OK, I did my own calculations based on the data Prober posted (thanks!) just to see if I would find anything different.

I used a different approach - I calculated the MOI of the stock flywheel based on the weight (using the formula for a flat disc), then calucalted to torque required to accelerate it based on Prober's data. Then, multiply by 0.55 to find the torque "freed" by the lighter flywheel.

I found that, during peak acceleration in second gear (5k-6k rpm based on Prober's data) a 55% reduction in MOI in the flywheel would give back about 3.75 ft-lbs of torque. Now, people are saying that the mass of the stock flywheel is concentrated somewhat at the rim; that would make my base MOI calc too low, and hence would result in a conservative estimate for the "free" torque. So perhaps it's actually between 4 & 5 ft-lbs.

For first gear, I estimated the increase in peak torque by simply multiplying by the ratio of first gear to second gear - the result is about 6.2 ft-lbs of extra peak torque. Perhaps 7 or so in the real world.

It's not going to set your '8 on fire, but it's a step in the right direction.

 

i quickly converted things to torque instead of hp and I pretty much agree with your numbers

 

Here is a Gtech comparision Polak and I made of his unmodified RX8 this weekend and mine from before. You can see the similarities and where the curve with the flywheel starts to increase above the baseline. The weights on the stats are different due to gas, both of us in his car, all the crap he had in the trunk and his nav system. I also posted this on the other flywheel discussion.

Enjoy...

 

For those interested, here are my LWF (Light Weight Flywheel) notes and conclusions for the Rx8. I had the car dyno'ed, one run each in 2nd, 3rd and 4th gear before and after the install.
First, with regard to flywheel weight(s), according to the US Postal Service's mail scales, the stock OEM flywheel w/ clutch bolts (at least the one removed from my car) weighed in at .... drum roll, 17 Lbs, 1.4 oz.
FWIW, Racing Beat's website lists the stock flywheel weight at 16.8 lbs, slightly less than my Postal scale results. Various other forum posting have suggested a "stock" flywheel weight as high as 21 lbs, others have posted the same approximate 17 lb weigh my trip to the postal scales came up with. But hey, I was missing one lock washer that fell out of the replaced parts box before I got to the scales :-)
Second, to shorten the learning curve fot those not yet aware, the manual transmission Rx8 stock OEM flywheel doesn't use or have a separate counterweight. It's apparently cast as part of the stock flywheel. All the aftermarket LWF solutions, i.e. MazdaSpeeds version, RacingBeat and SR Motorsports (are the others) that I'm aware of, use a separate counterweight. AFAIK, the counterweight in each case is the same (?) unless the suppliers do some kind of aftermarket balancing or machining to the part to make it specific to their LWF.
As others have pointed out, the OEM flywheel's 17 lbs should be looked at as a "gross" weight figure. For flywheel purposes, it is the way that 17 lbs is distributed across the diameter of the flywheel that is probably of as great importance as the flywheels "simple" gross weight. That said, a significant portion of all the LWF offerings, i.e. approximately 3.66 lbs based on my trip to the local Post Office scale, is in the separate counterweight. Unfortuantely, as far as I could determine, no aftermarket LWF purveyor either has, or if they do have, it isn't commonly available, a "moment of inertia" measurement for their flywheel, or have such a number or measurement for the stock flywheel.
FWIW, in the RacingBeat LWF case, the counterweight comes in a Mazda parts box, as PN# N3Z2-11-52X, with an additional RacingBeat lable affixed to the Mazda parts box identifying it as RacingBeat's "Part 11458, 09/16/04 Rear Balance WT '04 RX-8"
I suppose it's possible RacingBeat could have taken a stock counterweight, and tweaked, or otherwise removed weight to fit it's flywheel. I don't know, but I doubt anything like that was done. I didn't inspect the counterweight that closely for signs of aftermarket machining or balancing efforts before the shop slapped it in the car.
So, unless someone knows otherewise, for the moment, I'll go with the assumption that the RacingBeat, MazdaSpeed and SR Motorsports are all working with the same 3 lbs 10.6 oz counterweight.
Third, I haven't any direct "hands on" experience vis-a-vis either the MazdaSpeed LWF, or SR's product. The information I've been able to unearth remains a bit conflicted. Others posts on this forum have suggested the MazdaSpeed flywheel weighs @ 14.5 lbs, inclusive of the counterweight. The Rossenthal Mazda site posts a 15.5 lbs figure for the MazdaSpeed parts (I'm assuming that has to be both the flywheel and counterweight together.) There's a photo of the MazdaSpeed version on the Rosethal site. The MazdaSpeed version differs from the RacingBeat and SR LWF offerings in that it is still a one (? I think) piece, steel flywheel. Weight has been removed from the perimeter of the wheel (moment of interia) by the machining of a serries of holes. The other two flywheels are multipiece units made of aluminum disks to which a steel friction face plate and out steel ring gear have been attached.
I don't know how SR Motorsports would break out weight figures for the various parts their kit uses. SR's website seems to indicate they offer two levels of "light" flywheels, an 11 pound version and a "super" light, I'm not sure, but this may be SR's 9.75 lbs offering. SR's site is critical of others, un-named, "8 lb" flywheel offerings. I'm guessing that the #9.75 and #11 lb figures from SR's site would be exclusive of the counterweight, but I don't know that for a fact.
Fourth, courtsey of the US Postal Service scales, I can provide the following weight break down by component parts for the RacingBeat kit:
RacingBeat Flywheel, Alu. w/ steel friction surface 7 lbs 15.8oz
Counterweight, Mazda #N3Z2-11-52X 3 lbs 10.6 oz (including flimsy Box)
Misc. Bolts & washers 0 lbs 8.7 oz (inlcuding plastic bag)
TOTAL for RacingBeat's LWF @ 12 lbs 4 oz ..... +/- an ounce or two.
Last, I can share with you a set of "before" and "after" 2nd, 3rd and 4th gear dyno pulls (if I can just figure out how to attach these graphs) for the Racing Beat LWF vs. a stock flywheel. The dyno information is taken from a '04 Rx8 with approximately 11,500 miles on the odo, running an "M" flash ecu
Peak Stock Peak LWF
SAE hp / torque SAE Hp / torque
2nd 166.14 / 118.12 173.51 / 127.37 @ 5,950 rpm
3rd 173.91 / 126.95 176.91 / 132.86 @ 5,880 rpm
4th 179.81 / 130.68 178.76 / 135.35 @ 5,735 rpm
Hope this helps someone with their choices ~ attachment up loads, there should be 4 (if I figure out how to do this properly.)

 

The MS one weighs 6.92 Kilos, 1 Kilo = 2.205 lbs ~= 15.2586 lbs. Racing Beats reports the weight of the stock flywheel is 16.8 lbs. So MS "light weight flywheel only weighs 1.54lbs less then Stock? That seems as if it is hardly worth $620.00? No wonder some people complain that a light weight flywheel is not a rewarding HP mod.

The Racing Beat light weight flywheel is 12lbs ~ 4.8Lbs lighter.....$455...not sure if that is enough of a weight reduction either for me.

Lastly, I looked at the SR Motorsport one: It weighs 9.75lbs ~7.05lb (using the guidance that the OEM one is 16.8lbs) reduction off of the crank and it is for sale at $429.00 + 18 hardware install kit + $139 for counter weight. = $586 Philodox shared: "The SR Motorsports flywheel weighed in at 12.5lbs (flywheel, counterweight, and bolts"

Racing Beat has the lightest of these three contenders.

Full thread with LFW weights and Racing Beat's input.

https://www.rx8club.com/series-i-aftermarket-performance-modifications-23/light-weight-flywheels-points-interest-40514/

SR's measurement of the stock fly appears to be less then correct.

 

"... So MS "light weight flywheel only weighs 1.54lbs less then Stock? That seems as if it is hardly worth $620.00? ...."

I'd agree as a gross number, 1.5# lbs doesn't seem like much. But then again it's the redistribution of the remaining mass that's also important. It would be nice if some one else gets a chance to try to pull off a set of "before" and "after" dyno runs for somekind of comparison.

Total cost for my LWF install, including the parts, came to @ $750. That's what the dealership was quoting for labor alone to put the MS unit in.

I see from your thread at https://www.rx8club.com/showthread.php?t=40514, you recently bought a RacingBeat flywheel. I'll be interested hear what your SOP (seat of the pants) assessment of it is after you have a chance to drive it.

For me, at altitude (i.e. for instance 187 "SAE" bhp actually works out to about 145 real, actual, unadjsuted ponies at the wheels at 5,280 ft), every little bit helps... IMO.

 

hey MLX8. I just wanted to say nice post. Well put together with good supporting data.

 

Yes, distribution of mass is critical, and that's why the simple "flat disc" calculations that I and others have engaged in are flawed.

9 ft-lbs in second, and a nicely offset torque curve everywhere in the rev range - that seems pretty good to me. I'd be surprised if there is any other $620 mod out there that would have an equivalent effect.

Also interesting that you got more peak HP in hiugher gears. I wonder if that is a dyno effect or a result of the different ECU maps in different gears.

Thanks for the psot, MLX8.

 

GeorgH Questions "Also interesting that you got more peak HP in hiugher gears. I wonder if that is a dyno effect or a result of the different ECU maps in different gears...."

The dyno folks don't often (if ever) regularly do pulls in 2nd gear. They too remarked on the difference. Their innitial speculation was that it may have to do with possible losses as a result of the gearing selected. If that were the case, I would then guess/conclude that if one were to do a pull in 1st gear, one might expect a porportionately greater drive line loss? I guess I'll have to ask them if they want to voulunteer a dyno session with a 1st, 2nd and 3rd gear runs to test that hypothesis out.

Of additional concern to me, an the reason I added two 4th gear dyno charts, is the "ragged" nature of the power & torque curves that start to show up as the gear ratio moves higher. I don't know if this is something that is engine related, or the dyno induced. Certainly in the higher gears the dyno is spiniing a whole lot faster, than with 2nd or 3rd gear pulls. I didn't include or post the bhp curves on the charts just to keep them clean, simple and easy to read, but, for example, the "jaggies" in 4th gear actually have the car making slightly less "peak" bhp with the LWF (178.76 SAE bhp at @ 8,263 rpm with "smoothing" set to 0 vs. 175.77 bhp with smoothing set a t 5), than with the stock flywheel (176.85 at @ 8,455rpm with smoothing set to max, 5 ~ 179.81 @ 8,445 with 0 smoothing applied)

 

Yes, the curves get more ragged in the higher gears. One possible explanation is the sampling frequency on the dynos. In second gear, the rollers are accelerating very quickly, and, depending on the sampling frequency, the dyno may not catch all the local dips & peaks. On the other hand, in fourth gear, the rollers are not accelerating as quickly – the pull takes more time – so the dyno sensors pick up all the minutia of the torque curve.

Just a theory. I wonder what the sampling frequency of the dyno is?

 

mlx8,

Thanks for the data. I think the data clear shows that the effects of the RB flywheel are more pronounced in the lower gears. A 9 ft-lb increase in 2nd gear seems a little high but reasonable.

Jaguar,

It is not the weight that is important. It is the the moment of intertia that is important. The moment is determined by where the weight (or more accurately mass) is located.

The Mazdaspeed flywheel acts as basically the stock flywheel with holes drilled along the outer rim. Doesn't sound like much but it is. To show how important this concept is:

On a 12 inch flywheel (6 inch radius), removing 1 lb along the outer rim where the radius is 6 inches is equal removing 4 lbs from the wheel where the radius is 3 inches. Also no amount of weight reduction near the center of the wheel would ever equal 1 lb of reduction on the outside. (for this reason the weight of the counterweight is essentially meaningless, it will have almost no effect on interia.)

So while the Mazda speed flywheel which is about 13 lbs is only 4.5 lbs lighter than stock, most of that weight loss is along the outer edge. The effective drop in moment is probably close to half and that is the only thing that matters.

Also the amount of interia also has this square phenomena. At 9000 RPM the interia in the flywheel is 4 times that compared to when it is at 4500 RPM.

My guess is that since the transmission of the RX-8 was barrowed/modified from the Miata, the flywheel was borrowed as well (just a guess). I don't know what redline on the Miata is, but if were say 7000RPM, then the interia of the heavy stock flywheel would only be 60% of that compared to the RX-8 at 9000RPM (i.e. (7000*7000/(9000*9000) = .604 )

-Mr. Wigggles

Ps. They really need to sell "lightweight flywheels" as "low-interia flywheels". Weight is very much secondary.

Pps. I think this is the last time I try to explain this topic. My fingers are getting tired.

 

hmm... looks like the laws of physics still work pretty well


On another note, the reason for the increase of hp with higher gears is the same reason that the flywheel free's up more hp in lower gears - most drivetrain related losses are effected by the rate of angular acceleration... aka you rev much quicker from 3-9k rpms in 2nd gear then 3-9k rpms in 4th gear, therefore any inertial losses in the drivetrain will be bigger in 2nd gear then 4th gear. PLus gear frictional losses are slightly lower when there is less of a gear ratio.

Actually, the best dyno results of all should be in 5th gear because not only do you lower inertial drivetrain losses, but I believe 5th gear in our tranny is a straight through design which eliminates the frictional losses of transferring power to different shafts. Thats why most dyno's are (and should) always done in the 1:1 ratio gear, but some people get nervous when their car is doing 140mph sitting still.

 

hey MrWigggles I was listening....learned a lot of flywheel and benifts 9real tha tis) tha tcan be had from them....thanks fo rthe informative post. I think i'm going to settle on the Mazdaspeed flywheel. Thanks for all your effort.

 

I just got my ACT prolite flywheel and clutch installed. first impressions are not impressive. The revs are a little quicker but no noticable horse power increase. the clutch grabs sooner and harder...which will be good when i get supercharged. here are the weight measurments:
stock flywheel: 17.4 lbs
ACT prolite : 13.1 lbs

 

Either you got the 'streelite' or are quoting the Prolite with the 3.x lb counter weight.

Laters!
Edgardo