High idle problems

[Beezer]

Hi there folks new member here and kinda new NSX owner.

I've been having some idling issues with my car (92). The issues are a higher than normal cold start up and when the engine is near its operating temp the revs bounce up and down from 1500 rpms to 2000 rpms about 5 or six times then it'll settle down to about 1500. Also, when driving up to a stop my idle will hang at around 1500 to 2000 rpms and stay there for about 15 to 20 secs before dropping down to about 800-1000. Engine is stock except for an exhaust and it just passed CA smog recently.

I've read countless threads on here about the same kinds of situations but it seems like everything thats been mentioned to do has not worked for me.

What I've done so far is cleaned TB, EACV, tightened up FITV, added coolant, bled coolant, 2 new coolant temp sensors (ECU & temp gauge), main relay, ignition switch, all new plugs and coils, new battery, new negative terminal, new idler screw (have screw quarter turned opened from fully closed position), looked for vacuum leaks via carb cleaner.

Im tempted to get a new FITV and a new EACV but ive read about a couple of folks in here that done that and still had the issues.

I read a few posts about the clutch switch and was wondering if that could be the issue also because my switch has been bypassed by the previous owner.

I know i should just take drop it off to a professional to diagnose but just thought I'd ask here first to see if there would be any other ideas that i could get.

Thanks

 

[Honcho]

Hi there folks new member here and kinda new NSX owner.

I've been having some idling issues with my car (92). The issues are a higher than normal cold start up and when the engine is near its operating temp the revs bounce up and down from 1500 rpms to 2000 rpms about 5 or six times then it'll settle down to about 1500. Also, when driving up to a stop my idle will hang at around 1500 to 2000 rpms and stay there for about 15 to 20 secs before dropping down to about 800-1000. Engine is stock except for an exhaust and it just passed CA smog recently.

I've read countless threads on here about the same kinds of situations but it seems like everything thats been mentioned to do has not worked for me.

What I've done so far is cleaned TB, EACV, tightened up FITV, added coolant, bled coolant, 2 new coolant temp sensors (ECU & temp gauge), main relay, ignition switch, all new plugs and coils, new battery, new negative terminal, new idler screw (have screw quarter turned opened from fully closed position), looked for vacuum leaks via carb cleaner.

Im tempted to get a new FITV and a new EACV but ive read about a couple of folks in here that done that and still had the issues.

I read a few posts about the clutch switch and was wondering if that could be the issue also because my switch has been bypassed by the previous owner.

I know i should just take drop it off to a professional to diagnose but just thought I'd ask here first to see if there would be any other ideas that i could get.

Thanks

Welcome to Prime!

Wow, you've looked at just about everything that we would have recommended, but here's my thoughts:

A too-high cold idle and the hanging at 1,500 to 2,000 before dropping is generally the symptom of a bad FITV. Either the plunger is not sealing properly or it is just stuck open. The hanging behavior is because it's stuck open and the EACV eventually kicks in to drop the idle down. You can get a new FITV (don't forget the two O-rings) from Amayama or MITA for about 80 bucks. Easy replacement job.

Also, I noticed you replaced your idle screw. Did you set it using the service manual method and with a digital tachometer on the tach pickup in the engine bay? A mis-adjusted idle screw can also cause the behavior you are describing. Basically, it's letting too much air in and the EACV is stepping in to correct it.

 

[Old Guy]

Welcome to the club.

My first recommendation is that you download a copy of the 1991 service manual. Links to the manual can be found on Prime if you search. The idle control system description starts on page 11-66. In line with Honcho's comments, the service manual says that when the engine cold idle speed is out of spec, the fast idle valve is the first pick and the EACV the second pick. If the hot idle is out of spec, then the EACV is the first pick and the FIV the second pick.

Have you confirmed that the FIV is actually operating? The service manual describes a rather simple on car test that you can do to confirm that the valve is open when cold and closed when hot. If the FIV is not satisfying the fully open / fully closed requirements, then I think that replacement is a requirement. The valve looks like it can be partially disassembled; but, there are no internal replacement parts available. Full disclosure - I have a 2000 so I don't have a FIV or a EACV. On early generation MAP controlled EFI systems that have cold idle control via something like the FIV, it is not uncommon that when the FIV sticks owners will attempt to compensate for the problem by adjusting the base idle speed (the idle screw) which just screws everything up. So, confirm that the FIV is actually opening and closing at the proper temperatures and if it isn't, replace it. Then carry out the base idle speed adjustment as described on page 11-84 of the manual.

The base idle speed is set with the EACV electrically unplugged. This should cause the EACV to close. If the EACV is not closing then you will not be able to set the base idle speed correctly so remove the EACV and make sure that it is closed when disconnected. The base idle speed adjustment includes a test with the EACV disconnected and connected. This should allow you to diagnose whether the EACV is operating correctly. If the EACV does not appear to operate, check to confirm that the wiring to the EACV is OK (as per factory service manual) and that the EACV coil has not gone open circuit (should cause a CEL). However, an EACV that has an intermittent sticking problem may elude that test. I seem to recall that the EACV is a 12V PWM controlled device. If you are C code / one board controller savvy you could build a simple PWM controller which would allow you to run the EACV over its range of operation to sort of confirm that it works smoothly as opposed to doing a stick - lurch - stick - lurch operation (even though it has been cleaned).

The ECU does get a signal from both the manual transmission (neutral switch) and the clutch that puts the ECU into idle control mode when the clutch is pressed in or the transmission is shifted into neutral. The clutch switch that has been by-passed is likely the upper clutch switch which is part of the starter interlock system and has nothing to do with idle control. If the car idles erratically when in neutral its likely that the problem is not the clutch / neutral switch input unless both of them are buggered. Your symptoms don't really seem to fit with a clutch / transmission switch problem. When the clutch / neutral switch is a problem, it is common that the engine dies or idles too slowly because the ECU is not getting the signal that it needs to go into idle control mode.

If you have checked and ruled out vacuum leaks downstream of the throttle body then you have eliminated that as a potential problem. Check to make sure that your throttle cable is not sticking and preventing the throttle plate from closing. If you have vacuum leaks or an open throttle plate you can adjust idle to the cows come home and you will still have problems.

So, the Coles Notes version
- confirm no vacuum leaks and throttle plate is closed at idle
- check and confirm that the FIV operates as it should. If it doesn't, replace it.
- with an operational FIV, confirm that the EACV is actually closed when unplugged and do the base idle speed setting as described in the service manual
- the base idle speed setting procedure should tell you whether the EACV is operating.
- if the EACV appears questionable you can check the EACV operation as I describe or replace it

You can send it to the dealership; but, if you have the time and inclination it is more satisfying to diagnose and repair it yourself. The NSX is now a vintage / classic car which is like a hobby. Sending the car to the shop for repairs that can be done with common tools is like having somebody else build your model airplane kits for you.

 

[Beezer]

Thanks for the tips Old Guy and Honcho! I'll take a look at the service manual and check the FIV and EACV the proper way. For the FIV I just opened up the cover and tightened it down with a flat as much as i could and closed it back up. The EACV I took it off and sprayed it down best i could with carb cleaner.

As per the base idle procedure i didn't do it per the manual. The idle screw is basically a quarter turn from fully closed. That was how it was when i got it so i just marked it, took out the old one, replaced it with a new one and put it in the same position.

When i got the car it already had the idle issues and the previous owner said it was most likely due to a bad coil and an easy fix. In hindsight, I dont even think it really bothered him.

Will update with whatever i find. Thanks again.

 

[Honcho]

Thanks for the tips Old Guy and Honcho! I'll take a look at the service manual and check the FIV and EACV the proper way. For the FIV I just opened up the cover and tightened it down with a flat as much as i could and closed it back up. The EACV I took it off and sprayed it down best i could with carb cleaner.

As per the base idle procedure i didn't do it per the manual. The idle screw is basically a quarter turn from fully closed. That was how it was when i got it so i just marked it, took out the old one, replaced it with a new one and put it in the same position.

When i got the car it already had the idle issues and the previous owner said it was most likely due to a bad coil and an easy fix. In hindsight, I dont even think it really bothered him.

Will update with whatever i find. Thanks again.

Ah, ok, this makes more sense. First, you should set the warm idle with the idle screw. The process is on Page 11-84 of the service manual. You'll need a tachometer- I got a digital one from Amazon. Once you have a good base idle, you can see if the FITV is bad. If the car still does the same behavior, it's almost certainly the FITV.

Longer term, you may want to consider removing and deep cleaning the throttle body and intake manifold. As our NSX gets older, these components must be serviced to restore normal function. The design of the PCV and EGR systems means they get dirty and clogged up over a long time period.

 

[Beezer]

Honcho, what kind of tachometer did you get? The only one i found locally (harbour freight) was a multi metering type that had rpm pickup. In the service manual it says to hook up by the loop on the ignitor or thru the connector by the engine bay fan location but the tach meter i got just has a clamp that goes on the spark plug wires.

The service manual says to hook up the the loop of the ignitor secondary. What does that mean?

 

[Beezer]

I forgot to mention i tested the FITV last night per the service manual and it seemed to be working fine. I got suction during cold start and no suction when it got to operating temps. I also set the idle per the manual but i didnt have a tach so i just used the tach in the dash.

After the idle set i let it cool off overnight and started it this morning and im still getting some rpm bounce during a cold start but it did get better. Im not getting the 5 or 6 bounces anymore but i got 2 and a couple single bounces here and there until it fully warms up then the idle drops to about 800 to 1000.
I drove it last night and the idle is still hangs when i come to a stop.

 

[Honcho]

I forgot to mention i tested the FITV last night per the service manual and it seemed to be working fine. I got suction during cold start and no suction when it got to operating temps. I also set the idle per the manual but i didnt have a tach so i just used the tach in the dash.

After the idle set i let it cool off overnight and started it this morning and im still getting some rpm bounce during a cold start but it did get better. Im not getting the 5 or 6 bounces anymore but i got 2 and a couple single bounces here and there until it fully warms up then the idle drops to about 800 to 1000.
I drove it last night and the idle is still hangs when i come to a stop.

Here's what I use:

https://www.amazon.com/gp/product/B002Z32OHW/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

The idle delay coming off the throttle is normal for the 91-94 cars. They use a vacuum dashpot to slow the return to idle for emissions purposes. But, the idle shouldn't be hanging at 1,500 by the time you come to a stop. I suppose it could be a bad dashpot or also a "lazy" or sticky EACV solenoid. I mean, it is 30 years old...

 

[Beezer]

Thanks Honcho you're right it is 30 years old. I'll most likey replace both the FITV and the EACV.
This is the first time im hearing about a dash pot. Where would it be located and would it be something that's replaceable also?

 

[Honcho]

Thanks Honcho you're right it is 30 years old. I'll most likey replace both the FITV and the EACV.
This is the first time im hearing about a dash pot. Where would it be located and would it be something that's replaceable also?

The system is replaceable. The part number is 16111-PR7-A00 and you can still get them from Amayama. The way it works is it mechanically holds the throttle plate open and then slowly closes it. The suction of the intake overcomes the check valve, then when the throttle closes, the check valve locks in the pressure, holding open the dash pot therefore lowering the throttle body slowly. It's a pure emissions system and frankly it shouldn't affect the idle, but if it is malfunctioning it could in theory be holding the throttle open longer than it's supposed to. It's physically connected to the throttle plate.

 

[Beezer]

Thanks Honcho! Ive got parts arriving in the next few days hopefully they clear up my issues and if not ill try the dash pot.

 

[Beezer]

Fixed!!!

Ok so heres the update:

New fast idle valve and EACV delivered today and installed. Dont know which one out of the 2 was the issue but i reset the idle and everything is running smooth now. No more hanging idle at stops and no more hunting idle during cold starts. Thanks y'all!!

 

[Big McLargeHuge]

Here's what I use:

https://www.amazon.com/gp/product/B002Z32OHW/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

The idle delay coming off the throttle is normal for the 91-94 cars. They use a vacuum dashpot to slow the return to idle for emissions purposes. But, the idle shouldn't be hanging at 1,500 by the time you come to a stop. I suppose it could be a bad dashpot or also a "lazy" or sticky EACV solenoid. I mean, it is 30 years old...

Heya Honcho,

I've been preparing to do an FITV/EACV/what have you cleanout and idle screw adjustment to fix an issue where my NSX will rarely drop RPMs too much and stall while sitting at idle and heat soaked.

I bought the same 328 EZ digital tachometer you linked here but I can't get an RPM reading using the sensor clamp on either the ignition coil wire loop or the single wire tach connector next to the A/T engine fan as mentioned in the service manual.

I can however get a reading using the wireless mode on the tach and holding it next to an ignition coil, but my RPM reading is nonsense unless I set the tach to 4-stroke/1-cylinder mode. Did you have the same issue or am I just missing something?

Thanks-

 

[Honcho]

Heya Honcho,

I've been preparing to do an FITV/EACV/what have you cleanout and idle screw adjustment to fix an issue where my NSX will rarely drop RPMs too much and stall while sitting at idle and heat soaked.

I bought the same 328 EZ digital tachometer you linked here but I can't get an RPM reading using the sensor clamp on either the ignition coil wire loop or the single wire tach connector next to the A/T engine fan as mentioned in the service manual.

I can however get a reading using the wireless mode on the tach and holding it next to an ignition coil, but my RPM reading is nonsense unless I set the tach to 4-stroke/1-cylinder mode. Did you have the same issue or am I just missing something?

Thanks-

Hey Tyler-

I had the same issue with my tach so ended up using the Demon II to tune the idle rpm, since it is getting a direct read off the ECU. It works great on the snowblower though!

 

[Old Guy]

I won't weigh in on your FIV / EACV question; but, can advise on your tach issue. The 328 EZ tach has a clamp sensor; but, looking at the pictures in the Amazon listing the clamp sensor looks like it might be an electric field sensor rather than an inductive sensor which is measuring current flow. Inductive current sensors tend to look like the one on this tach

CT8027 Automotive Multimeter

The CT8027 is our best portable solution for troubleshooting electrical & electronic circuits and systems, and has a number of engine-specific features.

gtc.ca

which actually has a ferrite core and coil of wire in the sensor to sense the primary coil current.

If I am correct the clamp sensor on your tach needs to be on the secondary side of the coil where the voltages are much higher which on the NSX (and any COP equipped car) is a hassle because you have to extract the coil and add a high voltage extension between the coil and the plug so you can access the signal. The timing loop and tach connector test point are all low voltage. I have a timing light / tach with a true inductive pick up and it works just fine on the timing loop.

With coil on plug cars an individual coil is firing once every 2 rotations (just like a single cylinder lawnmower). That is why you have to select the four stroke / one cylinder mode to get a sensible reading. If you were using the inductive loop on the NSX I expect you would have the same requirement because I expect that the loop is just the ground connection for the #1 coil. So, your wireless function on the tach would appear to be working correctly.

 

[Big McLargeHuge]

Dangit! I did just get a gas lawnmower I could use it on but no need for a snowblower in GA for the most part .

Thanks for the explanation Old Guy, that makes sense to me. It seems like my measured RPM is about 100-200 RPM lower than the AP2 S2000 cluster I have installed which is probably to be expected but was confusing me a bit before. It was around 780-810 RPM a bit before the engine fan kicked on when I was messing around with it last night, but with the wireless mode on 4cyl/1str setting I should still be able to properly set the idle after I clean the FIV/EACV and replace the idle screw.

I'm also still running the A/T cams and ECU profile when I have a manual transmission so that might be somewhat related. The idle dropout/stall is pretty rare and the car always starts right back up so I've been putting off trying to fix it until now when I plan to attend the Mideast Meet in May.

 

[nis350]

Fixed!!!

Ok so heres the update:

New fast idle valve and EACV delivered today and installed. Dont know which one out of the 2 was the issue but i reset the idle and everything is running smooth now. No more hanging idle at stops and no more hunting idle during cold starts. Thanks y'all!!

Just curious if it is required to reset the idle after replacing these valves?

 

[Honcho]

Just curious if it is required to reset the idle after replacing these valves?

It's generally a good idea.

 

[Old Guy]

Dangit! I did just get a gas lawnmower I could use it on but no need for a snowblower in GA for the most part .

Thanks for the explanation Old Guy, that makes sense to me. It seems like my measured RPM is about 100-200 RPM lower than the AP2 S2000 cluster I have installed which is probably to be expected but was confusing me a bit before. It was around 780-810 RPM a bit before the engine fan kicked on when I was messing around with it last night, but with the wireless mode on 4cyl/1str setting I should still be able to properly set the idle after I clean the FIV/EACV and replace the idle screw.

I had a Honda gas lawnmower for 25+years and I have to admit I never had the urge or need to check the idle .

If you remain uncertain as to the RPM, you could purchase one of these hand held oscilloscopes or something like it. AliExpress / Amazon / Temu have a large selection of different options.

Professionelles digitales Oszilloskop, 2,8 Zoll IPS, Vollansicht, Farbbildschirm, Hochgeschwindigkeits-Oszilloskop, AC/DC-Spannungsprüfer, Messwerkzeug - AliExpress 1420

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

The tach out test connector on the side of the engine bay is a voltage signal which comes out of the ECU, not the ignition system. I expect the speed signal will be 0-5 volt pulses. It is the same engine speed signal as used by the cruise control, TCS, CCU and drives the tach display in the instrument cluster. @Heineken could probably advise as to whether it is one pulse every two engine rotations or 6 pulses every 2 rotations and the voltage level because he would have dealt with this when he did his cluster tester tool design. With the scope, you can measure the voltage pulses at the tach test point or more specifically the time between pulses. For one pulse every two rotations the engine RPM is just 2x60/T where T is the time between pulses in seconds (0.1 seconds between pulses = 1200 RPM).

I am a retired E Eng and for E. Engs oscilloscopes are sort of like a set of socket wrenches. That particular AliExpress oscilloscope is probably more like a set of sockets from the Harbour Freight bargain basket than a set of Snap-on sockets; but, probably good enough for this application. I am sure that @drew can probably point you towards the best of the cheapest all of which would be cheaper (pre tariffs) than that tach with inductive timing pick up that I linked or the 328 EZ tach.

 

[Heineken]

I'm on holidays so I can't check the formula

 

[Honcho]

@Old Guy I am a lawyer, not a E.E. , and I have the below. Could I get this to work as a tachometer?

https://www.amazon.com/dp/B07SL8KQPZ?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_1

 

[Heineken]

That one should work as well.
PS: If I remember correctly, the ignition signal is a 12 V rectangle.

 

[Honcho]

Errrr...maybe I should have said how do you operate an oscilloscope to get it to read as a tachometer?

 

[Old Guy]

Winging it here since I don't have the thing in front of me. Based upon Heineken's recollection that the RPM signal is a 12 volt pulse, do the following:

1- Set CPL to DC coupling
2 - Connect the scope + test lead to the tach out point and the - test lead to chassis ground.
3 - For vertical sensitivity set the SEN1 switch to 1volt and the SEN2 switch to 5x which should give you a sensitivity of 5 volts per division on the display.
3 - The Amazon webpage says the horizontal time base is selectable just like the vertical sensitivity; but, when I look at the pictures I see no switches for switching the time base. Maybe that is what those 4 push button switches on the side are for. Going back and checking the SM, it says the tach calibration is 300 pulses per minute = 100 RPM so 3000 pulses per minute at 1000 RPM or 50 pulses per second at 1000 RPM. 50 pulses per second is a pulse period (time between pulses) of 0.02 seconds. My post #19 assumptions were clearly incorrect!. Assuming you want to check engine speeds in the 700 RPM range the pulse period would be 0.0285 seconds so I would set the time base to 0.01 seconds per division or 0.005 seconds depending on what is available. It looks like the scope actually calculates and displays the period in milli seconds for you (they label it Cycl). The formula to convert pulse period T in milli seconds to RPM is 20000 / T

 

[Big McLargeHuge]

I'm a ChemE and know some of those words

Good news is that I took off the IACV & FITV and cleaned them out, and replaced the idle screw after cleaning the port out as well and seemed to have the idle dialed in well with some drivability testing yesterday.

The 328 EZ worked fine measuring the ign coil wirelessly and let me dial in the idle screw. The service manual wasn't too clear on this but it seemed like the RPM range targets are for when the rad fan is running; when the fan was off the RPM was in the normal 750+/- range but with the fan on it dropped to the 550+/- range with the IACV disconnected per the idle adjustment procedure.