Baulky Manual Gearbox

[Accelera Comigo]

Hi Folks
Can anyone offer a technical opinion on this?
I have just imported a 1991 NSX from Japan, a 20,000 mile manual, thought you might like a picture.
Looking through the history I think the car was used a bit by the first owner for 20 years, it has since done very little so probably stored somewhere for 10+ years.
The gearbox is "bauky," a bit stiff in the morning and a bit stiff when pressing on, on faster roads. So, difficult to pull the stick out of fifth and then difficult to pull into fourth, say. No crunching at all just a recalcitrant synchromesh. All seems fine at low speeds, when the synchro is not doing much.
Gearbox oil has just been changed, made no difference.
Am I seeing a symptom of long term storage? Would it be best to change the gearbox oil several times? Have you had similar problems? Does it look like a knackered synchro? Any other thoughts?
Thanks for listening! Accelera Comigo (Oliver)

[Kaz-kzukNA1]

You'll need to look into this from both the synchro and the CL disengagement.


What MTF used?
What was the state of the meshed filter? Any blass/yellow-ish metal powder trapped or just grey powder?

Any difficulties in selecting other gears?
For example, the 1st, 2nd or Reverse gear.
This will tell whether your CL Hyd Sys or other factors causing input/mainshaft not slowing down.

Kaz

[Accelera Comigo]

Hi Kaz

Thanks for the reply, I can answer a couple of those:

Oil used was Honda MTF. Done as part of a belt service at Nortonway.
Oil condition: I asked them to report back any colours or bits and they told me it was fine.
Oil filter: Sorry, didn't know there was a filter in the gearbox.
Other gears: yes, similar problem in all gears.

I've attached a sound file here recorded on a phone with a remote mic on the gearbox casing. With the engine at idle I was pushing the gearstick into first gear between 9 seconds and 15 seconds. Later in the sound file I raised my foot off the clutch, then pressed the clutch to the floor, then pushed the gearstick until a gear engaged.

You can hear the syncromesh working quite hard, I like the theory that it could be a bit of clutch drag. It more than likely has the original twin plate clutch. I had them bleed the hydraulics through as part of the service, made no difference.

Thanks, Oliver

[Old guy]

I am inclined to go with a dragging clutch, or at least rule it out first as the cause.

On a more conventional car, as a simple test you could put it into first or second gear and with the engine off have someone sit in the driver's seat and press and hold the clutch pedal to the floor. Then have volunteer(s) try pushing the car. If the car is hard / impossible to push you may have a clutch release problem.

The NSX doesn't have a lot of / any good places to push on so this type of test is difficult. Hooking a rope on to the front tow loop and having the volunteers engage is a tug of war with the car might be an interesting option.

There is no single synchro. Each gearset has its own set of synchros and it is usually the first and second gear synchros that start failing / grinding / baulking first unless the car has seen significant duty on a racetrack. It would be odd for all the synchros to fail at the same time. However, depending on how long the car was inactive the synchros and perhaps the shift selection mechanism may have become gummed up. I have never experienced anything like that. However, if it is a 1991 car and it has been operating on the same transmission fluid since 1991 prior to the recent service and has been inactive for 10+ years that would be terra incognito in terms of experience for me. Some operating time and subsequent flushes may or may not improve things.

When they bled the hydraulics, did they check the motion of the slave cylinder / clutch fork? The seals in the master cylinder and or slave cylinder may have failed after sitting idle for 10+ years. Internal piston seal failure in the MC may not manifest itself as external leaks, it just doesn't develop pressure. Unfortunately, I don't recall a spec in the service manual for clutch fork movement. Kaz may be able to advise.

Do you have the maintenance records for the car? Aside from the transmission fluid change records if there is no record of replacement of the engine coolant that is something that I would have done tout de suite. Coolant typically acidifies as it ages and that is a time based problem with use (heat cycling) be a contributing factor. If the records indicate 32 years with the same coolant in the engine I would include a new thermostat and water pump as part of that work and probably all new hoses in the cooling system.

Even being carefully stored for 10+ years is generally not a good thing for the moving parts of the car.

[Accelera Comigo]

Nice idea, thanks "Old Guy."

I think you are both right. On a fast road, as well as being baulky to push into gear it is very stiff to pull out of gear. Surely a worn syncromesh wouldn't affect it coming out of gear, just going in?

The only thing that would make it difficult to come out of gear while pushing on a bit would be clutch drag. My gearbox experience is limited to other gearboxes but other reasons for stiffness coming out of gear might be sticky oil deposits on the selector sleeve or on the detent bearings but they wouldn't be speed dependent.

I'm thinking I should change the clutch before it wears down my syncromesh.

Thanks, Oliver

[Old guy]

With only 20,000 miles, it would be unusual for the actual clutch to have worn out. However, having been to Japan the driving and traffic conditions do seem to be able to contribute to early end of life for clutches. Perhaps that is the reason that NSXs with automatics are more common in Japan.

As noted in my subsequent edit, despite the recent fluid flush you could have a clutch hydraulic system failure. Have that checked before considering actual clutch replacement. Replacement of the OEM dual disk clutch will be fairly expensive. It would really chap your butt if the clutch was replaced and you still had clutch drag.

[Accelera Comigo]

Thanks again, Old Guy. Will check clutch fork throw to avoid "chapped butt."

"However, if it is a 1991 car and it has been operating on the same transmission fluid since 1991"
Unlikely. First owner had it for 20 years and took it to the supplying main dealer several times. I had some fun translating with the app this time round.

"that would be terra incognita in terms of experience for me"
Nice. Do you all speak latin in Saskatchewan? We have it beaten into us at this end then drop it like a hot potato at 16.

When they bled the hydraulics, did they check the motion of the slave cylinder / clutch fork?
I'll ask Bryan, possibly not but he is pretty thorough.

"The seals in the master cylinder and or slave cylinder may have failed after sitting idle for 10+ years. Internal piston seal failure in the MC may not manifest itself as external leaks, it just doesn't develop pressure"
Good thought. Might be worth resealing the MC as a matter of course.

Do you have the maintenance records for the car? ...... tout de suite
Full history, in japanese of course. French is a whole other story, it's called the English Channel and that's that.

"If the records indicate 32 years with the same coolant in the engine I would include a new thermostat and water pump as part of that work and probably all new hoses in the cooling system."
All just done on top of the service, new hoses throughout. Old ones looked OK when they came off actually, I think it's been well looked after just not really used much.

If I've got an internal hydraulic leak I could leave the pedal depressed overnight to see if any fluid passes the seals. Would that tell me anything?

Oliver

[Old guy]

If I've got an internal hydraulic leak I could leave the pedal depressed overnight to see if any fluid passes the seals. Would that tell me anything?

Oliver

That trick will work with brake master cylinders. Press the pedal down and if the pedal eventually sinks to the floor you have an external leak or the MC piston seals are allowing leakage back up into the reservoir. That does not work with the clutch since the clutch MC generally allows full travel of the pedal right to the floor as normal operation so you don't get any 'eventual sinking' that indicates a leak. Pretty much need to check by measuring the fork movement - providing you know how much the fork is supposed to move. If you do have a leaky MC I expect that it will be much cheaper to purchase a new replacement rather than attempt a rebuild if the labor rates are similar to what they are here.

No Latin in grade school. A fair amount of French which is now exceedingly rusty. The Latin comes from working with lawyers drafting contract terms. I think they like to throw in some Latin terms 'just because'.

[Accelera Comigo]

I might have proved nothing, what do you think:
I pushed the clutch to the floor with a rod and left it there for two hours with the car in first gear. After two hours I released the handbrake and pushed it along the drive to see if there was any resistance. None at all, Nullo Resistente Omnio.
Did I prove anything? Oliver

[Old guy]

With the clutch pedal depressed to the floor, if you can push the car easily while it is in first gear that suggests that the clutch is releasing. The two hour waiting period also suggests that you are not getting any leakage past the master cylinder seals. I always hesitate to use the word 'prove'; but, the clutch does seems to be OK.

I think my final clutch related suggestion is to check the free play clearance at the pedal. The pedal does require a small amount of free play to make sure that the clutch is allowed to fully engage. However, if you have too much free play you need more pedal travel to get full disengagement. When you are driving if you always push the pedal right to the floor this loss of motion is probably not an issue. However, when I am shifting between gears I am pretty sure that I do not put the pedal to the floor (although its not something I really think about). If you drive the same way, any lost motion may be impeding a fast release of the clutch. See the attached diagram from the service manual. It is relatively easy to check the free play. Not so easy to adjust the free play.

[Accelera Comigo]

1483114832

Old Guy, Kaz

Thanks for your guidance. I fitted new master cylinder and new slave, gear change is much better. You can see from the shot of the old ones they were suffering a bit. Oliver

[Old guy]

It is good that the problem appears resolved. Master cylinder and slave is a much lower cost fix than a clutch replacement.