MadDog

Everyone has heard the argument that perhaps the 8 has horrendous drivetrain losses and that could possibly explain the low HP at the wheels compared to the claimed HP at the crank. Well, here is some anecdotal evidence to support the high drivetrain loss theory. The basis for this claim is a comparison of my RX-8 with my Audi A4 Quattro. For those of you unfamiliar, the Quattro designates all wheel drive.

I live near a road that goes up and down hill at an almost constant grade for about 6 miles. My GF lives at the top of said hill and I live about 3 miles down it from her. On my trip home I always bring the car up to 60mph, take it out of gear and let it coast down.

Without exception, the A4 will accelerate to about 65mph and then decelerate steadily to about 54mph when its time to turn on to my street. The RX-8 will always decelerate steadily to about 34mph at the same distance!!

Now, I know this is not a quantitative analysis, but, the RX-8 should have less drag than the A4. Plus, the A4 is all wheel drive. The A4 will have a little more momentum with which to battle the drag due to the additional mass. I think its in the 3500# neighborhood. All things considered, I would guess that they should at least perform similarly in my little test. But 20mph slower? This isn't even considering the tranny since its out of gear!

maxxdamigz

iTrader: (1)

While interesting, I don't think enough information is there to even signify that the drivetrain losses of an Rx-8 are what is making the difference. It could be that the tire pressure, suspension tuning, and road conditions simply dissappate the 8's lower kinetic energy. It could also be poorly lubed wheel bearings. From the dyno numbers I see and the current bhp# on record as 232, Mazda wants us to think the 8 suffers from roughly a 20-25% drivetrain loss on a dyno. What is the guesstimate of the A4's drivetrain loss?

I think the rx-8 is perfectly capable of 232 bhp, but maybe doesn't usually deliver it under normal conditions on the stock tune. Stock dynos for the 8 that I've seen have had much higher deviations than I would expect.

SmokeyTheBalrog

The weight is definitely helping the Audi there.

A weight difference of 500 lbs is quite a bit. Once two cars are in motion, everything else being similar, generally the heavier car with keep rolling long and faster.

But definitely check tire pressure. I know I can coast for ages (even up hill).

Also, which car has wider tires?

rotarygod

Speedsource Racing uses a completely stock Mazda drivetrain. At Sevenstock we asked them how much they lose through it. They have measured it and yes they told us. It's around 12%. Their engine dynos right around 260 at the crank and on a chassis dyno it's around 230 rwhp.

r0tor

iTrader: (1)

Wheel inertia, brake rotor inertia, bearings, lubricants, and flywheel inertia factor into drivetrain losses on a dyno...

they are not using the stock wheels/tires, not using the stock brake rotors anymore, and I'd be suprised if they have a stock flywheel...


I'd wager the thread starter may have a dragging brake

rotarygod

OK worst case scenario is that the stock car loses 15% and that's really pushing it. That's it. It's pretty efficient. They use the stock brakes with nothing more than different pads. Yes the flywheel is different and the wheels and tires are different. That's pretty much it. The car definitely loses nowhere near 20%.

Kewl

MadDog does this mean you are out driving the 8 again?

Not sure how much impact some of the mentioned factors might have on drivetrain loss...like tire size, pressure, brakes, bearings, diff gears, etc....my 8 is easy to push on level ground. It would seem to me that if those kinds of things are dragging down the car that much I would find it very hard to push. Like MadDog my 8 tends to slow way down on hills (downward) compared to my other vehicles. Yes the others weight more, but the delta between vehicles' speed is dramatic...IMO.

BTW T. I took the 8 to the track last week for two hours....w/instructor. We had the track to ourselves so no one else to worry about and had a blast. This track is much easier on tires. Had I known you were driving the 8 again I'd called. I will call next time we go, which should be soon.

MadDog

I have a MS lightened flywheel, AND the MS rims, which are a couple of pounds lighter than stock. That would mean a little less loss than a stock 8. My tires are inflated to 32 psi each. Now, I do have wider / stickier tires on the 8 than the A4. That could certianly mean more frictional losses for the 8's tires than the A4, but I don't know if this could explain the difference.

As far as the extra mass of the A4 goes, I dug through my fluids book and if drag coefficient and presented area were the same for the A4 and the 8, then the mass difference would only account for about a 8% reduction in velocity (according to my simplifed model where terminal velocity is proportional to sqrt(weight)). But, the 8 has a lower drag coefficient and a smaller presented area, so this probably more than offsets the mass effects.

Maybe it is losses due to the tire friction. I strongly doubt its a dragging brake, although I hadn't considered that, since the car doesn't pull at all - I can let go of the wheel and let it go straight down this hill.

Here's another thought. Those chassis dynos that attempt to measure HP at the flywheel make some measure of drivetrain losses - probably through a spindown type test. The few charts I've seen from these all get pretty close to the 238 crank HP advertised. Yet, the whp is closer to 170-180, right? This would also seem to support the large loss theory.

TeamRX8

iTrader: (25)

no A/C, serious pulley drive reduction, emission-less headers, catless exhaust, 5.12:1 rear gear, 2650# weight, Motec stand-alone, etc.

this all changes next year though ...

MadDog

Kewl, I finally got through emissions and I'm back terrorizing the streets again! Let me know when you're going tracking next time.

Why don't you do the same test on the same stretch of road? You could also compare to your other vehicles? (I can see this turning into a giant soapbox derby down that hill!)

SoFL_RX8

Wouldnt that have more to do with the relative gearing between the cars than the drivetrain loss?

MadDog

huh? The cars are out of gear...

maxxdamigz

iTrader: (1)

This test doesn't purely measure drivetrain losses. Losses of speed for the car are due to a host of factors which include drivetrain frictional losses. Inertial drivetrain losses actually act against the slow down. That is the purpose of a fly wheel. When you get the A4 rolling, you have sped up the car and drivetrain up to X speed. An object in motion stays in motion. Thus, the drivetrain (minus frictional losses) wants to continue spinning. Every rotating member of that drivetrain acts as a flywheel and PREVENTS the car slowing down. Not only do frictional losses have to stop the 3500 lbs of mass going down the road, but they have to slow down all the spinning components - in this case everything from the second half of the transmission to the wheels. The RX-8, spinning signficantly less mass, has less inertia to keep it moving. When you put foot to floor again, you have less mass to speed back up. So, under acceleration the A4's heavy drivetrain soaks up HP. Under decel, it provides forward inertia. The RX-8 probably has less drivetrain acceleration HP loss and less forward decel inertia.

Yes/no?

LionZoo

That is correct. When out of gear, the extra weight in the drivetrain will actually make the car roll a longer distance.

Maybe you should try this, on a hill let the car coast from a dead stop. See how quickly it accelerates.

MadDog

You're absolutely right. The drivetrain is acting like a flywheel and storing energy. You have to be a little bit careful though. Just because the A4 likely has a more massive drivetrain doesn't necessarily mean its storing more energy than that of the RX8. They are storing energy as angular velocity squared and only as inertia to the first power. The A4 diff is a 3.38 ( I think) and the 8's diff is a 4.44. The A4 has 235/45R17's at 821 revs per mile while the 8 is running 245/40R18's at 809 revs per mile, so the tire circumference is only 1.5% different - call them equal. But, because of the 8's final drive, its drivetrain components are spinning 31% faster than that of the A4. So, the A4 drivetrain could have 72% more inertia than the RX-8 drivetrain before it would store as much energy as the RX-8. Maybe that's possible, but I doubt the A4 has that massive of a drive train!

The bottom line is that the A4 looses 20% of the kinetic energy stored as translational motion, while the RX-8 looses 68% of its translational kinetic energy in the same distance! I find that remarkable.

olddragger

iTrader: (3)

I'm dizzy.----------------------------------------------------
But--casually speaking--i noticed the same thing 2 yrs ago and posted it. I dont have an A4(great car by the way and cool looking) but I was comparing my 8 to my other car--Honda Odessy -i know i know
My car was better at the coasting thing once I swapped fluids( to royal purple) but it still seems tight to me. I have owned and driven over 34 differant cars in my miserable life and the only other one that I can remember feeling this tight was a 1969 Corvette. Hell even a CJ-5 with 900 multimile tires (36" tall) coasted better. Still love her though! Maybe she will loosen up after I reach 50K
Olddragger

rotarygod

The exhaust might give 10 hp or so. Tuning obviously is everything. It still doesn't change the fact that they have the stock transmission and are only losing about 12%. As I said, worst case scenario, the RX-8 loses 15% in drivetrain losses. A coastdown test on the street will show more than just drivetrain losses. It's complete rolling resistance on all wheels. Lots of things can affect this such as camber, etc. The 2nd generation RX-7 was a fairly light car but it was very hard to push for the same reasons. The RX-8 driveline is actually quite efficient.

MadDog

Agreed. But don't loose sight of the fact that those same rolling resistances must be overcome when accelerating the vehicle, too. Same for a dyno test - except only the rear wheels come into play. The losses are the same when accelerating or coasting.

r0tor

iTrader: (1)

so you saying 12% vs 15% for a stock car... so you thing ac delete, serious pulley reduction, light as hell flywheel, light rotors, light wheels, a probably more efficient rear end, probably better bearings, and racing lubricants is only worth about 7hp in drivetrain losses...

rotarygod

The cars they race are essentially stock when it comes to drivetrain. Stock brakes. Just different pads. Light flywheel does a little bit, pullies doa little bit (very little!), a/c delete doesn't do crap as I've never seen anyone dyno with the a/c on. Stock rear end, stock bearings, etc. Their car isn't THAT special over the street car so I think 15% is a pretty reasonalbe estimate. If you have a car that loses 20% in drivetrain, something is either seriously wrong or you have all wheel drive. The RX-8 does not lose that much to the drivetrain. Stop trying to justify low power numbers by claiming it's an inefficient drivetrain. It isn't. The car just doesn't make crap for power from the factory. Even the RX-7's which had more rotating mass, a heavier flywheel, driveshaft, 4.10 gearing etc. still dyno'd at around 215-220 on a 255 hp crank rating. That's a loss of between 12%-15%. Drivetrains in cars are better than most give them credit for. That's why I laugh at people who try to add in 20% or more for it's loss when doing a chassis dyno. It's not that great and is a way to high number to use.

The FACT is that the RX-8 loses between 12%-15% through the drivetrain on a stock car. Period. A coastdown test tells absolutely you nothing useful in determining this.

maxxdamigz

iTrader: (1)

15% to 12% is a 20% reduction in losses. That's pretty decent. Given the amount of losses occuring in the transmission its self, I think that's a decent improvement. If you go from a 25# stock wheel to a 17# lightweight race rim (assuming tire weights remain the same), you're probably not decreasing the rotating inertia of the wheels by 20%. Race lubricants and bearings might be better than 20% more efficient. The tranny and differential probably are not 20% more effficient. The flywheel and pully are probably more than 20% more efficient. The driveshaft/axles are probably no change. AC delete (if the rx-8 uses a clutched compressor) is probably more effective as a weight savings than a drivetrain loss reduction. Light weight rotors might be 20% better. So, I think going from 15% to 12% is a very nice improvement. 7 hp from just spinning stuff is a lot of spinning.

Rhythmic

So the Speedsource car's only mods are: Cat delete, race pads, lightweight wheels, heavy underdrive pullies, a/c delete, 4.10 gears, and a different ecu? That's pretty interesting.

The 4.10 gears move the powerband up further than the 4.44 stock gears? As opposed to the FEED 4.77 that moves the powerband down in the rpm range? Is this correct? I guess that makes sense b/c their only using high rpms...?

Rhythmic

do you know how the ecu is different? Leaner, higher rev limit, etc?

rotarygod

They've got the stock rear end gears. The RX-7's all had 4.10. There isn't much done to the Speedsource cars in terms of engine or drivetrain. They are far more simple than most people realize. They are really just lightened with a different suspension setup and a full roll cage. Pretty simple really. Their 3 rotor car is another story altogether. That is a ground up full race car.

MadDog

Hmmm... that's an interesting assertion. I'd say that aside from aerodynamic drag, they only possible thing that could cause the vehicle to loose energy would be frictional losses. Those same losses would have to be overcome when accelerating, too. Include aero drag and you have a pretty accurate picture of all the forces that you have to overcome in order to accelerate the car, which is, of course, the goal.