PenguinCoder

So i just got this 89 GTU, and when i first start the car it will immediately jump to 2k until just after proper oil pressure is reached, then it drops to 1k and revs back up to 1.5k. It is pretty consistent in speed and regularity, and goes away once the car is warm. Does anyone know why the car would idle so oddly while cold?

introVert

tps, or possible vac leak. Mine does this as well, but I haven't bothered to fix it. I use it to gauge when the car is warmed up and ready to drive.

Aaron Cake

iTrader: (1)

Likely the typical vacuum leak, as well as a misadjusted TPS and possibly a dirty/sticky BAC valve.

When you start the car, it should immediately rev to around 3K for a few seconds, then drop to about 1.5, until it hits the proper 750 RPM idle when warm. Any deviation from that is generally due to hacking (though I agree with disabling the 3K startup) and the adjustments I mentioned above.

Icemark

I have seen this intermitently as well on one FC. I can tell you now, that it is not the BAC, or a vac leak, and the TPS is adjusted correctly in this case (but the TPS is old- 120K miles, so I can not rule out the TPS having issues when partially warm).

I'll should have access to the car this weekend, I'll try and see what I can find out.

PenguinCoder

I'm thinking it is probably not a vacuum leak. This only happens when it is _cold_. once it is warm, it will idle steady below 1K. There could be a leak, but in all of my experience a vacuum leak affects the engine all the time not just cold.

Could be the TPS, i will be getting the car back from GA this weekend, so i will take a look when i pick it up.

PenguinCoder

Would the fast idle have anything to do with this problem? i was looking through the Mazda Service Manual and it looks like that could need some adjustment...

introVert

Hmmm- awaiting the iceman's retort- I would like to get off my lazy butt and fix this problem...

Icemark

Through testing I have ruled out the problem being caused by the BAC, and AWS.

On cars that have this problem, do you have the cold idle assist on the throttle body???

introVert

Hmmm- s4vert owner here- not sure where it is until I go out and check-

but she also has 150,xxx miles on the odo, so my TPS is ancient as well...

sleejay

i have the SAME problem (89 gtu) ...
tps is set properly... NO vacuum leaks...

i know my BAC is bad, so i thought that was it, but yes i still have the cold start assist on the tb. (which id like to remove, i dont like shooting to 3k right at startup, cold or not)

side note: anyone want to point me to a way to get rid of CSA while keeping everything else normal? (idle, etc) thanks

SPDRACER

I also have the same problem

1990 j-spec t2 with cold start assist

I had the car at a rotary specialist a while back and he advised me that it was a faulty tps but I don't know how or why he came to that conclusion...

You dont think the problem could be due to a faulty coolant temp sensor? Considering the rpm fluctuation only occurs when cold...

Icemark

I have a new TPS for the car I have been tracing this on, but have not had a chance to put it on yet.

as for coolant sensor, I have not checked that, but the guy with the problem has advised me that surging idle is now happening warm or cold for the first 30 seconds, so I would tend to rule out the coolant sensor.

This weekend, I'll look at the TPS (even though his existing one specs perfect) and I am thinking the AFM and the cold start idle assist (not to be confused with the lame sub zero start assist that uses coolant to help start the car- which we luckly don't get in CA due to emissions).

PenguinCoder

I am pretty sure that the cold start assist is still attached to the car, so i warmed up the engine to test the TPS. I was getting in the range of .9K to 4.9K ohm resistance, so that seemed okay, but i went ahead and adjusted the low end to be at 1K... Then the car's idle worsened. Now it will dip down to 400 or 500 rpms and i have almost nothing above 6K RPMs. I even went so far as to return the adjustment to where i found it, but it appears that adjusting it in general does very little to the car either while running or at rest. I also checked the AFM and it seems to be resisting the proper amounts whilst closed.

Do i need a new TPS? or is there some more sinister afoot? reset the ecu or something?

Tournapart

iTrader: (4)

i dont really see it as a big problem or anything, it does it on my car as well (91 Vert). has done so since i bought it, 93K on the car. it doesnt really bother me, my honda used to do the same ****, but it would only do it if the engine had a full cool down, 4hrs or more of sitting, so i dont see the big problem yet. im sure something is going bad but ill replace it when i get to it

PenguinCoder

I am starting to think that this is mostly normal...

But here i am, not half an hour since the last readings and the car is idling fine, it is bouncing around no more than one notch on the tach in either direction... And i didn't do a damn thing! I just jacked the car up and changed the tranny fluid, i didn't fidget with any of the electrics, adjust the idle, or anything... now it is idling just fine... maybe the inclination of the car sealed a small leak or something... perhaps the rain lowering the ambient air temp almost 10-degrees helps it somewhat.. i dunno. Not sure i can complain either, my 6k power is back the way it was when i got the car.

Oddities...

Icemark

well I replaced the TPS on my buddies this last weekend, and it still had the same problem, however as an intresting note, the problem went away with the TPS unplugged altogether.

I have also ruled out any of the cold idle assist, and timing (the FSM and Hayes both call for timing to be checked first). I am going to try subsituting a fuel pump next as his car is doing it weither or not the car is warm- as in anytime the car is started, for the first minute or so.

The other thing I am going to look at next is the TPS reading at the ECU. To see if there is enough of a drop there to cause the problem.

Digi7ech

iTrader: (1)

I have a similar problem.

I have a S4 TII engine which will constantly idle surge no matter what temperature.
It surges between 1500 and 1900 rpm.

I replaced the TPS a month ago and it didn't do anything. I then set the TPS to 1 volt or whatever it's suppose to be. The car idle surged increase to go from 1k-2k rpm.
I then adjusted the tps up to 1.2-1.5 range I think to try and relieve the surging.

It still does it pretty bad though. I haven'tr tried driving around w/ it unplugged though.

On initial start up the car will often idle perfectly at 800 then bump to 1k then once I depress the clutch and touch the gas it will jump to 1.5k. Then after some driving it will idle surge. At street lights it sounds like I have a bridgeport.

It's been like this for 3 or more months now.

I known I have a small TID crack but it shoudln't affect it this badly.
I also checked my BAC so I know it's not the problem.

CarmonColvin

There are parts of the intake system that only get vaccum when the engine is cold.

My '91 had the idle surge when it was cold and it would go away and run perfectly once warm. I spent weekends trying to locate the cause of it. I finally went and bought a hand vaccum pump and started checking everything and found the cause. It was a vaccum leak on the "diapharm" (AKA: 13-W7Y; N350-13-W7Y) that is on the throttle body. The key is that it only sees vaccum when the engine is cold because its vaccum is controlled by the "thremo valve" (AKA: 13-980A; N304-13-980A). This is by design. The secondary throttle plates are controlled by the diapharam and is only pulled closed by vaccum when the engine is cold (to act like a choke) the rest of the time they are open.

The problem is that the diapharm on the inside is made of paper or rubber and they have cooked with the rotary heat over the last 1x number of years. They crack and leak. Try capping off the line going to the diaphram (not the diaphram itself, you can leave it open) and see if your problem goes away OR put a hand vaccum tester on the diapharm and see if it holds a vaccum.

At the time I discovered this I checked every one I had in my garage (4 at the time and all of them leaked). This leak NEVER showed itself when I was checking for leaks using brake cleaner. I think the leak was too far away from the intake track for the brake cleaner to get sucked far enough in to cause the engine to change sound.

Attached is a picture of the part I am talking about.

Icemark

Not on this car... already removed

CarmonColvin

That would eliminate that as a possible source.

Is there another part of the intake system controlled by the vaccum valves that only operates when cold or only when warm? I don't remember.

BGH

I have this exact same problem and I almost
just bought a new tps.

Icemark

Okay, it looks like I got this fixed on my buddies car, but I can not say without a doubt that what I did fixed it.

With my buddies 88 GXL, the TPS was new and set to spec with two different meters, Timing was dead on spec, and he was getting a 500 RPM surge when starting for the first 30 seconds to 1 minutes hot or cold engine (although initially the problem started appearing only when starting cold). Naturally I assumed TPS to start with, but that did not appear to solve it.

I ended up swapping out the 9's he had on the leading plugs, and newly cleaned injectors, and setting TPS by the plug at the ECU.

Bob_The_Normal

Damn.

This was hard to find even searching but glad I did search for it...

So no clue what it could've been mark?

Mine does it, start the car and it jumps up to 2.5k drops to 1-1.5k an back and forth really fast as if I was revving it...



History on this: I've had my TPS set wrong for a long time, it was adjusted about 30% rich... previous owner did this to shoot flames (sigh) and while I should have fixed it a long time ago, I've had a problem with the thermosensor being disconnected so cold starts wouldn't happen anyway... so it helped with keeping the car idling cold.

ANYWAY. So the other week I swapped out the harness and got everything back working great... then other day I rewired in my SAFC again (was wired into the old, bad harness) and readjusted the TPS to stock.

This fixed hot-idle problems that I'd been having since installing the new harness... and by hot idle problems I mean, pulse-y, torqueless low end (running far too rich), etc. But when I installed the harness (before adjusting the TPS) and it now had access to the thermosensor, I now have the cold startup (3k rev) procedure working. It did it perfect... start up, rev to 3k for a little... down to 1.5-1k.... but then wouldn't idle solidly once fully warm.

Adjusted TPS and now although I have a ROCK SOLID warm/hot idle (it's really nice hehe) and much better low end torque/power.... I have the problem you're all describing when cold... I'll take a video of it today.

So I'm guessing it's definately related to TPS, but I'm not sure why it's happening.

If you set the TPS by the plug (FSM style) Icemark, that might be the thing to do.

I don't have a checker light though... what voltage level can I look for on the checker plugs with a voltmeter that would indicate that it's set correctly??

--Gary

Rx7Ridah

iTrader: (8)

bob, your looking for 12v + at either side, (pin a or b) and grnd @ pin c. you will also want to make sure your tps is reading consistant though a sweep and 1k ohm at rest fully warm. if its not fully warm it will be slightly off due to the thermowax on the throttle bodies holding the throttle plates open. if you have a turbo model with stock intercooler i have found that making a home depot special out of some pvc piping and rubber plumbing 90's works great for diagnosing and working on stuff normally buried under there.

Bob_The_Normal

^ Thanks on the 12v on the pins, that's easy. I already checked my TPS of course... I check it with an analog MM then set it with a digital.

I'll see what setting it like this turns up, when I get a chance.

--Gary