Clutch Upgrade: Which one?
01-02-2012, 08:28 PM
#31
Rest in Peace
Rennlist Member
Rennlist Member
Join Date: May 2006
Location: Bird lover in Sharpsburg
Posts: 9,903
Likes: 0
Received 2 Likes
on
2 Posts
What do you see for the future for manual transmissions in the 928 once we get over that breaks everything in the transmission threshold?
Do you have plans for beefing up the 5 speeds, or is this just better parts for what we have?
Do you have plans for beefing up the 5 speeds, or is this just better parts for what we have?
I'm not giving up on the torque of the 928 engine....that would be a huge mistake, in my mind. I don't think that the weight of the rotating assembly determines the amount of torque the engine can make....so there is very little downside to reducing the weight of the reciprocating pieces, for me. I've always engineered my pieces to be able to spin at 7,500 rpms, even at their current weights. Correctly centered connecting rods, custom oiling in the crankshaft, removal of the two "extra" counterweights that are "completely useless (direct quote from Whitey at Moldex) below 7,500 rpms", current camshaft and valve train technology, all combine to make my engines function fine at 7,500.
However, once I reached over 500 ft lbs of torque at 5,000 rpms (and maintained that torque all the way to 6,000 rpms) the weaknesses of the transmissions became very apparent. Mark Anderson has broken virtually every single gear inside the 5 speeds.....and that is with an "older technology" engine. Making the automatic transmission work, behind all this torque, took me months, all by itself.
And although I am bringing several stronger transmission pieces into the equation, I think that moving the power range up perhaps 800-1,000 rpms will make the life of the drivetrain much better, as well as increasing horsepower, while not killing the great torque these engines make.
However, once I reached over 500 ft lbs of torque at 5,000 rpms (and maintained that torque all the way to 6,000 rpms) the weaknesses of the transmissions became very apparent. Mark Anderson has broken virtually every single gear inside the 5 speeds.....and that is with an "older technology" engine. Making the automatic transmission work, behind all this torque, took me months, all by itself.
And although I am bringing several stronger transmission pieces into the equation, I think that moving the power range up perhaps 800-1,000 rpms will make the life of the drivetrain much better, as well as increasing horsepower, while not killing the great torque these engines make.
01-03-2012, 02:04 AM
#32
Rennlist Member
01-03-2012, 02:29 AM
#33
Rennlist Member
No doubt your thought might be a consideration in some applications, but keep in mind that I'm not building a 22 lb. crankshaft.....
I just don't think it needs to be 63 pounds, for the relatively small stroke of our engines.
The stock harmonic balancer isn't very good technology. It isn't even very good bad technology. Nothing that fits that loose on the crankshaft can properly absorb harmonics....much less a rubber style damper. To say that there is a lot of harmonics that are "lost" between tthe stock crankshaft and the stock damper, is a severe understatement.
And yet, how many broken 928 crankshafts have you seen or even heard about?
I'd say that the lack of broken cranks, even with the terrible stock damper, would pretty much prove that we are no where near the "lower" weight limit of these crankshafts. Keep in mind that the 928 crankshaft isn't even built out of a "high end" piece of steel and the lack of broken cranks becomes even more of a consideration.
As far as what reduced mass does for the engine....reciprocating mass becomes very important as the rpms increase....especially with the terrible problems we have with the strength of the stock block. Turn the current technology at 7,200 rpms for extended periods of time and the block is going to be the weak link...that is a given, for me.
Sure, I've been working on a billet block, but perhaps a slightly lighter crank, rods, flywheel, and flywheel will solve the problem for the guys that are not going to be able to afford a $15,000+ block.
I just don't think it needs to be 63 pounds, for the relatively small stroke of our engines.
The stock harmonic balancer isn't very good technology. It isn't even very good bad technology. Nothing that fits that loose on the crankshaft can properly absorb harmonics....much less a rubber style damper. To say that there is a lot of harmonics that are "lost" between tthe stock crankshaft and the stock damper, is a severe understatement.
And yet, how many broken 928 crankshafts have you seen or even heard about?
I'd say that the lack of broken cranks, even with the terrible stock damper, would pretty much prove that we are no where near the "lower" weight limit of these crankshafts. Keep in mind that the 928 crankshaft isn't even built out of a "high end" piece of steel and the lack of broken cranks becomes even more of a consideration.
As far as what reduced mass does for the engine....reciprocating mass becomes very important as the rpms increase....especially with the terrible problems we have with the strength of the stock block. Turn the current technology at 7,200 rpms for extended periods of time and the block is going to be the weak link...that is a given, for me.
Sure, I've been working on a billet block, but perhaps a slightly lighter crank, rods, flywheel, and flywheel will solve the problem for the guys that are not going to be able to afford a $15,000+ block.
Actually, i I think we got that you are on the right track with your efforts, BUT, the rubber coupled damper is proven to deal with odd harmoics due to the fact that it ITSELF, has a natural frequency, that can counteract the harmoics of the engine better, than if it was solidly coupled. thats how it works, the rubber is there by design. there are also viscous dampers, and magnetic dampers that also work on a similar principle. Now, is it the best designed rubber coupled damper? probably not if they come apart as Fan has seen. But, it can do a good or even great job.
certainly we have not seen any broken cranks like we see often with the BMW guys having broken cranks at the rear end where the flywheels attach. And as you said, reducing the stress on the engine block for high rpm use is important. so, a ligher crank is where the efforts should be made to help with the stresses of high rpm operation.
you know im agreeing with most all you are pointing too, and certainly, just removing 3.6lbs off the flywheel is a rounding error in any performance metrix.
even throttle blips, yet some companys, individuals swear by the gains without a clue of the actual changes that will occure , and that those changes are so easy to calculate. it is those same calculations , which would be used to set up and calibrate a dyno , so the results of thoe calculations are even more accurate than if you tested them on the dyno.
01-03-2012, 02:56 AM
#34
Nordschleife Master
Running the lightweight DD flywheel (made by SPEC BTW), I did not notice much of a difference in running at all. Or "blipability", I found more of a difference with the changing of camshafts than I did from the lightweight flywheel.......
01-03-2012, 11:17 AM
#35
Developer
The CW on a lightened flywheel is that it improves throttle response. It does not make more torque, and it does provide an increase in BHP.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
01-03-2012, 12:05 PM
#36
Developer
What do you see for the future for manual transmissions in the 928 once we get over that breaks everything in the transmission threshold?
Do you have plans for beefing up the 5 speeds, or is this just better parts for what we have?
Do you have plans for beefing up the 5 speeds, or is this just better parts for what we have?
The hinderance here is the cost of development vs the probable number sold. This is an expensive R&D project (read R$D) and I cant quite justify how many 6-speeds I'll sell to 928 owners.
Especially given that the 928 transaxle does survive depending how its driven. I think Kibort is right - and he does drive with an emphasis on smoothness and his servioce intervals are lenghtened as a result. I like to think I emphasize smoothness also, and so far have only broken the spider gears in the diff on the stock box. Now, at my new power levels, I expect to start breaking more things like Mark A has. I have been carrying a fully built G28.13 transaxle with me for the last two seasons, haven't needed it yet... but I do expect that to change.
01-03-2012, 12:08 PM
#37
Rennlist Member
Thread Starter
I see a kit that will drop a modern 6-speed into the 928. Working on that here. I dont see "beefed up internals" getting us where we need to be.
The hinderance here is the cost of development vs the probable number sold. This is an expensive R&D project (read R$D) and I cant quite justify how many 6-speeds I'll sell to 928 owners.
Especially given that the 928 transaxle does survive depending how its driven. I think Kibort is right - and he does drive with an emphasis on smoothness and his servioce intervals are lenghtened as a result. I like to think I emphasize smoothness also, and so far have only broken the spider gears in the diff on the stock box. Now, at my new power levels, I expect to start breaking more things like Mark A has. I have been carrying a fully built G28.13 transaxle with me for the last two seasons, haven't needed it yet... but I do expect that to change.
The hinderance here is the cost of development vs the probable number sold. This is an expensive R&D project (read R$D) and I cant quite justify how many 6-speeds I'll sell to 928 owners.
Especially given that the 928 transaxle does survive depending how its driven. I think Kibort is right - and he does drive with an emphasis on smoothness and his servioce intervals are lenghtened as a result. I like to think I emphasize smoothness also, and so far have only broken the spider gears in the diff on the stock box. Now, at my new power levels, I expect to start breaking more things like Mark A has. I have been carrying a fully built G28.13 transaxle with me for the last two seasons, haven't needed it yet... but I do expect that to change.
And here it is: https://rennlist.com/forums/928-foru...ost-157-a.html
01-03-2012, 02:38 PM
#38
Rennlist
Basic Site Sponsor
Basic Site Sponsor
The CW on a lightened flywheel is that it improves throttle response. It does not make more torque, and it does provide an increase in BHP.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
Could you please explain this a bit further?
Isn't BHP the horsepower that I measure at the flywheel on an engine dyno, or is there a distinction?
I understand what you are saying about the engine reving quicker, when it is out of gear, with lighter components.
What I wonder is if there is anyone that can actually tell the difference if there is less than 3% of the total weight changed?
I'm pretty sure that I can't possibly detect this difference.
Let me explain why I'm pretty sure I can't tell the difference:
The change in weight between a stock single disc clutch and a dual disc clutch is certainly a larger percentage than the <3% of an aluminum flywheel, yet there isn't a chance that I (or anyone I know) can tell, by listening, what style clutch is in the car. I actually went out and tried this, with a group of friends. None of us could tell the difference by listening.
I simply can't walk by and say:
"Wow, listen to that thing rev. That GTS has a dual disc clutch in it."
Do you think that you can tell the difference between styles of clutches, in the same vehicle?
Or is a lightened flywheel any different, because it is located in a different place than the clutch?
I'd like to wish you a Happy New Year, Carl.
My New Year's resolution was to be able to freely exchange ideas back and forth with you, without being hostile. Should be much better for this Forum.
I think that this thread is a great start!
I'm really looking forward to more of these exchanges!
__________________
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
01-03-2012, 06:45 PM
#39
Rennlist Member
The CW on a lightened flywheel is that it improves throttle response. It does not make more torque, and it does provide an increase in BHP.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
Explanations: the increase is throttle response is because you have lowered the rotational inertia of the engine. Road racers like the increase in throttle response it provides as the engine can be matched to the gear during shifts in less time.
A lightened flywheel has no effect on the torque output of the engine. It appears to, but that is just because it is storing less kinetic energy at the start. Starts are something few racers give a rip about, there is only 1 start per road race. Its an easy trade-off.
There will be an increase in BHP as a result of lowering the inertial mass. BHP (Brake Horse Power) is the net remaining power available for use after the engine overcomes its own internal losses. Lowering the weight of the rotating assembly lessens the internal losses, and BHP goes up as a result.
Here is a good article that attempts to explain this further:
http://www.uucmotorwerks.com/flywhee...heel_works.htm
Colin: yes, our flywheels are made by SPEC. I designed them with SPEC, and we did the 928 fitment and testing for SPEC. After the early (no timing gear) flywheel could not be used on the LH-Jet equipped 16v motors, it was us that made the re-design and testing for SPEC on the improved version. And so on with the 32v single disk versions. I continue to have my flywheels made for us by SPEC because they do such a good job. Just wanted you to know.
that quacky BMW written paper is as full of it as it gets. No Carl, you dont get any gains by a lighter flywheel (measureable anyway) its calculable! its nothing to write about. you give me the change in rpm in time, and Ill tell you the hp it cost or saved. ( knowing the weight and size of the flywyheel) really simple stuff. dont over sell it. it makes you look bad
01-03-2012, 07:39 PM
#40
Rennlist Member
Flywheel stores energy, a heavy one stores more kinetic energy. A heavy reciprocating assembly takes more time to respond to RPM changes. It has nothing to do with one having more power. The difference is if you are accelerating in gear the light one will pull into higher RPM quicker and the power comes from the higher RPM output.
An example 5.0 928 heavy and light reciprocating assembly engine, 2000rpm...200 HP, 3000 rpm...300 HP. Heavy one takes 3 sec to pull 2000 to 3000 RPM in gear. Light one takes 2 sec to pull into 3000 rpm and now to have a second gain to get into the extra 100 HP. Do it many times on a race track and you have a big gain. On the street it is pointless.
I ride a new yamaha r1...1000cc with 175 HP & 14K RPM rev limit. At 2K RPM it is mild power, snap the throttle and 2 seconds later it is at 14 K RPM and ready to throw you off.
I hope this helps.
An example 5.0 928 heavy and light reciprocating assembly engine, 2000rpm...200 HP, 3000 rpm...300 HP. Heavy one takes 3 sec to pull 2000 to 3000 RPM in gear. Light one takes 2 sec to pull into 3000 rpm and now to have a second gain to get into the extra 100 HP. Do it many times on a race track and you have a big gain. On the street it is pointless.
I ride a new yamaha r1...1000cc with 175 HP & 14K RPM rev limit. At 2K RPM it is mild power, snap the throttle and 2 seconds later it is at 14 K RPM and ready to throw you off.
I hope this helps.
01-03-2012, 08:01 PM
#41
Rennlist Member
Flywheel stores energy, a heavy one stores more kinetic energy. A heavy reciprocating assembly takes more time to respond to RPM changes. It has nothing to do with one having more power. The difference is if you are accelerating in gear the light one will pull into higher RPM quicker and the power comes from the higher RPM output.
An example 5.0 928 heavy and light reciprocating assembly engine, 2000rpm...200 HP, 3000 rpm...300 HP. Heavy one takes 3 sec to pull 2000 to 3000 RPM in gear. Light one takes 2 sec to pull into 3000 rpm and now to have a second gain to get into the extra 100 HP. Do it many times on a race track and you have a big gain. On the street it is pointless.
I ride a new yamaha r1...1000cc with 175 HP & 14K RPM rev limit. At 2K RPM it is mild power, snap the throttle and 2 seconds later it is at 14 K RPM and ready to throw you off.
I hope this helps.
An example 5.0 928 heavy and light reciprocating assembly engine, 2000rpm...200 HP, 3000 rpm...300 HP. Heavy one takes 3 sec to pull 2000 to 3000 RPM in gear. Light one takes 2 sec to pull into 3000 rpm and now to have a second gain to get into the extra 100 HP. Do it many times on a race track and you have a big gain. On the street it is pointless.
I ride a new yamaha r1...1000cc with 175 HP & 14K RPM rev limit. At 2K RPM it is mild power, snap the throttle and 2 seconds later it is at 14 K RPM and ready to throw you off.
I hope this helps.
in gear, the times for acceleratoin is so slow (relatively speaking) obviosly, the ligher flywheel will have more of an effect in 1st than in 4th, but my point, partially in just, was to point that the shift from redline, and a speed shift , could transfer the extre KE of the 3lb mass going from 6000rpm to 4500rpm, to get a few extra oz-in of torque kick.
. its all rounding errors. as I said and as greg was pointing to , there will be no differnece measurable with a 3.5lb lighter flywheel. you give me the change in RPM time, and ill give you a HP equiv. its less than 1 - 2hp overall average.it it made a difference, i would have one on my race car. however, by going from a 50lb clutch to a 25 lb clutch (going dual disc) it made shifting easier, near perfect in my mind, and saved 25lbs of needless weight. 25lbs is significant, yet still not measureable on the dyno, (in 3rd or 4th anyway) and is a heck of a lot more than saving 3.5lbs.
Now, you make no sence of the light comparison vs the heavy 928 5.0 rotating assembly. the effects of that lighter mass is very slight. its the rate of speed change, or the rate of kinetic energy change that is really the def of HP.
yes, it can save hp, but you tell me the weight and dim's and the time you are acceleratting and i can give you a hp savings or cost. 3lbs off the flwheel, except for 1st gear or reving the engine in idle., is a rounding error at best.
01-03-2012, 10:06 PM
#42
Rennlist Member
I remember someone interviewing Jim Clark about racing in the old Tran-Am series. Seems they would build the transmission out of Magnesium and cover it with an Iron Foil so a magnet would stick to it to pass inspection. Said that's also why those cars would burst into flames when they wrecked. Said they would even strip the body, dip it in acid to eat away the thickness of the metal, then pull it out and repaint it. Said they got caught when an inspector set his empty styrofoam coffee cup on the roof and it dented it.
Juat a matter of the lengths you want to go to remove weight from the car.
Juat a matter of the lengths you want to go to remove weight from the car.
01-03-2012, 10:46 PM
#43
Rennlist
Basic Site Sponsor
Basic Site Sponsor
I remember someone interviewing Jim Clark about racing in the old Tran-Am series. Seems they would build the transmission out of Magnesium and cover it with an Iron Foil so a magnet would stick to it to pass inspection. Said that's also why those cars would burst into flames when they wrecked. Said they would even strip the body, dip it in acid to eat away the thickness of the metal, then pull it out and repaint it. Said they got caught when an inspector set his empty styrofoam coffee cup on the roof and it dented it.
Juat a matter of the lengths you want to go to remove weight from the car.
Juat a matter of the lengths you want to go to remove weight from the car.
$550.00 to remove 3.5 lbs is about $160.00 per pound.
Not how much I want to spend to remove weight ($16,000 for every 100 lbs.), unless there is a significant performance improvement to be had.
Hopefully, Carl will clarify shortly.
01-03-2012, 11:10 PM
#44
Nordschleife Master
We had a thread here a few months ago about a Corvette transmission installed in a 928. I unfortunately stopped following it, but I thought I recalled him getting it to work. I'll see if I can find it...
And here it is: https://rennlist.com/forums/928-foru...ost-157-a.html
And here it is: https://rennlist.com/forums/928-foru...ost-157-a.html
01-04-2012, 01:24 AM
#45
Rennlist Member
Ok, Here is the equation for those that dont trust the "I feel faster", "significantly faster", folks.
M=I x ( gear x final/2)^2
what happens is that for 7 lbs, its like you are losing 70lbs in 1st gear
30lbs in 2nd, 16lbs 3rd and 10lbs 4th.
If you go by the old rule of thumb (also referenced in that BMW UC article), of each 10lbs equals 1hp, you can see that even at 7lbs, the HP saved or lost is near as high as 7hp in 1st gear, but really about 3hp in the usuable 2nd, 1.6hp in 3rd and near 1hp in 4th. at 3.5lbs, the weight in question, you can see it would be significantly less.
so, if you lose 3.5lbs off your flywheel, an you are driving around in 3rd gear and are able to notice under 1hp, god bless you for your sensitivity!
carl, if you really want to make some cash and do everyone a significant upgrade. make double disc clutch kits out of stock parts . 20+ lbs savings is very significant, especially if spinning! you are not going to notice the acceleration difference though, heck, i didnt. (probably 3hp on the dyno in 4th gear.)
M=I x ( gear x final/2)^2
what happens is that for 7 lbs, its like you are losing 70lbs in 1st gear
30lbs in 2nd, 16lbs 3rd and 10lbs 4th.
If you go by the old rule of thumb (also referenced in that BMW UC article), of each 10lbs equals 1hp, you can see that even at 7lbs, the HP saved or lost is near as high as 7hp in 1st gear, but really about 3hp in the usuable 2nd, 1.6hp in 3rd and near 1hp in 4th. at 3.5lbs, the weight in question, you can see it would be significantly less.
so, if you lose 3.5lbs off your flywheel, an you are driving around in 3rd gear and are able to notice under 1hp, god bless you for your sensitivity!
carl, if you really want to make some cash and do everyone a significant upgrade. make double disc clutch kits out of stock parts . 20+ lbs savings is very significant, especially if spinning! you are not going to notice the acceleration difference though, heck, i didnt.
(probably 3hp on the dyno in 4th gear.)