The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support ExtremePSI
Please Support ExtremePSI

Clutch Too Aggressive? Here's a way to calm it down, requires a little fabrication

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

clutchtamer

Proven Member
49
13
Apr 29, 2015
Concrete, Washington
A little background first, here’s something to think about- if you know an engine’s average steady state torque output over a defined rpm band, as well as it’s rate of full throttle acceleration (without external load) over that same rpm band, you can calculate the approximate torque-seconds required to accelerate that engine’s rotating mass across that rpm band. From that, you can calculate how the required torque changes when that acceleration period is compressed or extended (twice the torque required to do the same work in half the time). This gives you useful numbers to estimate the torque spike intensity created from rpm loss during either a launch until clutch lockup, or during a gear change.

I’m using general numbers here, hopefully to make this concept easier to understand….
…Lets say a generic 450ft/lb engine is tested for how fast it can accelerate WOT with the clutch pushed in, no load applied. Acceleration rate is found to be 8500rpm/second thru the heart of it’s torque band. At this rate, the engine gains roughly 2000rpm in .235 seconds.
…Now the car is launched and .235 seconds later, rpm has dropped 2000rpm as the clutch locks up. If it took 450ft/lbs to accelerate the engine’s inertia 2000rpm in .235 seconds, it also took 450ft/lbs to remove 2000rpm from that launch rpm in the same .235 second time period. Where did this 450ft/lbs of inertia energy discharged over .235 seconds go? Into the transmission’s input shaft along side the engine’s WOT 450ft/lbs, That’s a total of 900ft/lbs for a brief .235 second time period, then torque drops below 450ft/lbs as the engine starts gaining rpm and some of it’s output is siphoned off as energy being recharged as inertia back into the rotating assembly. To the driver, this change in overall torque output might feel like a bog.

From there you can play with the rate of the rpm loss and how that effects the torque that the input shaft will see. To remove 2000rpm from the rotating assembly over twice the period of time requires ½ the torque, so doubling duration to .47 seconds will cut that torque spike in ½ to 225ft/lbs. In this instance, the input shaft will see 675ft/lbs for .47 seconds before dropping below 450ftllbs. Speed up that 2000 rpm loss to just .118 seconds (typical of what you might see with a grabby clutch), the torque spike increases to 900ft/lbs. In that case, the input shaft would see 1350ft/lbs for .118 seconds before dropping below 450ft/lbs. This would feel like a huge bog, and be more likely to break a transmission. Change the ultimate limiting factor from breaking a transmission to exceeding the ultimate traction potential of the tire, reducing this torque spike is beneficial either way.

As you know real life power application isn’t binary like my example, but I hope this helps illustrate where I’m coming from.

Another way of putting it, which path do you think would result in a quicker car?...

...if you are launching from 6000 and rpm dips to 4000 as the clutch locks up, you launched with inertia energy in addition to the power that the engine was making. That added inertia energy was only temporary though, and has to be repaid as the engine begins gaining rpms. But the drivetrain had to have enough reserve capacity to handle that inertia surge, so you are only really using the drivetrain to it's full capacity for the first few tenths of a second (during the inertia discharge).

...add clutch slip. If you don't lose rpm after launch, there is no added inertia energy being dumped into the drivetrain. You are now launching on motor power alone, so little reason to launch any higher than your torque peak (~4500rpms). The real advantage comes when you realize that your drive train now has extra reserve capacity that you are not using! Now you can add more engine power without breaking the drive train. Unlike inertia energy that only lasts a few tenths of a second, added engine power will be there for the entire duration of the run.

Any added launch energy from a 6000+ launch comes from an inertia energy discharge, which results in rpm loss. If your rpm does not dip after launch, the stored energy of a 6000+ launch will net you little beyond what you could get launching from your torque peak (maybe 4500?). The lower rpm launch will result in a lot less clutch wear, as a 6000 launch packs 78% more inertia energy than one at 4500. This might seem counter intuitive, but you may see less wear/tear on the clutch. Despite slipping for a longer period of time…
…launching from a lower rpm without bogging = less total revolutions of slip
…most of the slip will now be with less than full PP pressure on the disc.

Later today when i get some time, i'll post up some pictures of how you can take advantage of this with a little simple fabrication.
 
Your talking to a turbo crowd that does not have 450tq on the 2 step let alone most of the people make no where near 450tq during a 4th gear pull. Not to mention how much less power the engine makes due to flame front travel time in a lower gear in relation to ignition timing as as well as boost pressure normally being lower as you have a hard time making the boost in such a short unloaded environment.

I am interested to see what your going to show us, and by looking at your name "clutch tamer" if this is some unsupported sales pitch.

What kind of car do you have and how fast have you went with an AWD turbo 4 banger?
 
If it took 450ft/lbs to accelerate the engine’s inertia 2000rpm in .235 seconds, it also took 450ft/lbs to remove 2000rpm from that launch rpm in the same .235 second time period. Where did this 450ft/lbs of inertia energy discharged over .235 seconds go? Into the transmission’s input shaft along side the engine’s WOT 450ft/lbs, That’s a total of 900ft/lbs for a brief .235 second time period, then torque drops below 450ft/lbs as the engine starts gaining rpm and some of it’s output is siphoned off as energy being recharged as inertia back into the rotating assembly.

Huh?

I'm thinking that if it took 450ft/lbs of torque to free rev the engine to 2000rpm, that is a tight ass motor. :)
 
Very confused Craig. Took a long time to write it all too....
 
Very confused Craig. Took a long time to write it all too....

I've read it 3 times and still haven't got a clue what I'm reading. Maybe I'm just slow today.

That’s a total of 900ft/lbs for a brief .235 second time period...

I am however, EXTREMELY interested in some of those magic beans that somehow get 900 ft/lbs of torque out of a motor that is only capable of producing 450 ft/lbs. :D
 
I am however, EXTREMELY interested in some of those magic beans that somehow get 900 ft/lbs of torque out of a motor that is only capable of producing 450 ft/lbs. :D
I think he's trying to say that with the rotational inertia produced by a spinning clutch at high RPM, that it will add to the torque put on the drivetrain..... albeit for only a very short time. Anyways that's what I understood from it. Whether that calculation is correct or not is entirely different.
 
The clutch is directly coupled to the motor the instant it fully engages. Therefore, any torque that is measured at the input to the transmission is still just a function of total engine torque, once the clutch is engaged. During the time it is in the process of engaging, any excess power not absorbed by the drivetrain or transferred back into the engine as a shock load is being lost to friction and heat.

In order for the clutch to actually add any meaningful power to the system, it would have to have more rotational inertia than the engine's entire rotating assembly (including the flywheel and pressure plate) at the point of clutch engagement, and be able to maintain that inertia advantage throughout the engagement process without any losses due to friction and heat (where can I get THAT clutch? :D ). Not saying it can't happen, but I would want to see reputable engine dyno results before I can believe that anyone is spinning a DSM clutch fast enough to add 450 ft/lbs of torque to an engine's output. (Actually, I'd settle for 50 or 100 lbs of added torque). Think about it... how heavy would the clutch disk need to be to overcome the inertia of the engine, flywheel, pressure plate, and harmonic balancer, even if you somehow spun it up to a couple thousand RPM faster than the engine before engaging it?

I can guarantee you that my little Clutchnet disk isn't going to add a single solitary ounce of force to my drivetrain, no matter how fast I spin it. :)
 
Last edited:
Well, in any case, Id like to see this thing, as Ive been considering the thing Magnus sells for a while, it just looks like it might be kind of a pain to get adjusted just right.
 
In order for the clutch to actually add any meaningful power to the system, it would have to have more rotational inertia than the engine's entire rotating assembly (including the flywheel and pressure plate) at the point of clutch engagement, and be able to maintain that inertia advantage throughout the engagement process without any losses due to friction and heat (where can I get THAT clutch? :D ). :)
With the flywheel being an energy storage device, you are not having to overcome that because it is spinning and the momentum from that is also used to propel the car forward during a launch, but only for a short time. Look at those Fly Wheel toys they sell. You spin them up very fast and the inertia from it spinning translates to a torque that moves it forward. That force does not simply get canceled out once it touches the ground (or clutch is engaged) but is transferred from the rotating mass into in this case, the clutch, transmission, and ultimately to the ground.
 
A hit miss engine used to drive tractors back in the day. They used a heavy flywheel to keep the engine rotating until it fired off. inertia. Also kept the tractor moving. This is an interesting thread. Subscribed.
Even though I went automatic, I really enjoy seeing these manual threads.
Every bit helps. I still have my manual in the event I want to go back, and this might be an option.
Might want to hear Kurt out, you're gonna learn something.
 
Torque = MOI * angular acceleration. MOI for anything round is 1/2 Mass*radius*radius.

So an average n/a 4g probably makes about 100 ft*lbs of torque, it takes about .5s to go from idle to revlimiter. Our angular acceleration is 125-16/.5 218rps/s our MOI is 218/100 = .458lb*ft. Now if we dump the clutch from 6500, and it takes .1s to lock up the clutch and kill the engine. 108-0/.1 = 1080rps/s Plug that in 1080*.458 = 494ft*lbs spike on our input shaft. See this issue? Flywheels are energy storage devices.

I agree with you entirely. I think putting it in terms of angular acceleration per the mass of the all rotating part at their differential speeds is really the issue of efficiency we're talking about here. The balance between efficiently/quickly transferring torque and retaining angular momentum is a fine line and allowing the rate of that to be adjusted to our specific engine's and drivetrains is a key component to any successful launch.
 
The rate of heat release in an engine has a very very very minor effect the total power. Like less than 5%. Furthermore how does the "flame front travel time" have any idea what gear the engine is in? I understand that in the lower gears the angular acceleration is higher, and the piston speed is faster at tdc, than it was when combustion was initiated, but it's no where near as bad as you would think, at worst you might put peak pressure a degree or 2 behind optimum, and that barely hurts power..

As far as boost pressure, I don't know what your talking about, but my setup is very laggy, I used to launch at about 20psi, and as soon as the clutch came out it buried a 30psi gauge. I actually backed off my launch boost a bit to make it more consistant, and it works great, except my my flames are to slow, and they can't keep up with the car in first gear.

Just search it. There is real world info out there. It can easily be worth investigating for yourself if you want the most power in each gear. Engine accelerating 3 times as fast has a larger affect than one would think. Not to mention the less likely hood of running into any combustion instabilities. Most engines fail in high gear.

What do you think I am referencing to try to poke fun at me in regards to you interesting approach at launching your extremely "laggy" Hx35 setup. Maybe if you ran off the 2step and had timing trim by gear you wouldn't need 20psi to move a 1g off the line. I would assume that's why it has such high traps with such a slow elapsed time(wheel spin).

Its all in relation to sustained load and knock thresholds and peak cylinder pressures and rod angles and piston velocities and apparently putting 900 ft/lb of torque into the input shaft which would equate to 2700 lbs of force on first gear following roughly a 3:1 gear ratio.
 
I don't have to search it, it's what I do everyday, but I would appriciate if you could in any sort of an article backing your claims. As far as real world info, that's laughable. It's certainly not something your going to measure on any tuner shop's dyno.

So for a quick figure, my car typically pulls down to 4500 as the clutch is engaging, and I hit 9000 right at about the 60' in 1.5s I wrote and excel file to figure it out. over the about 40* of combustion duration, the acceleration allows the piston to be .006* farther ahead than it would be at steady state. The cycle to cycle variability is much much greater than this. The acceleration is just not a factor. I really hope you didn't hear this while at some tuning seminar. You've been had.

I see a lot of buzz words in your post, but no real understanding behind it.

As far as my car, it's a 5 speed street car, I don't think any of your tom-foolery would help. On the street tires it leaves just fine at 0 psi and without a 2 step. It leaves just fine on about 5psi on the slicks, I was just illustrating a point that anything should be able to make as much boost as wanted on the 2 step.

TRE says it takes 1200ft*lbs to do this, so does the math.... my car has stock rods and pistons right now, so either the intertia of the rotating assembly adds a f***load of torque, or I have the highest torque stock block evaarrrr!

You seem like a guy who looks at numbers all day and hates real world data. There have been many times where I have been involved in someones calculations and those have been fractions to what the real world results were(in a positive or negative way). AEM and other companies actually have this function already in there software. I have seen up to differences of up to 8* advance on a tuned map from 1-2 to 5th(they also do boost by gear and fuel by gear).


"If you tune a vehicle in a lower gear, it may knock in higher gears since the duration that it sits in each load cell is so much longer.

If you tune in a higher gear, the lower gears are generally less aggressive than they would allow.

  • Fuel trim per gear
  • Timing trim per gear
  • Boost per gear
This would be about as flexible as you'd ever need it to be to get the most out of each gear.

Here is an example. I had an Evo 8 with AEM EMS that was strictly for 1/4 mile racing. He ran a MBC and it was tuned for 4th gear since that is the gear it will be in for roughly half of the track. I then did pulls in 2nd and 3rd gear as well to see how the power was in each gear. The results were roughly (this is all off memory so don't take it as gospel):

2nd gear: 440whp
3rd gear: 490whp
4th gear: 505whp

After adjusting the fuel and timing trims for 2nd and 3rd gear:

2nd gear: 480whp
3rd gear: 515whp
4th gear: 505whp

To test it out, we did three 1/4 mile simulation pulls (waited 20 minutes between each pass) with no trim adjustments and looked at the average WHP, ET, and MPH. Then we did the same with the trims adjusted and did three 1/4 mile pulls again. I forget the exact results (I will dig them up later) but the average whp was about 20-30whp more and he shaved an average of .3 seconds off his times." Jamie worked at dynotech at the time of the post.

If you want I can ask for the real data, but I only looked for this because I was interested in the results again. It is obviously very laughable that you think real world data is worthless.

1200ft lbs the engine produces or through torque multiplication? I just was chatting with TRE the other day about this and would like to know if he said the engine produces 1200tq?
 
Last edited:
1200ft lbs the engine produces or through torque multiplication? I just was chatting with TRE the other day about this and would like to know if he said the engine produces 1200tq?
It's not that the engine is producing that much torque, its that that is the total torque put on the input shaft from both the engine and the rest or the rotating mass such as flywheel and clutch and anything else that spins at high speed. Notice that this ridiculous torque figure is only for .2 seconds or so. It's a very short time just to get the car moving. He never stated that you will have 1200ft/lbs all the way down the track or that it is solely engine torque.
 
Last edited:
We are off topic. Very few even jump into what we are talking about and obviously you have more knowledge in the technical part of it. I may not have the right words but the ideas on the topic in the world I live in which is on the dyno and at the track are on relevant to what I was trying to portray.

With that being said, I wish our topic would continue or have started in a different thread. This clutch idea stil baffles me and he never responded with any other information on top of seeming to sell a product without contacting Chris
 
...Basically, this DIY project is a simple, commonly available, hydraulic storm door closer cylinder installed on the clutch pedal. It allows tuning a bit of "slip" into the clutch's initial engagement, damping the peak shock loads as power is being transmitted to the rest of the drivetrain. It's adjustable for exactly where in the pedal travel that it becomes active, and adjustable for rate of release from that point on (it controls slip only during the final part of engagement). The cylinder is hydraulic (not pneumatic like most door closers), with characteristics similar to those of a 90/10 shock, pulling the rod out is easy, only the return stroke of the cylinder is controlled.

Here's a mock-up that shows a somewhat generic underdash install with a clamp-on pedal bracket...

You must be logged in to view this image or video.


With the cylinder installed on a clutch pedal, only the final bit of clutch pedal's release is delayed, not the whole release cycle. The rest of the clutch pedal's travel works without interference. During normal driving you will not even know it is there...no detectable difference in clutch feel. If you are using the clutch pedal during shifts, the cylinder will soften drivetrain shock during gear changes as well.

How does it work? Basically when the clutch pedal is depressed, it pulls the rod out of the cylinder. When released, the clutch pedal comes out unrestricted until the nuts on the cylinder's shaft contact the dash bracket. From that point, the rate of release is controlled by the adjustable orifice inside the cylinder.

How is it adjusted? Adjusting the nuts on the threaded portion of the shaft changes the point in the clutch pedal travel where it's release is delayed. From that point, turning the dash knob changes the speed of the pedal's final bit of release. There are 10 turns of adjustment on the knob. At "0" turns the pedal is delayed very little, barely noticable. At "10" turns, the pedal takes about a minute to return. Typically, the rate of release is usually set somewhere between 2-1/2 and 7 turns.

Here's a parts list for the DIY version...
...Wright Products VH440BL medium duty door closer
...Hilman 881357 bar knob (5/16" x 18)
...5/16 x 18 "all-thread" (comes in 3' lengths)
...5/16" x 18 flange nut
...5/16" x 18 jam nut
...Delrin plastic (for making the dash bracket slide bushing)
above items likely less than $50 at your local hardware store.

You must be logged in to view this image or video.


The VH440BL actually comes with attachment brackets that could be modified for use on a clutch pedal, but I made the bracket that attaches to the clutch pedal from scratch. The cylinder itself was modified by adding a threaded 5/16" coupler to it's shaft...

You must be logged in to view this image or video.


Here's a pic of an install that uses a simple pedal bracket that's welded to the pedal...

You must be logged in to view this image or video.


Here's a pic of a simple in-dash install that uses a couple back-to-back threaded nuts for making the "initial hit" adjustment...

You must be logged in to view this image or video.


Here's an alternative under-dash install...

You must be logged in to view this image or video.


Here's a pic showing the fabrication of a bolt-on dash bracket on a mock-up dash. This car has a removable trim panel, only a couple inches of this slipper assy will be visible in the final install...

You must be logged in to view this image or video.


Adjustment #1- Initial Clutch Hit...Adjusting the "stop nuts" on the threaded portion of the shaft changes the amount of initial clutch "hit", similar to adjusting the base pressure on an adjustable long style clutch. Initially I dial in 7 to 10 turns of "delay" (see adjustment #2), as this basically removes the secondary clutch application and allows focusing on just the initial hit. It is preferred that the clutch initially slips enough that the engine is not pulled down below it's "staged" RPM when the clutch is dumped. For this part of the adjustment we usually just make 30' to 60' test hits, no need to make a full run.

Adjustment #2- Secondary Clutch Lock-up Delay...Secondary lock-up delay is used to delay clutch lockup until the vehicle speed can catch up with engine speed. Turning the cylinder's winged dash knob (on the left in the pic above) basically changes how quickly additional clutch pressure comes in. There are 10 turns of adjustment on the knob. At "0" turns (fully counter-clockwise) our clutch pedal was delayed .181 sec, barely noticable. At "10" turns, the pedal takes about a minute to return.

I like to use enough secondary delay to keep engine's RPM up near it's torque peak until there is enough vehicle speed for clutch lockup.

Most of the clutch slipping that typically comes to mind is from a clutch that is operating beyond it's capacity, usually ends in a total meltdown. What we are doing here is making it possible to easily slip a clutch that has more than enough capacity.
 
thats the cheap diy launch control you can make at homedepot racing. Only problem is it slips on all gear changes unless you modify it to disengage after the first hit.

This is most definately spam unless youve already talked to Chris, even then you need to be a vendor. In before then lock.

http://grannys.tripod.com/clutchtamerdiy.html
 
thats the cheap diy launch control you can make at homedepot racing. Only problem is it slips on all gear changes unless you modify it to disengage after the first hit.

This is most definately spam unless youve already talked to Chris, even then you need to be a vendor. In before then lock.

http://grannys.tripod.com/clutchtamerdiy.html
I don´t think he´s selling anything as it's all DIY. This thread however is not in the ideal section and should be in the DIY or drivetrain tech section. Sounds like he might be testing a product to a new crowd maybe?? Either way it's pretty neat.
Edit: I found 1 ad and it's on his own site
 
Last edited:
If it stays that way im sure it'll be fine since its good information, but it directly ties into his own product.

Also, as i've said in my prevous post. Its engaged full time. You only want slip on the launch. All in all, if you're interested in a launch control device get the Magnus, or make your own, that can be released after the launch.
 
If it stays that way im sure it'll be fine since its good information, but it directly ties into his own product.

Also, as i've said in my prevous post. Its engaged full time. You only want slip on the launch. All in all, if you're interested in a launch control device get the Magnus, or make your own, that can be released after the launch.
He has a make your own version up. And some small slip between shifts other than launch can be good and is less destructive to drive train parts
 
http://www.dsmtuners.com/threads/witch-clutch-for-700-750-hp.490698/

Bastard, and here i though you said you werent using a device like this.

_____________

Slip is non-adjustable from gear changes. I don't know of anyone slipping the clutch in every gear change on a wot run. It might be useful in bastards case with him running a twin now with better streetability because of the slip but i don't want any after the launch.
 
Last edited:
I love how people always assume they have more experience then anyone else.

Ive know about the slipper for years and have used it extensively. In he end, its that fully time activation that made me take i off. Your experience might be different because you're runing a twin where you need full time engagement, but standard organics and pucks, imo, dont require it nor help too much in the realms of their power potential.

No experience, however, with the magnus piece.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top