Mitsubishi 3000GT & Dodge Stealth Forum banner

21 - 40 of 65 Posts

·
1994 Mit 3000gt VR4 6spd
Joined
·
1,290 Posts
Can it come off by causing the pulley bolt to back out? Yes, it's happened in other platforms, the reason is because it's doing it's job. I wanted to wait and see more pictures and such before giving my thoughts. It seems you either 1. Have install issues (you said something was stuck in between them?) or 2. You have some really bad harmonics or knock your not seeing.

If this would've happened at 15K of hard racing and never touching the bolt I would say yes. But at only 3,000 miles and 500hp, I would say something else is going on.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #22 (Edited)
You're off base on this one... The locating face is 100% on the gear. The gear sits proud of the crank and the crank never contacts the balancer. It's done that way to retain the gear. You can't have the gear rattling around beating up the crank and the keyway.

There's also no loctite residue or any evidence those threads had loctite on them. While I don't feel loctite is needed in that application on a properly installed bolt, if red loctite had been used there, not only would we see the evidence, I'm very doubtful that bolt would have ever fallen out.
Agreed on the Loctite, its clear none was used and it should not be needed. I think my engine builder suffered temporary brain fade - its 18 months since the cam drive was assembled and he has built dozens of race motors since then. Red Loctite would have been a disaster for any future disassembly because you cant apply heat to it without wrecking stuff.

You are correct I was looking at it all wrong so measured it. There is indeed a 15mm gap between the end of the crank and the contacting face of the pulley. All of the load is through the cam drive.

WOW, holy crap, this would never have passed design review in any engineering company I have worked for. Best practice is always to have the faces between and joined by a threaded fastener, to be in intimate contact and under a compressive load equivalent to the tensile load in the fastener. Having a gap between two faces through which a fastener is torqued down, will cause the outer and thinner face to act like a spring. With vibratory loads this will cause the fastener to unload and back out. So there should be an anti back-out feature like a locking tab in this assembly. You see this design compromise on wheel axle nuts and that's why they are always pinned.

I'm surprised there are not more crank pulley bolts backing out. I may add an anti back-out feature on reassembly.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #23
Can it come off by causing the pulley bolt to back out? Yes, it's happened in other platforms, the reason is because it's doing it's job. I wanted to wait and see more pictures and such before giving my thoughts. It seems you either 1. Have install issues (you said something was stuck in between them?) or 2. You have some really bad harmonics or knock your not seeing.

If this would've happened at 15K of hard racing and never touching the bolt I would say yes. But at only 3,000 miles and 500hp, I would say something else is going on.
I hope you are wrong on knock! For the level of build, its running a pretty mild tune. I'm not seeing conditions even close to cause knock and knock is not shown in the data logs.

I think it backed out because the joined assembly is a badly compromised design, that wasn't designed for the RPM range I'm at. See my long post above. On a race motor this would have an anti-back out feature - or better still, a less compromised design,

I guess its all part of the fun in pushing an 30 year old design to >twice its designed output, lol.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #24
Update - I have the replacement parts ordered, along with a couple of spare cam-belt drive sprockets that I will use to investigate some anti back-out ideas. I was initially concerned that the CAS may have been damaged but on visual inspection it looks good and untouched by the incident.

Meanwhile, I'll plan to make myself a tool to hold stop the crank turning while I torque the bolt down.
 

·
Registered
Joined
·
1,926 Posts
Update - I have the replacement parts ordered, along with a couple of spare cam-belt drive sprockets that I will use to investigate some anti back-out ideas. I was initially concerned that the CAS may have been damaged but on visual inspection it looks good and untouched by the incident.

Meanwhile, I'll plan to make myself a tool to hold stop the crank turning while I torque the bolt down.
Glad to hear damage was minimal. I'd definitely be interested to hear your ideas, on the anti-backout design. The only thing I could think of (which would be a horrible idea IMO) would be to torque the pulley bolt down, drill a hole through the crankshaft/bolt, then install a cotter pin. I'm sure that would throw off the crank balance and cause it's own set of issues.

As for the tool, you can always hit up Oohnoo. He has a crank tool that he designed and makes. Alan (bboyalan) showed it in his picture. However, you need to grind down the pins a little. It was designed for an OEM balancer and the Fluidampr's holes are smaller in diameter, for some reason.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #26
Glad to hear damage was minimal. I'd definitely be interested to hear your ideas, on the anti-backout design. The only thing I could think of (which would be a horrible idea IMO) would be to torque the pulley bolt down, drill a hole through the crankshaft/bolt, then install a cotter pin. I'm sure that would throw off the crank balance and cause it's own set of issues.

As for the tool, you can always hit up Oohnoo. He has a crank tool that he designed and makes. Alan (bboyalan) showed it in his picture. However, you need to grind down the pins a little. It was designed for an OEM balancer and the Fluidampr's holes are smaller in diameter, for some reason.
Thanks for the tip on Oonoo's tool. I already have something similar that I will attempt to adapt. If that doesn't work I'll be giving him a call.

Regarding anti back-out, I agree and don't want to throw off the balance. One of the options I'm looking at is to replace the OEM washer with an X series Nord-Lock. Nord-Lock Group - Secure Bolting Solutions

I'm in the process of getting spec's to see if there is one that may do the job.
 

·
Registered
Joined
·
1,926 Posts
Thanks for the tip on Oonoo's tool. I already have something similar that I will attempt to adapt. If that doesn't work I'll be giving him a call.

Regarding anti back-out, I agree and don't want to throw off the balance. One of the options I'm looking at is to replace the OEM washer with an X series Nord-Lock. Nord-Lock Group - Secure Bolting Solutions

I'm in the process of getting spec's to see if there is one that may do the job.

That's funny. I was just looking at Nord-Lock washers. I use Bolt-Depot for a lot of my SS allen bolts and they stock them:

 

·
Registered
Joined
·
278 Posts
Discussion Starter #28
That's funny. I was just looking at Nord-Lock washers. I use Bolt-Depot for a lot of my SS allen bolts and they stock them:

Yes, there are few proven options, if you need to select for vibration and balance. I looked at locking tabs, safety wire and even an end-plate that could be bolted down to the 3 threaded holes on the face of the Fluidampr - but they all raise balance and/or potential fatigue life challenges.
 

·
Money pit owner.
Joined
·
2,643 Posts
you guys are making me nervous, should I be re-torquing the crankshaft bolt frequently?
 

·
Registered
Joined
·
278 Posts
Discussion Starter #30 (Edited)
you guys are making me nervous, should I be re-torquing the crankshaft bolt frequently?
Yes absolutely, If you have a modified high RPM motor and/or a Fluidampr. If your motor is largely stock and hasn't been touched in >60,000 miles then probably not.

Checking the torque is a 20 minute job at most. It would be a 10 minute added task if you have the car on stands to rotate your tires, replace the pads, change the oil etc.
 

·
Money pit owner.
Joined
·
2,643 Posts
ok, do the accessory belts have to be loosened before re-torquing the crankshaft bolt? If so, then it's a bit more than 20 minutes in my experience lol
 

·
Registered
Joined
·
278 Posts
Discussion Starter #32
Looks like the M14 steel NordLock has more than sufficient capacity.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #33
ok, do the accessory belts have to be loosened before re-torquing the crankshaft bolt? If so, then it's a bit more than 20 minutes in my experience lol
No they don't need to be touched at all. The crank bolt is accessible through the driver side wheel well. Either from the side with the wheel off or from underneath with the wheel on
 

·
Registered
Joined
·
278 Posts
Discussion Starter #34
Update - I'm in conversation with Nord-Lock's US technical rep and hope to get some recommendations specific to this application. If this pans out I'll post up the spec of the Nord Lock I use.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #35

·
Registered
Joined
·
278 Posts
Discussion Starter #36
Another update. It looks like the cam drive sprocket walked out and back in again when the pulley came off and the CAS rotor may have impacted the CAS body after all - I had though id dodged that bullet :eek:

So yet more new parts ordered and I'll have to pull the cam belt after all :mad:
 

·
Money pit owner.
Joined
·
2,643 Posts
yeah, if the metal rotor on the back side of the crank sprocket moves forward, it will easily take out the crankshaft position sensor and that means taking everything off to replace it.
I'm still a bit confused by your initial diagnosis of the damper coming off. Are you thinking that it was not fully seated when you installed it and therefore it wasn't torqued down to the 130 ft/lbs? I'm trying to understand why the Fluid damper would come loose compared to an OEM damper.
 

·
Registered
Joined
·
278 Posts
Discussion Starter #38 (Edited)
I believe the Fluidampr was installed and torqued correctly. Its the unsupported cross-section of damper, under the bolt head that's the issue. It took me a while to figure out that it was unsupported by design,
Its a horrible compromise, which reduces manufacturing cost and steps, but creates what is effectively a sprung section under the bolt head - which is never a good idea.
This compromise is why the torque spec is 130 lb.ft, which is far in excess of what's expected to secure a drive flange. I believe that the reason it was set so high by Mitsubishi, is to pre-load the unsupported cross-section of damper underneath the bolt head.
Its sprung section that compromises the pully drive flange's ability to resist axial harmonic loads. It can deflect under the bolt and loosen it.
It turns out that the OEM damper does occasionally suffer the same kind of failure. I found it documented and there are posts in this forum from the 2000's. It sees to work but could be limited to OEM RPMs.
Now that I better understand the design issues, I could see that axial harmonic loads in the higher RPMs that I'm now running, could overwhelm the pre-load in the unsupported cross section of damper.
My plan so far, is to replace the OEM load washer with a Nord-Lock rated for 130 lb.ft and which can absorb 0.7mm of axial expansion. Then I will add checking it to my routine maintenance schedule.

IMPORTANT NOTE: all of the above is about axial harmonic loads. The damper is primarily designed to manage radial harmonic vibrations, which is seems to do an excellent job of doing.
 

·
Registered
Joined
·
1,926 Posts
So one thing you might want to verify, is if the Nord-Lock washer is thick enough, so the crank bolt doesn't bottom out. What Nord-Lock washer are you going to use? The one on Bolt-depot is only 3.2mm thick, vs the 9.2mm of the OEM washer.
 

·
Money pit owner.
Joined
·
2,643 Posts
I believe the Fluidampr was installed and torqued correctly. Its the unsupported cross-section of damper, under the bolt head that's the issue. It took me a while to figure out that it was unsupported by design,
Its a horrible compromise, which reduces manufacturing cost and steps, but creates what is effectively a sprung section under the bolt head - which is never a good idea.
This compromise is why the torque spec is 130 lb.ft, which is far in excess of what's expected to secure a drive flange. I believe that the reason it was set so high by Mitsubishi, is to pre-load the unsupported cross-section of damper underneath the bolt head.
Its sprung section that compromises the pully drive flange's ability to resist axial harmonic loads. It can deflect under the bolt and loosen it.
It turns out that the OEM damper does occasionally suffer the same kind of failure. I found it documented and there are posts in this forum from the 2000's. It sees to work but could be limited to OEM RPMs.
Now that I better understand the design issues, I could see that axial harmonic loads in the higher RPMs that I'm now running, could overwhelm the pre-load in the unsupported cross section of damper.
My plan so far, is to replace the OEM load washer with a Nord-Lock rated for 130 lb.ft and which can absorb 0.7mm of axial expansion. Then I will add checking it to my routine maintenance schedule.

IMPORTANT NOTE: all of the above is about axial harmonic loads. The damper is primarily designed to manage radial harmonic vibrations, which is seems to do an excellent job of doing.
ok, I had read about OEM dampers "coming apart" presumably due to age, but I didn't notice any mention of the OEM damper bolt backing out, so that is interesting. It also never even occurred to me to check and confirm that the damper bolt could bottom out without the OEM washer. In my experience, once that OEM bolt/washer is torqued to OEM spec (eg 130 ft/lbs), it takes more than double that to loosen it.
It may seem odd, but I only run my car at OEM horsepower/RPM, so I guess I shouldn't see the issue with my Fluid damper. :-}
 
21 - 40 of 65 Posts
Top