When I first start the car, it almost inevitably and immediately stalls right out after the 1st start. 2nd start usually holds, especially if I rev it. Reving after first start usually doesn't work. Sometimes it takes 3 tries. This has been happening for months, at least. Any ideas about what may be going on? 1991 5MT, no modifications.
-
Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.
Starting Oddity
- Thread starter Spottwd
- Start date
The typical cause of this is clogged throttle body idle passage. What is going on is your FITV is wide open when the car is cold so it gets enough air to start and idle. Once the car warms up, the FITV closes and now the car has to rely on the EACV passage and idle bleed air. If those passages are blocked, it won't start- not enough air.
Honcho probably has the diagnosis. Remove the EACV and clean it and clean out the internal passages in the throttle body. Unfortunately, to do a good job on cleaning the throttle body you really need to remove it. You can try something like CRC or 3M throttle cleaner with the straw jammed into the passages in the throttle body to try and flush them out; but, if that doesn't work it probably has to come off for a soak.
Thanks. This one may be beyond me.
It's really not that bad. You need to pull the throttle body off (4 nuts) and clean the idle air passages.
Then, pull the EACV (2 bolts) and clean or replace the screen.
Last edited:
Maybe not!
The straw jammed into the bleed holes requires a minimalist effort. All you need to do is remove the rubber bellows to access the air inlet side of the throttle body. A small extendable inspection mirror can help you see the bleed holes inside the throttle body.
Removal of the throttle body for cleaning is more involved; but, not incredibly technical. 11-116 in the 1991 service manual covers removal. The most technical part might be the correct adjustment of the throttle cable following re assembly.
Larry Bastanza had prepared a write-up describing the process for throttle body cleaning. I have the later drive by wire TB so I don't have that document. Perhaps some other member can provide a copy. I think @Honcho did a post or responded to a post discussing the electronic air control valve (EACV) which controls idle. Perhaps have a search for that and other posts with the search terms 'cleaning throttle body' or 'cleaning EACV'.
As an observation, at Acura dealerships, the number of first gen NSX qualified mechanics is getting small. If you are driving a 1991 you are effectively driving a vintage car which pushes you into the do it yourself category unless you have access to a somebody who specializes in the NSX. Where I live, there used to be a few of the old flat 12 Ferraris (Testarossa, 512s). For the guys that could not do their own work, maintenance involved a 1500 km roundtrip truck ride to the closest Ferrari specialist.
Edit - I see Honcho responded while I was tying
The straw jammed into the bleed holes requires a minimalist effort. All you need to do is remove the rubber bellows to access the air inlet side of the throttle body. A small extendable inspection mirror can help you see the bleed holes inside the throttle body.
Removal of the throttle body for cleaning is more involved; but, not incredibly technical. 11-116 in the 1991 service manual covers removal. The most technical part might be the correct adjustment of the throttle cable following re assembly.
Larry Bastanza had prepared a write-up describing the process for throttle body cleaning. I have the later drive by wire TB so I don't have that document. Perhaps some other member can provide a copy. I think @Honcho did a post or responded to a post discussing the electronic air control valve (EACV) which controls idle. Perhaps have a search for that and other posts with the search terms 'cleaning throttle body' or 'cleaning EACV'.
As an observation, at Acura dealerships, the number of first gen NSX qualified mechanics is getting small. If you are driving a 1991 you are effectively driving a vintage car which pushes you into the do it yourself category unless you have access to a somebody who specializes in the NSX. Where I live, there used to be a few of the old flat 12 Ferraris (Testarossa, 512s). For the guys that could not do their own work, maintenance involved a 1500 km roundtrip truck ride to the closest Ferrari specialist.
Edit - I see Honcho responded while I was tying
Thanks for the encouragement. I'll pursue giving it a go myself.
I do know what you mean about the number of NSX qualified mechanics getting small. The guy whom I've worked with, he is the one who did the original prep of the car in 1991, is grumpily retiring.
I do know what you mean about the number of NSX qualified mechanics getting small. The guy whom I've worked with, he is the one who did the original prep of the car in 1991, is grumpily retiring.
Well, that didn't work out so good. I took it apart the best I could, Sprayed everything the best I could. Put it back together the best I could - with no spare parts left over.
It now seriously surges on idle between say 500 rpms and 1000 rpms.
So, any idea what my "best I could" surely wasn't good enough?
It now seriously surges on idle between say 500 rpms and 1000 rpms.
So, any idea what my "best I could" surely wasn't good enough?
That sure sounds like an idle screw/passage issue. The surging means you opened up some airflow, but not enough. Did you remove the idle screw (count number of turns) and blow out the small passage between the screw hole and the throttle body passage?
Picture please, if you can.
At least I have plenty of cleaner left.
At least I have plenty of cleaner left.
Oh, I did not remove the idle adjusting screw. Manual Page ii-84. Can I just back it out and blow it out without removing the throttle body again?
Thanks Honcho.
I've got my idle back, but looks like I still have the stalling problem. If it is the EACV screen problem, this may be beyond my hands' abilities to get to. I was able to do the throttle body spraying with the assistance of Q-tips, but haven't been able to figure out how to detach the throttle cable, so getting to the EACV has become "problematic".
Any tricks?
I've got my idle back, but looks like I still have the stalling problem. If it is the EACV screen problem, this may be beyond my hands' abilities to get to. I was able to do the throttle body spraying with the assistance of Q-tips, but haven't been able to figure out how to detach the throttle cable, so getting to the EACV has become "problematic".
Any tricks?
First, on the screw, it's really hard to describe so I'll add this picture to try and show it. The idle screw passage itself should be blown out and cleaned. I would also replace the screw and/or the o-ring at a minimum. However, this passage gets its airflow from a tiny hole on the side of the screw port that connects to the inside of the throttle body passage. It's really hard to see the port because the throttle blade kind of covers it up, but it's there. The only way I found to clean it is to stick the straw of the carb cleaner bottle into the hole and start blasting (wear eye protection!) until the fluid comes out clean. You can't really get a tool in there and it's too small for a Q-tip.
As for your EACV issue, it's 90% sure your screen- when the solenoid opens there just isn't enough airflow to keep the car running. The EACV can be removed from the intake manifold with the engine in the car. You'll have to remove at least the rear strut bar, potentially some of the wire harnesses on top of the IM and the little coolant hoses, but you can do it. It's just two bolts.
Thanks for the encouragement. I'll give it a shot.
As for only two bolts, I keep remembering that Frankenstein's monster had only one bolt.
Oh, you can get a Q-tip in if you peel off most of the cotton.
As for only two bolts, I keep remembering that Frankenstein's monster had only one bolt.
Oh, you can get a Q-tip in if you peel off most of the cotton.
Guys, thanks for all of your suggestions. I did as instructed, including removing the idle screw and cleaning out its channel. Cleaning the EACV, throttle body, etc. In fact, I got quite facile at removing bolts and hoses. Wiring was a bit harder, but got all all of those apart and back together correctly.
Unfortunately, the starting and stalling problem has not gone away. So, I guess that it is off to the dealer where the old NSX mechanic has decided he only wants to change oil. But they said that "my guys are good" and we'll get it fixed ---- whenever.
Sigh.
Thanks again.
Unfortunately, the starting and stalling problem has not gone away. So, I guess that it is off to the dealer where the old NSX mechanic has decided he only wants to change oil. But they said that "my guys are good" and we'll get it fixed ---- whenever.
Sigh.
Thanks again.
Your EACV cleaning efforts have likely eliminated the EACV as the cause.
As a final kick at the can, since this is a 1991 have the fuel injectors ever been cleaned? When you power up the ECU and the fuel pump goes through its prime cycle, most fuel injection systems open the injectors briefly to prime / wet the inlet ports with fuel. This brief priming is typically based upon a temperature based look up value. When you crank the engines the ECU will then typically fire the injectors based upon a different temperature dependent look up table value. The ECU uses look up values because the MAP signal during cranking is too flakey to use the running MAP based fuel calculation. Once the engine gets above a certain RPM (probably 300 - 400 RPM) the MAP measurements smoothen out and the ECU transitions to calculating fuel based upon the MAP signal.
If the injectors are a little bit dirty the fuel flow will be less than desired. This can cause a few starting problems. During the prime and cranking phases fuel delivery may be less than optimal. This leads to a case where the engine seems to start; but, the fuel delivery during this phase is insufficient to make the transition to the MAP based fuel delivery and the engine does not sustain operation. You try this a couple of times and on the 2nd - 5th time the engine catches and sustains operation. This will likely be dependent on ambient temperature. What is happening is that each time the engine catches and then dies you heat the heads and the attached intake runners a bit. This changes starting conditions - typically requiring less fuel for start up. Once the engine is running and up to temperature it goes into closed loop operation which may be sufficient to hide and problems associated with dirty injectors.
Two other fuel system factors to consider. When you turn the key to run; but, not all the way to start you can hear the fuel pump run for a couple of seconds to prime the fuel system. It is during this phase that the ECU fires the injectors to wet the intake ports. If the check valve on the fuel pump is leaky or the fuel pressure regulator is leaky the fuel system looses pressure while parked. If you immediately turn the key to start you pretty much skip the fuel pump prime cycle and you are effectively trying to start the engine on reduced fuel pressure. I normally turn the key to run, listen for the pump to prime and then start the engine which probably gives me a close to 100% first try success rate for starting. Sometimes I forget and turn the key immediately to crank. This normally gives me a first try start; but, if the engine is cool sometimes I don't get an immediate start. I expect that the 23 year old check valve on the pump or the fuel pressure regulator is allowing loss of pressure which is compromising the start. 2nd try usually does the trick.
Do a simple experiment. Turn the key to run; but, not start and let the fuel pump prime. Turn the key off and then repeat the key on - key off exercise again before trying a start. This wets the manifold more than normal and helps insure that the fuel system is fully pressurized during the start sequence. If this improves the starting problem the problem could be
- dirty injectors
- fuel system is losing pressure when parked (pump check valve or leaky fuel pressure regulator). You can live with this problem once you know the symptoms
- check the fuel pressure as per the service manual. A tired fuel pump or a clogged fuel filter or a faulty fuel pressure regulator can cause low fuel pressure which can compromise starting. If the fuel pressure problem is not way below spec closed loop fuel control can hide this problem during normal operation.
As a final observation. If the injectors have never been cleaned, having the injectors cleaned on a 32 year old engine would never be a complete waste of money even if they are not the cause of the problem.
As a final kick at the can, since this is a 1991 have the fuel injectors ever been cleaned? When you power up the ECU and the fuel pump goes through its prime cycle, most fuel injection systems open the injectors briefly to prime / wet the inlet ports with fuel. This brief priming is typically based upon a temperature based look up value. When you crank the engines the ECU will then typically fire the injectors based upon a different temperature dependent look up table value. The ECU uses look up values because the MAP signal during cranking is too flakey to use the running MAP based fuel calculation. Once the engine gets above a certain RPM (probably 300 - 400 RPM) the MAP measurements smoothen out and the ECU transitions to calculating fuel based upon the MAP signal.
If the injectors are a little bit dirty the fuel flow will be less than desired. This can cause a few starting problems. During the prime and cranking phases fuel delivery may be less than optimal. This leads to a case where the engine seems to start; but, the fuel delivery during this phase is insufficient to make the transition to the MAP based fuel delivery and the engine does not sustain operation. You try this a couple of times and on the 2nd - 5th time the engine catches and sustains operation. This will likely be dependent on ambient temperature. What is happening is that each time the engine catches and then dies you heat the heads and the attached intake runners a bit. This changes starting conditions - typically requiring less fuel for start up. Once the engine is running and up to temperature it goes into closed loop operation which may be sufficient to hide and problems associated with dirty injectors.
Two other fuel system factors to consider. When you turn the key to run; but, not all the way to start you can hear the fuel pump run for a couple of seconds to prime the fuel system. It is during this phase that the ECU fires the injectors to wet the intake ports. If the check valve on the fuel pump is leaky or the fuel pressure regulator is leaky the fuel system looses pressure while parked. If you immediately turn the key to start you pretty much skip the fuel pump prime cycle and you are effectively trying to start the engine on reduced fuel pressure. I normally turn the key to run, listen for the pump to prime and then start the engine which probably gives me a close to 100% first try success rate for starting. Sometimes I forget and turn the key immediately to crank. This normally gives me a first try start; but, if the engine is cool sometimes I don't get an immediate start. I expect that the 23 year old check valve on the pump or the fuel pressure regulator is allowing loss of pressure which is compromising the start. 2nd try usually does the trick.
Do a simple experiment. Turn the key to run; but, not start and let the fuel pump prime. Turn the key off and then repeat the key on - key off exercise again before trying a start. This wets the manifold more than normal and helps insure that the fuel system is fully pressurized during the start sequence. If this improves the starting problem the problem could be
- dirty injectors
- fuel system is losing pressure when parked (pump check valve or leaky fuel pressure regulator). You can live with this problem once you know the symptoms
- check the fuel pressure as per the service manual. A tired fuel pump or a clogged fuel filter or a faulty fuel pressure regulator can cause low fuel pressure which can compromise starting. If the fuel pressure problem is not way below spec closed loop fuel control can hide this problem during normal operation.
As a final observation. If the injectors have never been cleaned, having the injectors cleaned on a 32 year old engine would never be a complete waste of money even if they are not the cause of the problem.
Similar threads
