ImThatGuy

Hey everyone-
I am having some major idle hunting issues with my sr redtop.
quick info then onto to the problem:
Sr redtop, 264 cams, enthalpy tune, 2871 turbo, 255 fuel pump, adjustable fpr set to 43 psi per Enthalpy spec, front mount, 3 inch turbo back, brand new wiring specialties harness, 850 injectors, z32 maf.
I have replaced, tested(voltage and resistance) and checked: timing, tps, cas, coil packs, coil pack harness, z32 maf, coolant temp sensor, manual boost controller(removed and running straight wastegate @ 7psi) boost leak test, compression test, spark plugs ( gap and type ), intake manifold gasket, all couplers for fmic, o2 sensor, injectors.
Everything is in spec and set perfect.

Ok, here is the scenario:
Cold start the car, it starts and idles to like 1800 for about 5 seconds and falls to 900ish. Should be the iacv. Once the car is warm and to operating temp, the idle will go from 2000 to 600, car stumbles and basically dies. If I tap the throttle it will rev back up to 1800-2000 and idle down again. Once it "tapers off" it will idle from 1000-1400 and drop to 700-900. It basically does this constantly.

I can't get into timing mode. Martin from Enthalpy said because the idle is fluctuating so much, it wont go to timing mode. I've tried multiple times with no luck.

Resistance on all sensors, injectors, coils and what ever else you can check is all spot on. Voltage for tps is .45 closed and 3.5 WOT. Resistance on IACV is to spec but I have a feeling its not working so I plan to buy a brand new one.
AFR at idle ranges between 16-20 because of the idle fluctuation. Hard to get a solid reading. If the idle ever tapers to 900-1000 and stays steady, afr reads 15-16.

If I go for a drive, as soon as I give the car any throttle the afr drops to 9-10 and car stumbles.....totally different issue. I need this idle hunting issue to not be here first and then I will address the afr issue.

Any ideas? Things to check? I've pulled all my hair out trying to chase this one down.

pacotaco345

Do you have the coolant lines hooked up to your IACV? My car did this for a while and that was my problem...

__________________

ImThatGuy

I do. Everything is hooked up properly to the iacv.

Jay

ImThatGuy

Also, checked resistance of the iacv and its at 9.5. Fsm says 10 ohms so its pretty close.

Kingtal0n

procedure:

1. Boost leak test
2. compression test
3. timing verify
4. wideband a/f verify


If still have problem:
Disconnect IACV, get engine to idle without IACV activity.
If it still hunts with IACV disconnected, disconnect the TPS to disable the idle timing control feature and re-test.

SR20 is the simplest engine I have ever had the pleasure of owning. I guarantee with no boost leak, balanced cylinder compression, and proper a/f ratio at idle speed will result with idle quality.
IACV is only used to compensate for varying environmental/internal conditions. If you had a ball valve from home depot hooked up inside the car between engine intake and intercooler you could do the same thing from the drivers seat manually. Engines produced before computer controls idle just fine without IACV, the idle speed wanders more but overall no hunting/problems occur because of lack of an IACV.

TheRealSy90

In addition to Kingtal0n's post; why do you think that an idle at 900rpm indicates a bad iacv? 900 rpm is a perfectly fine idle speed.
Personnally I preferred my cam'd sr20s to idle around 800-850rpm.

Also, you never stated your bov configuration, or if you are running one at all. Martin can kind of compensate for a vta bov with his tunes. I still recommend recirculating the bov to the intake post mafs as it comes from the factory. Or have Martin tune it for vta, OR delete the bov altogether, at least until you figure out your other issue you may have.

Lastly, are you running an SAFC of any sort piggy backed with the Enthalpy ecu? Even with the settings all zero'd out, I still had weird idle and afr at times just with an SAFC2 being connected. I removed it and the Enthalpy tune was perfect. They do not need to be "fine tuned".


I agree with Kingtal0n, the SR20 was always easy for me to troubleshoot. Hell I never NEEDED to get one to go into "timing mode" to figure out what was going on, but I guess it comes with the territory of starting out with an sr20 s13. When I eventually owned a kade s13 8 years later that ran weird I couldn't troubleshoot it for the life of me, I hate ka's.

__________________

ImThatGuy

I will do my best to answer questions in order:
1. Boost leak test
2. compression test
3. timing verify
4. wideband a/f verify
Done. verified all above items are perfect with the exception of afr bouncing around do to idle hunting.
Using innovate lm-1 wide band to check afr

If still have problem:
Disconnect IACV, get engine to idle without IACV activity.
If it still hunts with IACV disconnected, disconnect the TPS to disable the idle timing control feature and re-test. What am I retesting, timing?

why do you think that an idle at 900rpm indicates a bad iacv? I don't think its bad, in fact 900-950 would be ideal, hell I'd even go 1000 rpm to get the cam lope reduced.

Bov setup is vta. Martin is away and has tuned for it. I had it recirc years ago but changed it because it was an ugly setup that seemed to not make a difference in cars running. Martin re-tuned for vta.

AFC is connected. Martin told me to dial in afr with it. I have un-hooked it about a week ago to see if it made a difference and it did not. Well, not a significant difference. AFRs actually were more lean with it disconnected. I informed him of such scenario and he said to re-connect.

What would you both consider " proper AFR" at idle? I'd say 15-16 or am I crazy?

I was only trying to achieve timing mode to see if it would smooth out the situation. Martin told me not three days ago the car will not go into timing mode with the idle jumping all over the place.


So, in short...disconnect the iacv and see if idle hunts. If so, disconnect tps and check timing? If car idles fine with out iacv connected likely it is bad then? I did verify last night that it has battery voltage with key on and if I apply 12v to it, the plunger does move.

TheRealSy90

15-16 is fine at idle, and it IS normal for the afr to fluctuate at idle and low throttle cruising. This shows that the open loop is functioning.

Hypothetically you should be able to delete the iacv and set idle air flow with the throttle stop, which some people do for "race use only", but I've never recommended this myself. I stand by the throttle plate should be fully closed at idle, and iacv regulating idle air.

__________________

ImThatGuy

Sweet.
I'd prefer to leave throttle plate closed. In fact, I just went through the process to make sure it was closed completely and tps voltage was set according. I don't like "putting a bandaid" on it to make the car run...I want it to be as intended.

TheRealSy90

That's good. I was only saying if you wanted to try taking the iacv out of the equation, that you'd still need a source of idle air. Maybe see if you can borrow a known good iacv from someone, they aren't especially hard to remove.

__________________

Kingtal0n

I am not suggesting removing the IACV lol. I am saying only to disable it to see if it was the source of the problem. If you remove something from "moving electronically" and the symptom dissipates it gives us a clue about which part is bad.

You surely want the throttle body water tight when shut. The IACV is a source for all fresh air entering due to idle on the sr20det by intention. They usually have one setting (by set screw) for idle air, and there are two plugs iirc that add additional air when the computer activates those solenoids. One is for A/C idle kick up, you can activate this one manually on a switch to raise the idle if you need to. The other thing it usually has is a coolant loop which raises and lowers idle as the hot coolant heats up some kind of wax inside the IACV. It sounds really strange but the bottom line is: if you don't connect the coolant line the idle may always sit high when fully warm, because the wax is still cold.

The diagnostic procedure I outline is to help eliminate possible causes for idle fluctuation. For example, A/F ratio, if it is swinging around wildly it means the fuel map or maf sensor situation is terrible. Normal jumps from 15.1 to 15.3 or 15.5 on a loping engine is normal. If the injector is large, and the ECU attempts closed loop, it will run like shit, because the factory injector Pulse-step size is too large for a large aftermarket injector. Therefore I disable closed loop until at least around 2000rpm when using large injectors. To eliminate this as a possible cause, disconnect the O2 sensor to make sure the ECU doesn't try to step the closed loop injector lean/rich to compensate.

If you see a/f fairly stable, and the idle is changing alot by itself, then the culprit is airflow. Airflow controls idle speed. This is why we do the boost leak test. If the throttle blade is water tight, and we disable and block the IACV, the engine should die. If it continues to run, it means air is coming from a leak somewhere (but we found that leak with the boost leak test). So you would hold the throttle open manually with the blade to simulate a steady IACV airflow stream and idle the engine proper with a nice stable a/f ratio, thus proving the engine is capable of such a thing and flushing the IACV out as the culprit. If on the other hand, the engine still has difficulty running when using air through the blade as opposed to IACV (For, to the engine, there is no difference where it comes from really) it would indicate a problem in the valve train, piston, compression, that sort of thing. But we would have found that with our compression test, most likely, so it is unlikely in this case.

ImThatGuy

ok gents. Heading to the shop right now to test iacv and tps scenario. Cold start and will get car to idle then report back.

ImThatGuy

side note-
what is acceptable pressure drop when doing a boost leak test? psi per second or what ever time frame? What do you look for in a "sealed known non leaking" setup?

ImThatGuy

I followed this to the T

Disconnect IACV, get engine to idle without IACV activity.
If it still hunts with IACV disconnected, disconnect the TPS to disable the idle timing control feature and re-test.


Started car with iacv unplugged. car ran but like crap then died. Had to unplug injectors to clear fuel from cylinders and let it die. Plugged injectors back in and iacv to get to run again then let it warm up, unplugged it and idle dropped significantly. car ran though. Unplugged TPS, checked timing and it was 4th mark from the left. I adjusted CAS to 5th mark from the left. Idle stayed steady between 650-750. Plugged TPS and iacv back in and timing went back to 4th from the left and idle increased.
Car was warm by then.

TheRealSy90

So, problem is fixed?
You did the boost leak test? Theoretically you shouldn't have any pressure drop, as you're inputting pressure through the turbo intake into what should be a sealed pressure system.

__________________

ImThatGuy

Im going to try again tomorrow. So I need to get the idle set higher because 650-750 is just to low.

Ok so when I do the boost leak test, I get pressure from the T fitting on the valve cover and the crank case return from the T fitting hose. If I plug both of those the pressure remains stable for longer but eventually tapers off to about 10 psi. Im testing at 20-22 psi. I can hear the turbo wheel spinning while I'm filling the system with air ad it makes it difficult to hear leaks. I have a bottle of soapy water to check things but we all know that only goes so far before gravity starts working against me.

TheRealSy90

Sounds like your throttle is not fully closed. Air is getting from your compressor intake and into the cylinders, out the exhaust valves and spinning the turbine. I mean it’s not going to be literally air tight, but I’ve never heard of the turbo spinning during a boost leak test. Also air out of the valve cover and crank case sounds like some pretty good blow by, either rings or valve stem seals.


Sent from my iPhone using Tapatalk

__________________

ImThatGuy

Ok, complete cold start:

Came in this morning, unplugged the iacv and started the car. Kept tapping the pedal to keep it alive. as it warmed up I was able to back off from tapping the throttle. Once it tapered off and stayed running on its own I let it get to operation temp. Once there, I checked timing and adjusted it to 5th mark from the left. Revved it up a couple times and check AFRs. They were 19-20 and the exhaust smelled horrible. Revved it up a couple more times, went back and checked timing because the idle sounded it changed a little bit. Idle was 800 +/- 50 and the timing was clear past the mark on the right. So i adjusted is back to where it needed to be, idle settled in the 800s consistently and the afrs returned to 16.5-17.5.
This is all with the iacv unplugged and the TPS plugged in.


As far as the boost leak test, I have one coupler with the air nozzle and gauge. I connect it to the turbo intake pipe and charge it. I would "assume" that is why my turbine is spinning. However, as far as the T fitting leaking, wouldn't my compression test show bad results if I have crazy blow by, rings or valve seals that are bad?

addition: plugs look perfect. They are wet, burnt, carbon'd up or anything else alarming.

TheRealSy90

I guess I didn’t feel like a compression test is a “flow” of air that would spin the turbo. More like pressurized stagnant air.


Sent from my iPhone using Tapatalk

__________________

Kingtal0n

okay couple notes

1. never pressurize the crank case because it can blow out an oil seal. The crankcase at most should have 1psi MAX of air pressure inside it. Always disconnect any PCV port that might pressurize it during a boost leak test

2. The air pressure between turbo compressor -> intake manifold should only leak out very slowly, perhaps 1psi per every couple seconds at most. You'd like to see it just sit at some pressure and stay there, leaking around 1psi per 30 seconds or more if possible. Tiny leaks are not a big deal though, just as long as it holds for a bit and you detect no large leaks. If it seems like an exhaust valve is open and air is leaking into the exhaust simply turn the engine a little bit to shut the valve.

3. Need to set timing first. It sounds like you are chasing your tail here. Stab the CAS as per FSM and get it right, be sure 100% because everything you do after that will be screwed if you don't make sure timing is correct. CAS position should be nearly center when correct (center the bolts more or less if properly timed). Use a high quality timing light and make sure you reverse the pickup (flip it both ways) because some pickups are directional and only show correct timing when used in one specific direction. Timing needs to be exact. Dont move on from this until 100% positive you got it to exactly 15* BTDC and CAS is nearly centered etc...

4. you get timing, compression, and boost leaks done; Now you can move on to diagnostics.

At this point you might as well post pics of the engine bay so we can point out any obvious problems with plumbing. Hopefully you are using OEM intake tubes etc... and it is correct. basically maf needs to be placed a good distance from the turbo and needs a couple bends in the tube, like the OEM one.

Make sure fuel pressure is correct (around 44psi with engine off / no reference);
Unplug O2 sensor narrowband to eliminate closed loop;
then I would probably try to dial in idle a/f ratio now that you have proper timing and no air leaks, to around 13.8 to 14.5:1 using either fuel pressure or SAFC. I assume you have a rom tune?

With the a/f ratio around 13.8 to 14.5 the engine should calm down a fair bit and allow you to adjust, finally, the Idle control valve air bypass setting, using the set screw adjust idle speed to around 850rpm fully warm. You do this with the TPS Unplugged to eliminate idle timing control feature (just like we disabled closed loop feature by unplugging O2 sensor) also when setting timing with a timing light you will want the TPS unplugged for the same reason, it will say this in the FSM procedure.

ImThatGuy

Excellent. So how do I pressurize and eliminate the crankcase being pressurized?
I mode a post about doing a boost leak test but replies were minimal.

I will get to the rest of the items King posted as soon as I can. Probably will not have time tonight due to stupid holiday parties I was signed up for.

Very much appreciate both of you guys helping me out on this. I have been chasing this for YEARS and gave up but now I am back at it after the car sat for five years collecting dust.

Kingtal0n

Look for hoses connecting intake air path to the crankcase

There is one fresh air tube, the crankcase receives air from behind the air filter but before the compressor. That is the tube you are pressurizing so it needs to be disconnected. Then there is the rear of the valve cover PCV valve which connects intake manifold to crank case. The PCV is a check valve and should not flow from intake -> crank case, but if it is bad it will and needs to be replaced. I always change the PCV valve on every sr20 engine the minute I see it. If this check valve is bad it will cause oil to blow from the crankcase as boost enters the crankcase and blows it out, same as using an air compressor it can blow an oil seal if pcv valve is bad.

Finally the piston ring and dip stick should be considered. Dipstick enters crankcase same as piston does, through a seal. The dip stick tube has a rubber seal and the piston ring is a metal one which we hope represents a high quality oil/compression seal. Piston has multiple rings, top ones can be damaged by running hot/lean, they can lose tension and when compression goes up during WOT, combustion pressure will pass the rings and into the crankcase. They can also become worn out from mileage.

Be aware of those additional crankcase seals for valuable diagnostic clues. For example if you feel air blowing out of the dip stick it means that you are putting air into the crankcase and the dipstick is one possible exit. I always pull the dipstick tube before beginning a boost leak test because it will tell me if I've forgotten to pull off a tube somewhere that can damage an oil seal. This also explains why the dipstick (hopefully) pops out when an engine has bad blow by during WOT. The theme is to bandaid this condition with a catch-can and keep driving the car but the real problem is often the piston ring and the engine needs a rebuild (which is why I dont put cans on my cars, to me it looks like the engine has an uncontrollable blow-by problem and needs to be rebuilt). If an engine is in good condition but still needs a can because it has higher output than factory, the real solution is to improve/revise the baffle.

I put more info about pcv system here that might be educational:
http://www.nsxprime.com/forum/showth...=1#post1909285

ImThatGuy

Will do. I will get to looking tomorrow. I am also going to order a new pcv to make 100% I can eliminate it from possible problems. Great read you sent in the link also. Tons of great info in there.
I will check the dipstick o-ring. I have never noticed it popping out, ever.

Now that there is snow and salt in Chicago, I am done road testing the car. It will be difficult to address running/boost issues but I can do idle/rev testing. I ran the car today and noticed the stumble is some what back. I still have the iacv unplugged and the tps is plugged in. I have not made any changes to the arc fuel compensation to richen up idle. However, I did check the timing again today to make sure it was good still and it was spot on. My shop is nearly 90 feet long so I backed up by the door and did a rolling launch to see if I can make the car stumble. It leaped off the ground and had tons of power but basically died when came to a stop. I also noticed low vacuum on my boost gauge so I am thinking I have created a leak when I removed my intercooler piping the other day. I will also get for this tomorrow and possible order all new couplers to also eliminate this a possible problem.

Happy holidays to you both

J

ImThatGuy

Working on this today. Got an extra day off from work so I am at the shop wrenching. Updates to follow.

I got my conzult plug wired in over new years. Plugged in to my laptop to check everything as the car is running. Timing is dead set at 15. When I go into idle adjustment tab, car revs up voluntarily and then come back down.
I am not skilled at using the conzult program but figured it would help me double check things.

ImThatGuy

Update:
I do not have the new pcv valve yet because fr sport did not have it i guess.
did another boost leak test. results look way better after I verified the throttle plate was completely closed.
I removed the stop screw, plugged in the iacv and started the car. It stumble and I feathered the throttle to keep it going but once it stabilized the idle was good and even and the wot throttle stumble seems to have cleared up a bit. I can't road test the car because of the weather.
Once i receive the new pcv, I will replace and test again.

So my next question is...why would I not have a cold start high idle? As you know when you typical "cold start" a car, the revs will be 1500 or slightly higher until the car begins to warm up and then taper down to normal. I do not have that at all. I turn the key and car starts, revs to 1300 ish and then immediately drops to idle.
Years back I had a major electrical issue that burned up my main harness. Since then I have replaced the entire harness, ecu and all the sensors we have discussed in the thread.
By cold start, I mean the car is completely cooled, been sitting over night or for 4+ hours with temps around 50-60 degrees F.

TheRealSy90

Does your iacv have the factory coolant lines plumbed to it? People tend to delete them.


Sent from my iPhone using Tapatalk

__________________

ImThatGuy

If I recall, I re-routed them due to turbo upgrade and greddy intake manifold. I can go snap a pic of them, if I can get my phone back there.

TheRealSy90

So your IACV is not adjusting for cold start because it has no source of coolant temperature to manipulate the wax diaphragm.


Sent from my iPhone using Tapatalk

__________________

NiSilS14

In my experience, I've always had to "crack" the throttle blade a tiny bit on KAs and SRs. There's kinda a sweet spot where it's not too much to where it idles past 1k, but it doesn't stumble on its face when it's coming down from revs. That is given everything else is in working order. no intake leaks? good maf?

__________________
@adamcs14

ImThatGuy

I am going to go grab a pic and post it. I also snagged a pic of the conzult screen yesterday while the car was running. I had some interesting encounters with the car and adjusted the afc to run slightly more rich at idle. I am at a 10% correction, low throttle (set points are 0-40% Throttle) to get the afr into the 16s