1991 Wont rev past 5200

[Acuratech214]

As the title states I have a all factory 91 15k miles that wont rev past 5200 rpm. Hits a hard fuel cut. Vehicle otherwise idles great, pulls hard up to that rpm, no lights on, no codes. Recent work goes as follows: t belt, plugs, oil change, brake flush, air filter, fuel filter, radar system removal, trunk cd changer removal, radio head unit removal. I do not know if the vehicle did it before the service as it sits on a trickle charger and does not move for years.

Diagnostic steps goes as follows: checked compression to confirm timing 185-190 all cylinders, checked fuel pressure= fuel pressure is at spec according to manual, Checked oil pressure at idle, 3k, 4k, 5200= readings are good at all intervals, pulled clock fuse, checked for soft set connectors. fresh fuel.

kinda stuck and out of ideas at this point and seeing if anyone has any other ideas

 

[Valhalla]

You have a VTEC issue. This could be caused by a number of items. If the VTEC controllers don't have a signal from the ECU or if the oil is low, they won't engage. Check to make sure there is the proper amount of oil in the pan. Then check the connectors on the VTEC controllers. If the car has set for a long time, it's possible for the VTEC controllers to gum up or have debris clogging the screens. Look it over and report back.

 

[Old Guy]

If it hasn't moved for years, I expect that you have some gummed up stuff.

I don't recall whether the service manual calls out for a fuel / ignition cut for Vtec problems or just does not switch fuel maps. If the system is failing to generate sufficient oil pressure to operate the VTEC mechanism there is a pressure switch on the front and back cylinder banks that will not operate preventing VTEC engagement. This will generate an error code 51 or 52. If the actual VTEC solenoid has failed that will generate an error code 21 or 22. Note that even if the solenoid tests OK I think it can still be stuck closed causing an oi pressure problem. Normally I would say find your service check connector and use it to trigger the display of any error codes that may be stored in memory. That may give you some info as to what is going on. However, since you have pulled the clock fuse you have wiped out any stored codes.

Check to make sure that the water temperature sensor that the ECU uses (which is not the gauge sensor) is accurate. VTEC will not engage if the engine temperature is below 60C so a faulty sensor can screw things up.

Since you referenced the fuel pressure spec in the manual I presume you have the factory service manual. You could drive around to see if the engine generates any of the VTEC related codes. Failing that the next test would be to set up to measure the actual VTEC oil pressures as described in the service manual.

A couple of comments

- be aware that there are other things that can cause fuel cuts turn on limp mode
- if the car has been sitting for a long time, I would not be taking it out and zooming around. I would be driving very carefully gradually testing higher RPM limits and engine loads. Non operation of engines is not particularly good for them. There are a whole bunch of things such as plugged injectors and stuck VTEC solenoids that can go bad during a long period of non use.

 

[sca037]

I had this exact issue on our 1992 when the Vehicle Speed Sensor (VSS) failed, and it's not an expensive or terribly difficult part to swap out.

A big sign it's the VSS is if your speedo also stops working, otherwise follow the other learned advice given in this thread!

 

[Old Guy]

I had this exact issue on our 1992 when the Vehicle Speed Sensor (VSS) failed, and it's not an expensive or terribly difficult part to swap out.

A big sign it's the VSS is if your speedo also stops working, otherwise follow the other learned advice given in this thread!

Yes, vehicle speed is one of the enabling inputs to VTEC. However, my recollection is that it is enabled at a really low speed (less than 10 mph?). Your VSS would have to be really far out on its calibration or dead which should be apparent on the speedo.

 

[Acuratech214]

Yes, vehicle speed is one of the enabling inputs to VTEC. However, my recollection is that it is enabled at a really low speed (less than 10 mph?). Your VSS would have to be really far out on its calibration or dead which should be apparent on the speedo.

So I will tell you I am a master Honda/acura tech with access to any and all publications that acura has put out on this vehicle. I have no ego asking here for help and welcome all comments.

Oil pressure spec 10psi or above at idle and 50psi or above at 3k as per service manual.

My readings were 25psi idle 75psi 3k. 90psi 4K 90psi 5200

I checked for codes before clock fuse and none where there.

I have also checked all grounds and they are tight, clean.

VSS is working as designed

Coolant temp pcm reads also working as designed

Service manual gives nothing as to what causes fuel cut other than a old bulletin about fuel which didn’t pertain to me.

 

[Old Guy]

I agree that in the service manual (at least my 1997 SM), the only description of fuel cut is for trialing throttle and red line limit. It would be odd to initiate a fuel cut because the VTEC is inoperative. I was of the impression from the characterization done on the pre OBDII ECU that the ECU runs two (at least two) fuel maps. One with VTEC off and one with VTEC active. If the ECU does not get the signal that the VTEC has engaged it does not switch maps and the engine should continue to run up to red line; but, on the no VTEC map. No reason to do a fuel cut; but, perhaps @MouterMouth93 who has done more work on the characterization can advise whether there is a fuel cut if the VTEC pressure pressure switches don't operate.

VTEC operation is also dependent on engine MAP. If 1st or 2nd gear if you do a slow press on the throttle you should be able to get the engine RPM above 5200 RPM without entering hte VTEC operating range. Do you get the fuel cut with this type of slow RPM increase? If so, the problem may be completely unrelated to VTEC.

When you measured oil pressure, did you measure pressure at the VTEC measurement ports in the head? You have to run the engine above 5000 RPM and manually operate the VTEC solenoids to enable the correct pressure measurement.

 

[Zoom]

Maybe something seized over the time the car was stored ? Looks like the oil pressure doesnt reach the VTEC system preventing it from activating ? Also an ECU in limp mode would indeed reduce the fuel cut off but you'd get an error code out of it. Looks like mechanical to me. Another idea is gunk somewhere blocking enough oil flow from reaching the heads.

 

[Acuratech214]

I measured oil pressure at the sensor port by the filter. I’m thinking mechanical too but still don’t know how the ecu would detect that. It isn’t misfiring. There is no info out there as far as the factory as to what can or will initiate limp mode or even mention of it. Let’s say I was a tooth off on the back head I figured compression would show it or even some time of drivability which other than not revving it runs perfect. The only weird thing is it idles around 1k hot, now it sounds like it’s around 750 but gauge says 1k and I don’t know if the gauge is just off. When the car starts cold it revs up super high around 2200 and idles down slow. It’s different for sure from the many others I have worked on but it still idles down as it warms. Then I thought ok maybe the vtec plunger was seized from being dry but it would code. Gave thought to the idea of when I went back together maybe some Honda bond got loose on the back of the heads by cam seals and plugged something but I don’t use a lot and that seems super far fetched.

 

[Acuratech214]

Oil pressure was measured at the sensor port by the oil filter. No matter how I press the throttle 5200 is a hard cut. I thought to myself ok maybe the plunger in the vtec solenoid is seized from being dry, well that should code. Ok maybe when I did the belt some honda bond plugged something, I use the bare minimum and that seems way far fetched. Maybe I am a tooth off on the back head- Figured compression for sure would show that, Would think I would have some running rough concerns, How would pre obd ecu even see that and cut fuel at high rpm.

One thing I have noticed that is different than the many others I have worked on is this car idles down very slow when cold and only gets to 1k. starts at like 2300. The vehicle doesnt sound like its running that high so I figured maybe the gauge is off.

 

[Old Guy]

One thing I have noticed that is different than the many others I have worked on is this car idles down very slow when cold and only gets to 1k. starts at like 2300. The vehicle doesnt sound like its running that high so I figured maybe the gauge is off.

The base idle speed on a manual transmission equipped NSX is 650 RPM when hot. I have a 2000 model year with the aux blower / catalytic converter heating system so I can't really comment on what the typical warm up RPM is for an NA1. The dash tachometer will deviate from the actual RPM due to ageing of the capacitors in the instrument cluster. You should use a tach connected to the test point in the engine bay to confirm actual RPM. Tach circuit drift where the engine is running at 8300 RPM; but, reading only 5200 RPM would be extreme.

The capacitors can be replaced and the tach recalibrated if it is suffering from drift.

The correct way to confirm that the VTEC system is getting the correct oil pressure is to measure pressure at the ports in the head at 5000 RPM. This requires manual activation of the VTEC solenoid to do the test. However, it seems unlikely that your problem is VTEC related.

 

[Kaz-kzukNA1]

MT or AT?
Never converted from AT to MT?

Does the hard fuel cut happens at the same rpm even in different gear positions?

If the gauge cluster was never serviced before, very likely to have big offset at the tacho.

Once fully warmed up, check the resistance of the TW sensor.
Should be less than 400ohm.

There are 3 x coolant temperature sensors on our engine.
You need to check the one mounted at the front bank just below the EGR valve.
It has grey round 2pin connector with tiny black sleeve around the sensor body.

If the ECU thinks that the engine is still cold, it will disable the VTEC and will lower the rev limit but I thought it was around 7,000rpm and not like 5,200 that you were experiencing.

You won't get CEL if the TW sensor got huge offset unless open/short circuit.

There is a small connector near the INJ resistor block at the right side of the eng bay.
It's the rpm signal pickup.
Use the automotive multimeter or multimeter with frequency function.
It's 6 cyl so 3 tdc pulse per rpm so at 800rpm, it's 40Hz.
Even after fully warmed up and you are getting 50Hz (1,000rpm) or higher, your ECU is thinking that it still needs warm up cycle and hence, no VTEC and lower rev limitter.

For the rpm measurement, you can also use the #1 cyl loop wire at the ignitor connecotr.

If fuel pump relay related (enable the bypass of the fuel pump resistor), you would get hesitation and not hard fuel cut so for now, I excluded the fuel pressure from the consideration.

Don't know at what altitude you are in, the battery condition (cranking speed) or the equipment used for the compression check but the data (185 - 190) seemed to be fairly low although the importance is the deviation between the cyl and not the absolute figure.
Normally, I would get around 230 at just over 300ft altitude.


Kaz

 

[Honcho]

Oil pressure was measured at the sensor port by the oil filter. No matter how I press the throttle 5200 is a hard cut. I thought to myself ok maybe the plunger in the vtec solenoid is seized from being dry, well that should code. Ok maybe when I did the belt some honda bond plugged something, I use the bare minimum and that seems way far fetched. Maybe I am a tooth off on the back head- Figured compression for sure would show that, Would think I would have some running rough concerns, How would pre obd ecu even see that and cut fuel at high rpm.

One thing I have noticed that is different than the many others I have worked on is this car idles down very slow when cold and only gets to 1k. starts at like 2300. The vehicle doesnt sound like its running that high so I figured maybe the gauge is off.

I'd hook a vacuum gauge up to the engine and measure during idle. If the needle is bouncing, you're off by a tooth. This is usually not enough to throw a code, however and, as far as the ECU is concerned, the crank/cam position is within spec. Many NSXs are out there driving with a tooth off on the belt. I doubt this is your problem, because again, the ECU would only trigger a limp mode if there was a fault in the cam position.

I'd also take a look at your VTEC system. Inspect the VTEC plungers and check for 50+ psi at the 10mm inspection ports on the head in VTEC. If there is an oil pressure fault within the VTEC passages (they feed off of Main #4 ), the car won't throw any codes because the ECU will just think it's in non-VTEC mode and run on Map 3. Same with if the solenoids are stuck- you'll just be on Map 3 with a redline of 7,000 rpm. The 5,200 cutoff is odd because unlike the "non-VTEC" limit of 7,000, the 5,200 appears to be a hard-coded limp mode.

Now that I think about it, 5,200 rpm is right around where the fuel pump switches to "high" in the fuel maps (it's variable). If there is a problem with the fuel supply system, the ECU will hold it in "low" mode and not let you past the RPM threshold to protect the engine from a lean condition. It may or may not throw a code. I would run all of the fuel system tests (pump, resistor, injectors, etc.) to see if you can isolate it. The fuel pumps are getting so old on these cars that we are seeing a lot more failures lately. And, the pumps don't just die, they get weak and intermittent. Also, if the injectors are gummed up with varnish, that could be causing it too- might be worth it to send them out for cleaning and balancing.

 

[Old Guy]

I'd hook a vacuum gauge up to the engine and measure during idle. If the needle is bouncing, you're off by a tooth. This is usually not enough to throw a code, however and, as far as the ECU is concerned, the crank/cam position is within spec. Many NSXs are out there driving with a tooth off on the belt. I doubt this is your problem, because again, the ECU would only trigger a limp mode if there was a fault in the cam position.

Honcho

How big are the fluctuations? My NSX has always had a bit of vibration when cold and I have often wondered whether it might be a one tooth out problem. I am due and have all the parts for a timing belt / water pump replacement; but, I have also been really good at finding excuses to do other things. If I check the MAP and it looks like I am off by a tooth that might motivate me to get my act together.

What port do you use for the MAP measurement?

 

[Honcho]

Honcho

How big are the fluctuations? My NSX has always had a bit of vibration when cold and I have often wondered whether it might be a one tooth out problem. I am due and have all the parts for a timing belt / water pump replacement; but, I have also been really good at finding excuses to do other things. If I check the MAP and it looks like I am off by a tooth that might motivate me to get my act together.

What port do you use for the MAP measurement?

Noticeable for sure. You can use the PCV vacuum source on the top of the intake manifold- the car can go without PCV for a few minutes. The gauge needle should be steady. If it's "dancing" as the great LarryB said, you're off by at least a tooth. Many NSXs are off by one either because the belt is 30 years old and loose, or the tech was sloppy on the install and did not tension the belt properly. The whole "loosen the bolt and allow the spring to tension the slack" step is more art than science in my opinion, because if the tensioner is even a little cockeyed on its pivot point, you won't get the right spring tension. I had to do it 3 or 4 times on this current engine before I was happy with the tension.

 

[Old Guy]

I will see if I can find my vacuum gauge and give it a test.

Update

Well that was a treat!

Doing the actual vacuum measurement was easy. Setting up for the measurement - not so easy. That PCV hose has been on that manifold nipple for 22 years and it did not want to part company. It took me about 15 minutes, two seal picks. a pair of narrow pliers and a punctured finger to get the hose to part from the nipple. The PCV hose is fairly large diameter, much larger than the adapter on the end of my vacuum gauge so I had to cobble together some intermediate hoses and adapters to get the gauge connected. The vacuum test came out at a steady 18 in Hg. There was a very slow fluctuation from about 17.8 to 18 in Hg; but, I noticed that my idle was also fluctuating slowly so I assume that the loss of the air flow through the PCV system was upsetting the drive be wire idle control system. Reconnecting the PCV system restored the steady idle speed.

So the steady vacuum reading suggests no one tooth off problem.

Based upon my PCV hose experience I am not looking forward to my coolant hose replacement project.

 

[Acuratech214]

Vacuum is 21.5hg steady at idle which rpm gauge says 1100.

Every gear cuts out at 5200 (this is a m/t vehicle)

t/w sensor 376 ohms fully warmed up

 

[docjohn]

you have some very knowledgeable folks helping...one of which is "factory trained"

 

[Acuratech214]

map sensor reads as follows

idle .90v
2k .75v
3k .75v
4k .81v
5200 .86v

 

[Old Guy]

Your Tw sensor reading is towards the high end; but, within the service manual allowable range of 200 - 400 ohms. If you wanted to rule it out as a possible cause you could unplug it and shove a 220 ohm resistor in the plug to 'fake' a higher temperature and see if it makes the problem go away. However, I suspect Tw is not your problem.

Your vacuum reads higher than mine; but, that would be consistent with idling at 1100 RPM. My measurement was at 650 RPPM. That idle speed is very high for a hot engine. Is that the RPM according to your dash gauge or an external tachometer? Have you confirmed that the dash tach is or is not reading high?
[MENTION=18194]Honcho[/MENTION] mentioned this; but, Kaz seems to suggest that it is not the cause. Have you checked to confirm that the relay that by-passes the fuel pump resistor is working? If you check the schematic diagrams, you will see that the ECU monitors the fuel pump voltage. If the resistor by-pass relay (Honda calls it the fuel pump relay) does not operate the voltage on the pump stays low and the ECU may block operation above the fuel pump switch point or maybe it just triggers an error code (the service manual is not specific about that). I think that if you shunt the fuel pump resistor so that full voltage is applied to the pump all the time you should be able to test whether this is causing your problem.

 

[Honcho]

you have some very knowledgeable folks helping...one of which is "factory trained"

That's an understatement
[MENTION=38866]Acuratech214[/MENTION], even though Kaz does not suspect it, I would try to jump the resistor and force the fuel pump to high and see if your rev limit goes away, as [MENTION=26435]Old Guy[/MENTION] suggests. This feels more and more like a fuel system issue to me.

 

[Acuratech214]

My vac readings were done with the gauge saying 1100 but i confirmed the gauge was off by 300 rpms. I would think my 21.5 is perfect with a low mileage vehicle and your 18 would be more indicative of a higher mileage vehicle. my current elevation is about 500 above sea level according to google for dallas tx. going to dig further into the fuel system and see what i find

 

[docjohn]

out of curiosity is this your car or a customer's..?

 

[shingli2k]

As the title states I have a all factory 91 15k miles that wont rev past 5200 rpm. Hits a hard fuel cut. Vehicle otherwise idles great, pulls hard up to that rpm, no lights on, no codes. Recent work goes as follows: t belt, plugs, oil change, brake flush, air filter, fuel filter, radar system removal, trunk cd changer removal, radio head unit removal. I do not know if the vehicle did it before the service as it sits on a trickle charger and does not move for years.

Diagnostic steps goes as follows: checked compression to confirm timing 185-190 all cylinders, checked fuel pressure= fuel pressure is at spec according to manual, Checked oil pressure at idle, 3k, 4k, 5200= readings are good at all intervals, pulled clock fuse, checked for soft set connectors. fresh fuel.

kinda stuck and out of ideas at this point and seeing if anyone has any other ideas

At what MPH is it cutting off in gear? Is it engaging VTEC? I had this happen to my vehicle where I thought my car wasn't revving past a certain RPM, but it turned out to be a malfunctioning gauge cluster. The Capacitors leak over time and causes failures so they need to be changed out. I was cutting off at like roughly 6k rpms but car felt normal and VTEC was engaging. It wasn't till I looked at what the actual speed that the fuel cut off was till I figured out my tachometer wasn't working.

 

[Acuratech214]

Couldnt rap my head around why I havent found the issue already with the checks i have done. so went over everything again, found its the tach gauge thats off.

gauge reading 800rpm car reading 800rpm
gauge reading 2k car reading 2200
gauge reading 3k car reading 3600
gauge reading 4k car reading 4800
gauge reading 5k car reading 6500


thanks all that helped going to find someone that can rebuild the gauge now