up), I would take the BAC apart. I have not opened mine, but it's not a big deal at all. Just remove and clean the innards. It could just be carbon buildup preventing the diaphragm from closing properly on occasion.

Also, under the dynamic chamber (the box on top of the engine, sorry I don't know if you know the terminology) is a black, circular "chamber" (about 1.5" by 3", ) that is in series with the BAC. If that part or the vacuum lines to and from it are leaking, that will obviously cause problems with the BAC. If the vacuum lines are original, that would be a possible source of the problem.


"you know how we both (Nitsuj and myself) mentioned that if the car is turned off while this is occurring, and you turn it back on after a few minutes or a few seconds, the rpm shoots right up to 3000 and holds steady? Well while messing around the other day, I was able to make that happen by myself while the problem was not occurring. I did it by pulling the small vacuum hose to the lower left of the BAC assembly. As soon as I reattached it things went back to normal."

The hose you disconnected is manifold vacuum going to the pressure regulator solenoid. It doesn't have anything to do with the BAC, but disconnecting any vacuum line will cause anything from a high idle, to a bouncing idle, to a very rough idle and stalling. Depending upon the severity of the leak. I disconnected my hose earlier to check the result, at the same location, and the idle surged. But that's not your problem.

"The throttle linkage itself all seems fine,..."
That was my original point, actually. It seems fine when you operate the throttle by hand, but it can and does stick when you're not looking. A little oil solves that potential irritant. I don't mean to suggest that's the cause of your problem, but it's a possibility if the linkage is anything but well lubricated.

"cleaning out the throttle body and making sure the butterflies open and close freely did help the most out of everything I've done, but it continues to happen occasionally"
I do hate to keep bringing it up, but did you lubricate the throttle linkage sufficiently?

I don't claim to have the solution, just a few thoughts.

And I wanted to keep this short and to the point...so much for that theory...

-John.

FJ

Yes, that's what should happen if it's working properly...It's not your AFM it appears, but it was worth considering...

-John.

FJ

gump, before the forum lost a few posts, I read your last one before it disappeared. If I recall correctly, there were two points:

Yes, I meant disconnect the V&V solenoid connector as they control the BAC. You would also want to disconnect the ASV for the same reason. If it then idled normally, as it does when you pinch the elbow, the best bet I think would be to check all of the signals at the ECU. Tough to do when the problem is not always present, but I don't see how else you are going to be able to narrow it down.

I wasn't very clear about the chamber. It's black plastic and is in the vacuum line to the BAC. You can see it under the Dynamic Chamber. It may have a crack along it's seam that opens with heat on occasion, or the vacuum lines attached to it may be leaking.

-John.

gump

Thanks, yeah I was not going to retype that whole message I left after the board crashed and I lost it

I'm doing some work today on the car, I'll let ya'll know what I find later.

gump

Ok, I have a temporary solution to fix the wacky idle problem. I tinkered for awhile unsuccessfully then as a last ditch, I plugged the elbow that allows air to run from the dynamic chamber to the BAC. Drove around a whole bunch and could not make it happen (usually I can make it act up within a few minutes if i try)
So that solves that for now. Don't know where my problem is though.

I checked my voltages, they were much lower then yours. 5.65v and 10.22volts. So maybe its the computer, or one of the solenoids?!? I might hit a Junk yard and see if they have an SE I can pull a few things off of to try (BAC, V&V solenoids, ECU)

I'll take a picture of how I plugged the elbow and post it.

I also took off the BAC/ASV assembly again and pulled the V&V solenoids out and looked at that plastic piece underneath the dynamic chamber. It looked like it was in one piece. There are many short vacuum lines down there and they all slip on and off very easily, so I'd like to replace all of them.

Zulu

At least now you're making some progress. Good luck, and I'll be following in your footsteps shortly

gump

Ok, here's the picture. It was a real simple job. Just disconnected the hose at one end, plugged it. Attached another hose I had laying around to the other end and plugged that. Temporarily fixed.

FJ

But not really far enough off to be sure. They may well be O.K. Different engines, different conditions, plus the fact that we are not measuring DC voltages. And your values relative to each other are not that different from mine. Out of curiosity, we'll have to note how close yours are to mine once it's running properly.

Two things:
-I also would have checked the voltages when it was idling normally. If they are the similar to when it acts up, it would point to the BAC itself.

-Did you try unplugging the V&V connector while it was idling badly? If not, might be worth the time to try it. If the idle returns to normal it would point to a bad signal from the ECU.

Also, you could run a piece of hose from the V&V solenoid to the BAC, bypassing the two vacuum lines and the plastic chamber. If nothing changes, that would eliminate them as the cause. I haven't opened a chamber but I suspect it's just a hollow tank to dampen the pulses in the vacuum line. It shouldn't change much by bypassing it. I'll try it on mine soon.

-John.

Zulu

Long-term impressions?

Zulu

Well I went and tried that, no more crazy behaviour. Cold starts are a bit of a pisser tho. And the car seems to warm up quicker.

*shrug*

Any negative effects from blocking it off?

gump

Well I'm not sure but it's possible my blocking that off has caused my cold start up to be a bit different. It runs great otherwise! That idle problem never ever happened, over 400 miles now. I'm just debating now on what steps to take first to get everything to work right with the BAC tube attached.