my car has always had a jumping idle when i start it back up after running it for a little while since i've owned it (usually when i get gas and shut it off for a couple minutes.) it wasn't anything more then embarrassing when everyone looks at me like me like im doing it on purpose. None of the rotary mechanics ive dealt with over the years really know how to souce the issue. recently, it went from having a bouncing idle for a minute or two tops until it gets back to a normal idle. now, it does it while i am driving. it is a weird/scary feeling, cause i will let off the gas pedal, and the car bucks and hesitates like i am pushing it, even at highway speeds; as if the car is giving itself quick jabs on the gas pedal.

i brought it to KDR a couple weeks ago for a dyno tune. the problem seemed to have switched to this new hesitation bucking mode there on the dyno. (just to clarify, this issue was happening before i brought it to KDR, a new similar kind of issue relating to this unknown problem just showed itself on the dyno. KDR did not cause this problem.) while on the dyno, we also noticed that it was smoking coolant near behind the UIM, and dave figured it was the turbo coolant hose. we also noticed that the aftermarket upper radiator hose wasn't fully sealing at the top of the radiator (note: i have a FMIC with an upright radiator set up.) he thought that the issue with the bucking and the idling could have been related to the leaking turbo coolant hose throwing something off (forgive me for lack of the technical word- we had just made 348rwhp with my car and i was on cloud 9 and i cant remember exactly what he said.)

i ordered the turbo coolant hoses and new stock radiator hoses. sure enough, as he said, the turbo coolant hose was practically busted right open. we removed my airpump, since i have a midpipe that its not hooked up to, so it was just sitting there useless, threw on Pettits airpump delete pulley and belt set, and saw the faulty hose immediately. we replaced it, and i figured i was home free. we installed the upper radiator hose, as the lower radiator hose was in good shape and didnt need to be replaced (so i have one sitting around when i need it), filled the car with coolant, water, and water wetter and got the air pockets out of the system. i get in the car and start it up and im letting it idle and the temperature starts to rise. at first, it rose quickly and started to have its bouncing idle (1500 to 2500 rpm, back and forth repeatedly.) i shut it off, and added more mix to the radiator to top it off due to the air pockets.

i hop back in the car and start it up and let it run again. it rises slower this time. im watching the temps slowly rise on the PFC commander, and it holds at 85*C for a little while. im thinking ok, thats good. so i give it a little gas to get the temp up a little to see what happens. i jab the throttle and hold it and it starts to slowly rise to 88*C. it stays there for a little while then starts to slowly go down. i give it some more gas to make sure the problem of the ghost idle is fixed and all it required was a well pressurized coolant system and the turbo coolant hose (a lot of mistaken wishful thinking on my part- that would have been too easy.) it then raises back up to 88*C and then to 89*C. the instant (and i am talking millisecond) it hits 89*C, the car starts to have a bouncing idle.

i let it run without giving it gas to cool it down to 87*C. i give it gas again, and the temp rises. once again, the instant it hits 89*C the bouncing idle starts. so i shut it down and we pack up and decide to call it a night as we feel we've done all that we can for now with the knowledge on the issue that we have (which sad to say, isn't much- which brings me to this point: what's going on??!?)

on the way home, i test our theory that it is an electrical issue of some sort. my car is running nice and cool on the drive home so i give it some gas to see how it reacts with the higher temperature and to see if it is a temperature sensitve issue. sure enough, i get it to 89*C and the instant it changes from 88 to 89 on my commander, the car starts bucking while driving it at 40mph. it bucks until the temp gets back below 89*C. the instant it hits 88 again, it stops bucking completely.

i have recently done a lot of work on the car, so i am trying to trace the work to diagnose if anything i have done recently might have caused the change from bouncing idle to bucking while driving. i have recently added a resonated midpipe, added a grounding kit (which now that i think of it, the car wasn't doing this on the dyno before the groudning kit), took off the airpump (which has been useless for a couple years since it wasn't hooked up to the hi flow cat i had before), replaced the longer turbo coolant hose, replaced the upper radiator hose on the upright radiator, and i am running a 95* FC thermostat.


i tried to find similar instances on the forum. rough idle issues i could find, but temperature sensative bucking issues while driving i couldnt find.

thanks in advance for any help. its really hard for me to troubleshoot it while standing in front of the engine bay or sitting in the driver seat, so if you can give me anything to get me in the right direction that would be great.




CLIFFNOTES:
Car has a bouncing idle (about 1,200 to 2,400 rpm repeatedly) while parked
Car has a bucking throttle (as if i was tapping the gas and quickly letting off but the car does it on its own) response the instant my car hits
Both happen the instant my car hits 89*C on the Commander.

What is significant about 89*C that might cause this to happen?

 

Assuming no mechanical problems with the car, this problem WILL NOT be solved without a datalogit, a wideband, some street tuning, and a lot of time on your hands. So go buy a datalogit. Yes, I know you just paid someone to tune your car, but buy one anyway if you don't have it. The PFC idle control circuit is very sensitive, but it DOES do a good job if you get it set up right. Many people remove the ISC and disable PFC timing control. When you do that, you lose A/C and electrical load idle up, as well as increase the chance of the car stalling on decel. But I have gotten the PFC idle system to work right before, it's just a tedious process.

As far as your situation goes, I have encountered temperature-related surging. It is was on my friend's FD which I tuned, streetported with stock sequential twins (550/1300) and stock vacuum routing. airpump was removed, but I think the ACV was hooked up... I didn't do the vac job. Assuming there are no vacuum leaks and the ISC isn't dirty/bad, idle surging on a PFC is usually caused by

1) lean condition, especially on hot starts (which is what you are having). lean condition is usually leaner than 13-13.5:1, measured with a wideband on a car with no airpump. This is a function of IAT correction and the INJ map in the PFC, especially under the low vacuum areas usually reached during a hot start.

2) too much bypass air. This is a function of TB adjustment (primary TB bore butterfly stop screw, secondary TB bore stop screw, air adjust screw) and the idle learning procedure. Of course a tight throttle cable can throw things off too, the spec is 1-2mm of freeplay. I measure freeplay in the cable by marking the cable with a piece of masking tape and measuring the freeplay with a dial caliper.

3) O2 feedback being turned on without a functioning airpump and careful tuning.

In the case of this particular car I was working on, it would warm up and after the fast idle cam disengaged the idle would cut down to right around the target set in the PFC. Then at temperatures in the high 80s C or any time electrical load was detected (fan relay switch point in the PFC especially) the idle would freak out. The car would surge, especially on decel or hot start, and usually the idle would stick at around 1200 or so despite the target being set in the 800 range.

Solving this problem took about 12 hours of trial and error. Several NC rotary shops had taken a stab at it and didn't solve it during the limited time period they had. If it really is just a tuning/TB adjustment issue, the only way a shop could solve it would be with a blank check for however much hourly work it would take.

Anyway, the first step is to begin adjusting the TB screws. You can PM cewrx7r1 , he sells a PFC tuning document which covers TB adjustment in more detail. The first thing everybody does, besides checking the throttle cable and fast idle cam, is tighten down the screw under the lower studs of the TB elbow. It's recessed in there. You tighten that all the way and back it out 1/2 turn.

Set your idle speeds in the PFC. I prefer to set my base idle speed and electrical load idle speed to the same value, and I set the fuel cut recovery to 1500. My idle settings on that FD:

F/C AE 1500
F/C EL 1500
F/C AC 1500
Idle AE 850
Idle EL 850
Idle AC 950

The EL detector switches on any time the fans switch on, in addition the headlights and blower motor. The fan relay (and E/L indicator in the PFC) switches on at 1 C below the target temp in Settings 3 of the FC edit software. This is just speculation here, but if your tuner set the fans to come on at 90 C, the E/L detector comes on (you can see this on the commander inputs screen under the "etc" menu) and the ISC circuit will react differently. Base timing advances from -5 advance to about +6, and the car can start flipping out then too. Even if the E/L detector doesn't come on, the PFC idle control circuit is very sensitive (even to temperature) as I said, especially when there is too much bypass air.

Anyway, after resetting the PFC (unless it's possible to do the idle learn without a reset) make sure that O2 feedback is off and your idle speeds are set. Now start the car and perform the idle learn procedure where you idle for 10 minutes with no load, then 10 minutes with E/L, and 10 minutes with A/C. Check that your idle AFR's range never get leaner than 13:1 in this case. Sometimes leaner will work, but we'll set the upper acceptable limit here to 12.9:1 . Since there is no O2 sensor correction, the AFR will fluctuate with ambient conditions, and I usually aim for a range of 12.2-12.9:1 . Don't touch anything until the learn procedure is done, even if the idle is rough while the PFC is learning.

When the procedure is done, drive it around town for about 30 minutes, as the PFC will continue to learn as you drive. Monitor the AFR's, and if they get too lean you need to adjust the INJ map and/or the IAT correction map. See if any sticking or surging improves. The idle may stick high on decel, until vehicle speed is below something like 5-10 km/h. That is the ISC preventing a stall. It should slowly drop down to the target idle speed however once the car reaches a complete stop.

If the idle is still sticking or surging after all this, and you are positive there are no ISC problems or vacuum leaks, you have to adjust idle air and fuel further. If the AFR's are fine, you will have to mess with the TB again. Tighten the air adjust screw (recessed under the TB elbow) all the way and back out to 1/4 turn, at least if you had it set to 1/2 turn before. Don't close it all the way. Now loosen the primary bore stop screw (on the front of the TB) slightly to close the throttle plates. Double check that the fast idle cam is fully disengaged when the car is warm.

Now perform the idle learning procedure again (all 30 minutes of it). Assuming no other issues with the car, repeat the process over and over again, reducing bypass air until the idle settles down. If you reduce by bypass air a bit too much, you may have some stalling on decel unless you add more bypass air back or adjust the dashpot. And again, if you don't have a wideband, a datalogit (with some basic knowledge of how to use it), and 12 hours of your life, don't expect to get the idle right. If you want the most help from us here, wire in a wideband's output to the datalogit, and start posting AFR logs of you driving around.

 

also make sure your TPS and other sensors are in spec/adjusted. this seems obvious but people forget to do so...

 

thanks a lot for taking the time to respond. i do have an accurate AFR gauge. i dont have the datalogit (anymore) and sadly have run out of project money for a little while. all of this procedure has to be done through the datalogit and can't be adjusted through the commander? when you say "reset the PFC", do you mean i will lose all of my settings (including the tune) and have to retune the car again? where is the ISC (idle speed control) located to check if it is clean or dirty? is the the idle learn setup in the datalogit or is it in the commander as well? I had my A/C pulled when i had my first PFC installed. the thottle cable isn't overly tightened but has very little play. what exactly can i do without the datalogit (as i am reading your procedure as a play by play with the datalogit. correct?)

here are my current settings on the PFC (note i havent adjusted anything yet, as i have to get back to school and cant quite get into it. plus i dont have a datalogit so i feel im up against a rock and a hard place until i get some money to have a rotary specialist with a datalagit take care of it):

F/C AE 1100
FC EL 1200
FC AC 1200
Idle AE 820
Idle EL 875
Idle AC 900

which screws are you talking about backing out:




again, thanks a lot for taking the time to write that up. i really appreciate it.

 

i just took my car out to try to video tape what it does. its a fairly cool night, and no matter how much i tried i couldnt get the temps above 89*.

i drove it around for 45 minutes or so and brought it home. when i got home, i parked it. shut it down. let it sit for 2 minutes and started it back up. sure enough, the commander read 94 and it was acting up. I have a AEM UEGO AFR Gauge and it was pretty accurate to the readings we were getting from the dynojet. i videotaped the AFR gauge and some of the tach so you could see how it bounced.

here are the videos:
Idle Video 1
Idle Video 2
(im sorry for the poor quality and short videos. i had to do it on my camera phone and the sending file limit is 15 seconds. i dont know if that will help but it's worth a shot.)


what are the consequences of driving it like this, if it isnt running lean while it happens when im driving it? it tends to cool down while driving but stays that way in bad traffic (as i found coming back from KDR through NYC in 30 minutes of stop and go traffic.) while its parked, it stays that way for a couple minutes until it gets below 89* then comes back down to a even 900-1000 rpm 11.8 AFR idle and runs cool.

i really dont want to drive it this way (but would love to drive it every opportunity i get, but not at the expense of a new motor) but will need to get it somewhere to have it diagnosed and fixed.

thanks.

 

May sound ghetto but turn your fans on at a lower temp for now. Not a fix but a bandaid while you sort it out.

Could it be his temp controlled wax push rod thing ( forget the name) on the back of the throttle?

 

holy **** it's lean as hell. Classic hot start lean idle, assuming your wideband is ok. i had a UEGO, when it's ---- that means off the charts lean. If you didn't have your ISC installed the car simply would have died. look through the shop manual for the ISC stuff, I'm pretty sure it's on the back of the UIM but I am more of a 2nd gen guy (I have a PFC on my 2nd gen) and the location is different on those cars.

Here's a test. Right before you restart it, go to IGL/T, Fi screen on the commander. That's diagnostic mode; any changes you make will go away when you turn the car off. Go to INJ. crank that bitch up to like 1.10 at least. Start the car up. if it's still lean, keep increasing that number and see if it does anything. it should be between 12.5-13:1 steady.

when I say reset, I mean reset the PFC and lose all your settings. that's what the datalogit is for partially, to save and restore your map, unless there is a way to perform the idle learn procedure that I am unaware of.



the air adjust screw is under the TB elbow piece, recessed in there. feel around for it with your finger or use a mirror to see it. you need a flathead screwdriver to adjust it. be careful and don't strip it.

in your situation a lean condition when not under load isn't necessarily going to hurt anything. it will just drive like **** and embarass the hell out of you. you need to get a datalogit man, i know money's tight but you need it. you can't adjust IAT settings without it. the IAT sensor is under the UIM. on hot start it heatsoaks, and makes ECU think the air is less dense than it actually is. so the ECU leans out the mixture. on a stock car that's not a big deal because the O2 feedback takes care of it, but once you remove the airpump and cat the O2 feedback doesn't work anymore.

 

ok in that pic, I think I labelled the primary and secondary butterfly adjuster screws correctly:



don't go messing with them right now though.

 

Sounds like throttle body hard settings. If those settings aren't right or close to right, you get odd idle issues like you are experiencing. Incorrect TB hard settings also affects the accuracy of the TPS sensor. You want to get the TB hard settings right and then check/adjust the TPS.

Tuning sometimes also plays a part so trying a different map might once you have checked and adjusted the hard setting would be a worthwhile test if you can borrow a datalogit from a friend.

 

thanks again arghx. im not going to mess with it at all. i have a cruise set up in New Hampshire in a couple weeks and the final destination is a rotary shop (Synaptic3.) I am pretty sure they have a dataloggit so ill see what they can do. its definitely a little embarrassing, but i can deal with that. ill throw the silencer on to tone it down a little bit too.

i know i keep saying it, but honestly, i really appreciate the feedback. i will look into a Datalogit and get one as soon as possible. im kicking myself for selling it now.


thanks gracer7-rx7. ill have to have all of this checked out.


any more feedback is still appreciated, and ill post up what the problem was and how we fixed it when we get to it.

 

My friend took my Power FC out, cut the EGR Solenoid wire, AWS Solenoid wire, Split Air Bypass wire, and EGR Switch wire from the harness. I backed it out of the garage, (for the first time this year. Can't wait to drive it!) let it idle, idled it at 3,000, and stabbed the throttles a couple times to get it up to temperature.

She got up to 88*, I watched anxiously, held the throttle to get that extra 1*c to see how it acted at 89*; my temperature threshold when troubles start. It didn't do anything this time. I think it was just that the wires weren't clipped or cut when my 2nd Power FC was installed.

I don't want to speak too soon, as I haven't been able to drive it on the road yet, but I think it's fixed. I'm planning on putting her back on the road within a couple weeks, doing an autocross event, and see how everything is. I will do a follow up post.

Thanks to Dale Clark for the help with the Power FC FAQs to get the job done.

 

Bringing this back from the dead.

I've been having this exact issue for over a year now. I'm just getting around to attempting to fix the issue.

There's some excellent info here.

 

Well, did you cut the 4 wires?

 

I haven't yet...but I'm going to.

 

I'm not sure about the OP's car, but cutting the 4 wires listed in DaleClarks FAQ fixed the issue for me. I'm still gonna have a local tuner check the maps for me just in case...

 

The 4 wires MUST be cut if you have a USDM car. This is mentioned in the installation instructions from Apex'i.

 

In my case, the PFC was already installed and tuned when I bought the car. I had the car retuned after breaking in a fresh rebuild...then the problem started... which I don't get why it did if everything was fine prior to the rebuild