Drivetrain Loss: Fixed or %
Thread Starter
Super Member
Joined: Apr 2004
Posts: 723
Likes: 0
Drivetrain Loss: Fixed or %
OK so I was thinking & had this discussion on another forum where I got shot down. But I'm hoping you guys will be more conducive to a good debate. Why should the drivetrain losses be a percentage of the power? I mean you lose a certain amount of power through the gearbox, another amount through the torque convertor (if you have one), a few more HP through the diff, & some through the tyres. If you add all of that up you get the drivetrain loss.
It might work out to be a certain % of the power by coincidence. When you have more power stock the manufacturer puts in beefier components that sap more power. So the losses go up. SO it may seem like the losses are a % of the power.
Check this I got from googling:
Now it is true that not every component in a transmission system absorbs a fixed % of the input power. Some components like oil seals and non driven meshed gears (as in a normal car multi speed gearbox) have frictional losses which are not affected by the input torque. These losses do increase with speed of course but at a given rpm can be taken to remain constant even if the engine is tuned to give more power. We'll look at real world transmission loss percentages later. Finally, the biggest source of loss in the entire transmission system of a car is in the tyres - they account for half or more of the total losses between the flywheel and the rollers. Each set of driven gears, i.e. the final drive gear or the particular gearbox ratio that you happen to be testing the car in, only absorbs about 1% to 2% of the engine's power.
So over HALF of the drivetrain losses are due to the tyres.
Here's more info:
Gearboxes - losses in gearboxes are similar to those in an axle or a differential. Most of the losses are simply, bearing, and oil drag. These are engine RPM dependant, not power or torque dependant. Some of the losses ARE dependent on the power and increase as the power transmitted goes up. Marginally. So again a simple percentage cannot be used, and measuring the losses by letting the dynamometer "push" the vehicle with the clutch disengaged also does not allow for these load dependent values either"
"Rear (or front) diff / axle (car) losses - This is both load (torque) related as well as Road speed related. Both components need to be considered here. Oil drag, and grease in CV joints and wheel bearings and prop shafts, as well as internal windage losses and external windage losses on prop shafts etc, all increase as road speed rises. They are NOT engine RPM related or Power related. These all increase with road speed only, and are independent of the gear the vehicle is in. These are the largest cause of the total losses in differential, prop shaft, drive shaft, or wheel bearings.
However some other losses also occur in diffs and axles that are load (torque) related... Gear teeth, side loads caused by helical cut gears on thrust bearings etc do experience greater losses as the torque going through them increases. This effect is pretty minor compared to the above, but does occur. There is obviously an oil film present, or the components would destroy themselves, but the increase is measurable."
"Tyre / roller interface - for any particular tyre / drum the following is true... To begin with there is quite a high value from 0 to say 1 mph. This stays as speed increases but the power needed slowly increases with the road speed, (not engine RPM!). So this loss is greater in taller gears and at higher speeds. Again it almost completely unrelated to the unrelated to the power being applied. So is not a percentage of power! UNLESS they spin...
What each individual car loses is an unknown - it will depend on tyre sizes and pressure, suspension angles and other things, but it shouldn't be far from the figures above. For sure though, no 2wd car in the world, unless it has flat tyres and a gearbox full of sand, loses anything like 30% of the engine's power in the transmission and tyres as many rolling road operators would try to have you believe. In general though it is fair to say that low powered cars have higher % losses than high powered cars. This is because some of the frictional losses are independent of engine power and so represent a bigger drain on a small engine. For example, a 60 bhp Fiesta will have around 14 to 15 bhp total transmission and tyre loss (25%) whereas a 90 bhp XR2 will only have about 17 to 18 bhp loss (20%) - a smaller % obviously. By the time you get to RWD cars with engines in the 300 to 500+ bhp range, losses can eventually drop to as little as 12% to 14% or so.
We have frictional, rotational and viscosity related losses. The quick summary is that there are mechanical portions of the transmission and differential whose losses remain fairly constant with power changes and portions which vary in relation to power transfer. The combination of the two yields a loss curve which is neither a straight percentage or a fixed amount but somewhere in between. It is a curve. When you make small changes in output, moving a small distance along the curve, could yield either assumption. But when large power changes occur, you will see that the power output is neither the straight percentage or the fixed amount.
The basic proof is the conservation of energy.
A constant loss would yield a constant temperature, as power/torque was changing at the same gear/rpm etc., in both the transmission and the differential.
This is not true. Large temperature changes occur at greater power outputs with all other variables held equal. This temperature change (energy) is coming from the increased frictional losses as power or torque transfer increases cause increased friction and increased energy losses. That increased energy loss comes from the output, the only place it could come from, proving a fixed value incorrect.
Fixed percentages are also proven incorrect by the fact that the power losses assumed under a straight percentage assume that bearing losses, gear seperation force losses, viscosity and rotational forces all increase with torque transmission proportionally as does gear tangential force losses which is clearly not the case.
The quick easy answer is, it is a combination of nearly fixed losses and power dependent losses. But most of the losses are fixed. A small percentage is related to the input torque.
What say you?
It might work out to be a certain % of the power by coincidence. When you have more power stock the manufacturer puts in beefier components that sap more power. So the losses go up. SO it may seem like the losses are a % of the power.
Check this I got from googling:
Now it is true that not every component in a transmission system absorbs a fixed % of the input power. Some components like oil seals and non driven meshed gears (as in a normal car multi speed gearbox) have frictional losses which are not affected by the input torque. These losses do increase with speed of course but at a given rpm can be taken to remain constant even if the engine is tuned to give more power. We'll look at real world transmission loss percentages later. Finally, the biggest source of loss in the entire transmission system of a car is in the tyres - they account for half or more of the total losses between the flywheel and the rollers. Each set of driven gears, i.e. the final drive gear or the particular gearbox ratio that you happen to be testing the car in, only absorbs about 1% to 2% of the engine's power.
So over HALF of the drivetrain losses are due to the tyres.
Here's more info:
Gearboxes - losses in gearboxes are similar to those in an axle or a differential. Most of the losses are simply, bearing, and oil drag. These are engine RPM dependant, not power or torque dependant. Some of the losses ARE dependent on the power and increase as the power transmitted goes up. Marginally. So again a simple percentage cannot be used, and measuring the losses by letting the dynamometer "push" the vehicle with the clutch disengaged also does not allow for these load dependent values either"
"Rear (or front) diff / axle (car) losses - This is both load (torque) related as well as Road speed related. Both components need to be considered here. Oil drag, and grease in CV joints and wheel bearings and prop shafts, as well as internal windage losses and external windage losses on prop shafts etc, all increase as road speed rises. They are NOT engine RPM related or Power related. These all increase with road speed only, and are independent of the gear the vehicle is in. These are the largest cause of the total losses in differential, prop shaft, drive shaft, or wheel bearings.
However some other losses also occur in diffs and axles that are load (torque) related... Gear teeth, side loads caused by helical cut gears on thrust bearings etc do experience greater losses as the torque going through them increases. This effect is pretty minor compared to the above, but does occur. There is obviously an oil film present, or the components would destroy themselves, but the increase is measurable."
"Tyre / roller interface - for any particular tyre / drum the following is true... To begin with there is quite a high value from 0 to say 1 mph. This stays as speed increases but the power needed slowly increases with the road speed, (not engine RPM!). So this loss is greater in taller gears and at higher speeds. Again it almost completely unrelated to the unrelated to the power being applied. So is not a percentage of power! UNLESS they spin...
What each individual car loses is an unknown - it will depend on tyre sizes and pressure, suspension angles and other things, but it shouldn't be far from the figures above. For sure though, no 2wd car in the world, unless it has flat tyres and a gearbox full of sand, loses anything like 30% of the engine's power in the transmission and tyres as many rolling road operators would try to have you believe. In general though it is fair to say that low powered cars have higher % losses than high powered cars. This is because some of the frictional losses are independent of engine power and so represent a bigger drain on a small engine. For example, a 60 bhp Fiesta will have around 14 to 15 bhp total transmission and tyre loss (25%) whereas a 90 bhp XR2 will only have about 17 to 18 bhp loss (20%) - a smaller % obviously. By the time you get to RWD cars with engines in the 300 to 500+ bhp range, losses can eventually drop to as little as 12% to 14% or so.
We have frictional, rotational and viscosity related losses. The quick summary is that there are mechanical portions of the transmission and differential whose losses remain fairly constant with power changes and portions which vary in relation to power transfer. The combination of the two yields a loss curve which is neither a straight percentage or a fixed amount but somewhere in between. It is a curve. When you make small changes in output, moving a small distance along the curve, could yield either assumption. But when large power changes occur, you will see that the power output is neither the straight percentage or the fixed amount.
The basic proof is the conservation of energy.
A constant loss would yield a constant temperature, as power/torque was changing at the same gear/rpm etc., in both the transmission and the differential.
This is not true. Large temperature changes occur at greater power outputs with all other variables held equal. This temperature change (energy) is coming from the increased frictional losses as power or torque transfer increases cause increased friction and increased energy losses. That increased energy loss comes from the output, the only place it could come from, proving a fixed value incorrect.
Fixed percentages are also proven incorrect by the fact that the power losses assumed under a straight percentage assume that bearing losses, gear seperation force losses, viscosity and rotational forces all increase with torque transmission proportionally as does gear tangential force losses which is clearly not the case.
The quick easy answer is, it is a combination of nearly fixed losses and power dependent losses. But most of the losses are fixed. A small percentage is related to the input torque.
What say you?
MBWorld Fanatic!
Joined: Jan 2004
Posts: 7,092
Likes: 13
From: Salt Lake City (but not Morm)
2003 E55 & 2014 GL550
Don't quote me as this being Gospel, but I believe when I was out at Klee, Cor said they have calculated a 19% loss from crank to wheels.
Mine dyno'd at 585hp...thus 473 rear wheel.....sounds pretty close?
Mine dyno'd at 585hp...thus 473 rear wheel.....sounds pretty close?
Thread Starter
Super Member
Joined: Apr 2004
Posts: 723
Likes: 0
OK that sounds fair. But lets say you double your power to 1000hp or so. Do you expect your Drivetain loss to double? You currently lose 112 to the wheels. Let's assume you do not change any components on the drvetrain & they hold up fine to 1000hp (I know it's probably not possible, but hey). Let's also assume you change the tyres & they don't slip any more than before (also a stretch).
So you know expect to lose 224hp to the wheels. But why? It's the same drivetrain. Most of the losses are fixed. SOme components will lose more power as they heat up. But most of them stay constant at a given rpm. So I say the losses will go up to maybe 140hp, but they won't double.
So you know expect to lose 224hp to the wheels. But why? It's the same drivetrain. Most of the losses are fixed. SOme components will lose more power as they heat up. But most of them stay constant at a given rpm. So I say the losses will go up to maybe 140hp, but they won't double.
MBWorld Fanatic!
Joined: Aug 2004
Posts: 1,732
Likes: 1
From: NJ
2005 E55
Originally Posted by M&M
OK that sounds fair. But lets say you double your power to 1000hp or so. Do you expect your Drivetain loss to double? You currently lose 112 to the wheels. Let's assume you do not change any components on the drvetrain & they hold up fine to 1000hp (I know it's probably not possible, but hey). Let's also assume you change the tyres & they don't slip any more than before (also a stretch).
So you know expect to lose 224hp to the wheels. But why? It's the same drivetrain. Most of the losses are fixed. SOme components will lose more power as they heat up. But most of them stay constant at a given rpm. So I say the losses will go up to maybe 140hp, but they won't double.
So you know expect to lose 224hp to the wheels. But why? It's the same drivetrain. Most of the losses are fixed. SOme components will lose more power as they heat up. But most of them stay constant at a given rpm. So I say the losses will go up to maybe 140hp, but they won't double.
MBWorld Fanatic!
Joined: Mar 2002
Posts: 2,934
Likes: 387
From: Kenilworth, il usa
GT63S ePerformance 4dr, '25 Bentayga Black Ed.(wife), Wrangler and 250 MLB(kids)
The more power you make the more cooling you need.
Cooling costs horsepower,so it is not exactly just fixed losses but a combination of it and the loss that increases as a cost of hp increase.
Cooling costs horsepower,so it is not exactly just fixed losses but a combination of it and the loss that increases as a cost of hp increase.
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
From all the comparisons between chassis dynos and crank dynos that I've seen, the driveline eats around 15% to 20% of the power. The parasitic losses increase from the extra heat that extra hp creates on the transmission, differential, and tires. Not to mention, the exhaust pressures increases as hp increases and that too adds to the incrimental increase of the parasitic losses seen on a chassis dyno. Are the losses fixed, NO.
Trending Topics
Banned
Joined: Sep 2005
Posts: 1,023
Likes: 0
From: Lone Star State
w210 E55
Absent's Car Collection
....................Damn it man! When do you find time to drive your sweet cars? 
'05 G55(Kleemann headers,ECU and pulley)-wife's
DB9 Volante (wife's) in Winter
997S Cab daily driver
575 replacement on order :v
M5 in October/Nov (well..........almost sweet!
)
MB World Stories
The Best of Mercedes & AMG
Manual Mercedes? 6 Times Sindelfingen Let Drivers Have All The Fun
Verdad Gallardo
Mercedes SLR McLaren 722 S Is Extremely Rare Example Modified by McLaren
Verdad Gallardo
8 Classic Boxy Mercedes Designs That Have Aged Like Fine Wine
Verdad Gallardo
Flawlessly Restored Mercedes 190E Evo II Heads to Auction
Verdad Gallardo
Electric Mercedes C-Class Unveiled: 11 Things You Need to Know
Verdad Gallardo
Mercedes EQS Gets A Major Update: Everything You Need to Know
Verdad Gallardo
5 Underrated Mercedes-Benz Models That Don't Get the Love They Deserve
Verdad Gallardo
Mercedes 300D Has Pushed Well Past 1 Million Miles and It Ain't Stopping
Verdad Gallardo
10 Most Reliable Mercedes-Benz Models You Can Buy Used
Verdad Gallardo
Super Member
Joined: Aug 2005
Posts: 665
Likes: 0
The answer should be obvious - there is no answer. The only true way to determine the power delta is to stick the engine on an engine dyno and then plop it into the car on a chassis dyno. But that's only good for that particular dyno setup on that particular chassis dyno. Drive it down the street to another dyno and the answer will change. Take it back to the original dyno and it could change again. Chassis dynos should only be used to indicate probable power deltas from mods.
However, if similar cars are dyno'd on similar chassis dynos, and given a statistically valid sample size, you can then say that the mean of that population divided by the factory HP rating is the answer for that car on that chassis dyno. But there would be error bars around that answer.
Even the dual-roller chassis dynos that measure deceleration cannot give an accurate measurement of drivetrain loss. They only measure losses while the car is coasting and cannot account for the increased frictional losses due to the application of torque. But isn't it amazing how they always seem to come up with the correct number for crank HP when measuring a stock car. Can you say software?
However, if similar cars are dyno'd on similar chassis dynos, and given a statistically valid sample size, you can then say that the mean of that population divided by the factory HP rating is the answer for that car on that chassis dyno. But there would be error bars around that answer.
Even the dual-roller chassis dynos that measure deceleration cannot give an accurate measurement of drivetrain loss. They only measure losses while the car is coasting and cannot account for the increased frictional losses due to the application of torque. But isn't it amazing how they always seem to come up with the correct number for crank HP when measuring a stock car. Can you say software?
Thread Starter
Super Member
Joined: Apr 2004
Posts: 723
Likes: 0
Yeah Grumpy that's how I started with this line of questioning a while back. A buddy of mine has an RS6. Dyno'd it. On the "coast-down" test it measured the DL to be 98hp (4WD). He chipped it, decatt, induction, etc. It gained another 70 AWHP. So it went from 352 AWHP to 422AWHP according to the tuner.
But then they did the normal coast down test. The dyno showed almost the exats same loss. 97hp actually on the modd'd run. But I feel that just measure the drag on the tyres with the drivetrain not under load. So yeah, the modd'dcar might not have had more losses under no load.
But then they did the normal coast down test. The dyno showed almost the exats same loss. 97hp actually on the modd'd run. But I feel that just measure the drag on the tyres with the drivetrain not under load. So yeah, the modd'dcar might not have had more losses under no load.
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
Coast down test is BS. It will not show you any thing other than the drag on the engine and driveline while it is not under load. The load, heat, friction, and exhaust pressure is what causes the hp loss to increase as hp is increased. The loss is usually 15% to 20% of total hp.
Thread Starter
Super Member
Joined: Apr 2004
Posts: 723
Likes: 0
YEah Blown , that's what I'm saying. The coast down is BS.
But how can the DL be a straight % of the power? 50% of the losses are where the tyres touche the ground. I don't believe one can say its a %.
Most people that mod car put in aftermarket drivetrain components. What it you fit an LSD over your stock open diff? Uou get some LSD with the cooling fins that will reduce the temps. An efficient LSD such as this might give less loss. Lightweight flywheels, heavy duty clutch, upgraded suspension etc. In some cases you may find an modd'd car has LESS drivetrain loss than a stock car.
But how can the DL be a straight % of the power? 50% of the losses are where the tyres touche the ground. I don't believe one can say its a %.
Most people that mod car put in aftermarket drivetrain components. What it you fit an LSD over your stock open diff? Uou get some LSD with the cooling fins that will reduce the temps. An efficient LSD such as this might give less loss. Lightweight flywheels, heavy duty clutch, upgraded suspension etc. In some cases you may find an modd'd car has LESS drivetrain loss than a stock car.
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
I don't believe one can say its a %.
Senior Member
Joined: Sep 2005
Posts: 432
Likes: 40
From: Savannah GA
2020 G550
Originally Posted by BlownV8
Coast down test is BS. It will not show you any thing other than the drag on the engine and driveline while it is not under load. The load, heat, friction, and exhaust pressure is what causes the hp loss to increase as hp is increased. The loss is usually 15% to 20% of total hp.
One thing you forgot.
Simple physics.
As the time decreases to accelerate a mass the amount of energy required will go up.
Put simply, if your drivetrain and all rotating parts make the car go from 0-100 in 9 secs it will take more energy to get there than if it took 11 secs.
So....the more hp you are putting out, the more you are spending accelerating a mass in less time.
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
A small part of it is a %. The rest is fixed. It's more like a curve. Your drivetrain loss doesn't double if your hp doubles.
This is an exerpt from Lingenfelter Perfromance Engineering:
Lingenfelter Performance Engineering, Inc.
--------------------------------------------------------------------------------
March / April 1996
Race season is well underway and the shop is as busy as ever. The engine dynamometer has been especially busy with the drag race engines - including our own - as well as the street engines and R&D projects. Our Dynojet chassis dynamometer continues to get more and more use as well. We have now chassis dynamometer tested many stock and modified ZR-1 Corvettes along with a variety of other cars and trucks.
As part of this testing, we have been able to measure many different variables that effect rear wheel horsepower. One such variable that we have been able to quantify is the horsepower losses due to elevated vehicle operating temperatures. The LT5 engine seems especially sensitive to this. This highlights the need to cool the intake manifold and get the water temperatures down between drag runs. We have also been able to measure and quantify the losses between an engine dynamometer test run and a chassis dynamometer run of the same engine. Even when an engine is run on the engine dynamometer with all of the accessories and the catalytic converters (the way we usually run most engines), you will usually see losses of between 15% and 20% between the engine dynamometer and the chassis dynamometer. These losses are due to the additional backpressure of a complete exhaust system and the frictional losses in the transmission, differential and tires. We have even been able to measure the differences in friction due to changes in tire pressure - in one case a customer had left his rear tires roughly 15 psi low after leaving the dragstrip and then came to the shop. We chassis dynamometer tested his vehicle with the low tire pressure and then brought the tires up to proper street pressures and gained close to 15 horsepower at the rear wheels.
The Dynojet chassis dynamometer is an inertia type chassis dynamometer. This means that it calculates horsepower and torque based on how quickly a given inertia - in this case, a set of rolls of given mass and dimensions - is accelerated. The length of time it takes to accelerate from one rpm level to the next is the sweep time and the rate that you accelerate from one rpm to the next is the sweep speed or sweep rate. Because it takes more power to accelerate the mass faster, you will see lower horsepower figures when a car is tested in a lower gear (1st gear for example) then when it is tested in a higher gear (3rd gear for example). This is because at higher speeds and higher gears, it take longer to accelerate from one rpm level to another (for example, 2000 rpm to 6000 rpm). This remains true until the gains in horsepower from increased sweep time are offset by the increased frictional losses of the transmission, differential and tires. As speeds increase, the frictional losses in the transmission, differential and tires increase. The higher the horsepower of the car, the faster it will accelerate the rolls and the higher the speed (and therefore the higher the gear) will need to be to get the sweep time long enough to give an accurate reading. Because the Dynojet chassis dynamometer is an inertia type chassis dynamometer it does not allow you to perform fixed rpm or step type horsepower tests - you can not hold the vehicle at a given rpm or speed and check the horsepower level. Despite this limitation, the inertia type dynamometers give you a very accurate measurement of what the vehicle sees in real world situations. An inertia type dynamometer will show the effects of reduced driveline inertia (such as lightweight flywheels, driveshafts and wheels) while a steady state test does not show these improvements.
Because the Dynojet chassis dynamometer uses one large diameter roll per wheel, overheating the tires and the tendency of cars to try and jump off the rolls is not a concern. This design also means that the cars do not need to be loaded down against the rolls - further reducing tire heat build up and increased frictional losses through the tires. This means that the risk of tires exploding from too much heat and load is virtually nonexistent.
In our continued effort to offer more horsepower for the LT5 engine - we will soon be offering larger, lightweight stainless steel valves for the LT5 engine packages. The larger valves will offer improved airflow and the reduced mass will provide higher rpm capability to the LT5 engine. Keep an eye on this Shoptalk section to get more information on the improvements provided by the larger valves as well as from the titanium rods mentioned in the last issue. Many other improvements for the Corvette and the LT5 engine are also in the works.
For those ZR-1 owners that would like to upgrade to the newer 1994 through 1995 ZR-1 wheels or want an extra set of wheels, we have available a limited number of the GM 5-spoke ZR-1 wheels. A set of the 17 x 9.5" front and 17 x 11" rear wheels sells for $1099 while a set of four of the 17 x 9.5" wheels (for 1988 through 1996 Corvettes other than the ZR-1) sells for $1079. We also offer polishing and powder coating for $995 a set.
In our continued brake package improvements, we will soon be offering a brake package designed to bridge the gap between the stock ZR-1 brakes and our complete 13.5" Alcon brake package. The new packages will keep your stock 13" rotors but use the 4 piston Alcon calipers. These calipers offer increased stiffness over the stock PBR two piston floating calipers and should provide better heat dissipation and pad wear. You will be able to later upgrade to the 13.5" two piece rotors without having to change calipers. These Alcon caliper/13" Corvette rotor packages are also available for the Camaro, Firebird and Impala SS as well. The Alcon calipers are now also available powder coated. We now also offer the complete line of Performance Friction "-4" and "Z" compound street pads along with the Performance Friction racing compound pads.
With the weather finally starting to look better and spring already here (and summer not far away), it is time to start enjoying our cars again.
--------------------------------------------------------------------------------
March / April 1996
Race season is well underway and the shop is as busy as ever. The engine dynamometer has been especially busy with the drag race engines - including our own - as well as the street engines and R&D projects. Our Dynojet chassis dynamometer continues to get more and more use as well. We have now chassis dynamometer tested many stock and modified ZR-1 Corvettes along with a variety of other cars and trucks.
As part of this testing, we have been able to measure many different variables that effect rear wheel horsepower. One such variable that we have been able to quantify is the horsepower losses due to elevated vehicle operating temperatures. The LT5 engine seems especially sensitive to this. This highlights the need to cool the intake manifold and get the water temperatures down between drag runs. We have also been able to measure and quantify the losses between an engine dynamometer test run and a chassis dynamometer run of the same engine. Even when an engine is run on the engine dynamometer with all of the accessories and the catalytic converters (the way we usually run most engines), you will usually see losses of between 15% and 20% between the engine dynamometer and the chassis dynamometer. These losses are due to the additional backpressure of a complete exhaust system and the frictional losses in the transmission, differential and tires. We have even been able to measure the differences in friction due to changes in tire pressure - in one case a customer had left his rear tires roughly 15 psi low after leaving the dragstrip and then came to the shop. We chassis dynamometer tested his vehicle with the low tire pressure and then brought the tires up to proper street pressures and gained close to 15 horsepower at the rear wheels.
The Dynojet chassis dynamometer is an inertia type chassis dynamometer. This means that it calculates horsepower and torque based on how quickly a given inertia - in this case, a set of rolls of given mass and dimensions - is accelerated. The length of time it takes to accelerate from one rpm level to the next is the sweep time and the rate that you accelerate from one rpm to the next is the sweep speed or sweep rate. Because it takes more power to accelerate the mass faster, you will see lower horsepower figures when a car is tested in a lower gear (1st gear for example) then when it is tested in a higher gear (3rd gear for example). This is because at higher speeds and higher gears, it take longer to accelerate from one rpm level to another (for example, 2000 rpm to 6000 rpm). This remains true until the gains in horsepower from increased sweep time are offset by the increased frictional losses of the transmission, differential and tires. As speeds increase, the frictional losses in the transmission, differential and tires increase. The higher the horsepower of the car, the faster it will accelerate the rolls and the higher the speed (and therefore the higher the gear) will need to be to get the sweep time long enough to give an accurate reading. Because the Dynojet chassis dynamometer is an inertia type chassis dynamometer it does not allow you to perform fixed rpm or step type horsepower tests - you can not hold the vehicle at a given rpm or speed and check the horsepower level. Despite this limitation, the inertia type dynamometers give you a very accurate measurement of what the vehicle sees in real world situations. An inertia type dynamometer will show the effects of reduced driveline inertia (such as lightweight flywheels, driveshafts and wheels) while a steady state test does not show these improvements.
Because the Dynojet chassis dynamometer uses one large diameter roll per wheel, overheating the tires and the tendency of cars to try and jump off the rolls is not a concern. This design also means that the cars do not need to be loaded down against the rolls - further reducing tire heat build up and increased frictional losses through the tires. This means that the risk of tires exploding from too much heat and load is virtually nonexistent.
In our continued effort to offer more horsepower for the LT5 engine - we will soon be offering larger, lightweight stainless steel valves for the LT5 engine packages. The larger valves will offer improved airflow and the reduced mass will provide higher rpm capability to the LT5 engine. Keep an eye on this Shoptalk section to get more information on the improvements provided by the larger valves as well as from the titanium rods mentioned in the last issue. Many other improvements for the Corvette and the LT5 engine are also in the works.
For those ZR-1 owners that would like to upgrade to the newer 1994 through 1995 ZR-1 wheels or want an extra set of wheels, we have available a limited number of the GM 5-spoke ZR-1 wheels. A set of the 17 x 9.5" front and 17 x 11" rear wheels sells for $1099 while a set of four of the 17 x 9.5" wheels (for 1988 through 1996 Corvettes other than the ZR-1) sells for $1079. We also offer polishing and powder coating for $995 a set.
In our continued brake package improvements, we will soon be offering a brake package designed to bridge the gap between the stock ZR-1 brakes and our complete 13.5" Alcon brake package. The new packages will keep your stock 13" rotors but use the 4 piston Alcon calipers. These calipers offer increased stiffness over the stock PBR two piston floating calipers and should provide better heat dissipation and pad wear. You will be able to later upgrade to the 13.5" two piece rotors without having to change calipers. These Alcon caliper/13" Corvette rotor packages are also available for the Camaro, Firebird and Impala SS as well. The Alcon calipers are now also available powder coated. We now also offer the complete line of Performance Friction "-4" and "Z" compound street pads along with the Performance Friction racing compound pads.
With the weather finally starting to look better and spring already here (and summer not far away), it is time to start enjoying our cars again.
Thread Starter
Super Member
Joined: Apr 2004
Posts: 723
Likes: 0
Originally Posted by BlownV8
Why are you asking questions when you think you know the answer and will not accept one that differs from your opinion.
Gearboxes - losses in gearboxes are similar to those in an axle or a differential. Most of the losses are simply, bearing, and oil drag. These are engine RPM dependant, not power or torque dependant. Some of the losses ARE dependent on the power and increase as the power transmitted goes up. Marginally. So again a simple percentage cannot be used,
As for the diff, the losses go up with speed, but at a given speed say the same irrespective of the input power:
Oil drag, and grease in CV joints and wheel bearings and prop shafts, as well as internal windage losses and external windage losses on prop shafts etc, all increase as road speed rises. They are NOT engine RPM related or Power related. These all increase with road speed only, and are independent of the gear the vehicle is in. These are the largest cause of the total losses in differential, prop shaft, drive shaft, or wheel bearings.
But you are correct that some losses definitely are related to the input torque:
However some other losses also occur in diffs and axles that are load (torque) related... Gear teeth, side loads caused by helical cut gears on thrust bearings etc do experience greater losses as the torque going through them increases. This effect is pretty minor compared to the above, but does occur. There is obviously an oil film present, or the components would destroy themselves, but the increase is measurable."
MBWorld Fanatic!
Joined: Apr 2002
Posts: 1,288
Likes: 3
From: Frederick, MD
'72 Suburban
M&M, I think one problem you're having in this debate is that you're accepting everything you read as being true (so long as it supports your claims). How do you know that 50% of the drivetrain loss is caused by the tires?
My opinion, based on anecdotal evidence, is that it's a curve. It's always going to be a percentage of power, but that percentage isn't fixed. So if you're making 500rwhp and losing 15%, it may only be 12% at 1000rwhp. You won't have a fixed hp loss present in any driveline, regardless of the power being pushed through it (ie... 100hp lost at 500rwhp, 100hp lost at 1000rwhp).
My opinion, based on anecdotal evidence, is that it's a curve. It's always going to be a percentage of power, but that percentage isn't fixed. So if you're making 500rwhp and losing 15%, it may only be 12% at 1000rwhp. You won't have a fixed hp loss present in any driveline, regardless of the power being pushed through it (ie... 100hp lost at 500rwhp, 100hp lost at 1000rwhp).
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
I don't know the answer. No-one does. It is my opinion that it is not a % as there are losses that are engine RPM dependant, not power or torque dependant. The gearbox is the biggest source of loss after the tyres & most of the losses in there are not related to the input powerL
The loss is absoultely hp and tq dependent. As power increases, the friction off all the moving parts increases and the pressure of the exhaust gasses increases. You can't say for certain down to an actual percentage what the car looses on the chassis dyno but you can keep it within a good margin of error at 15% for manual cars and 20% for automatic equipped rear wheel drive cars. The loss for all wheel drive cars is even higher. The only true way to know down to the last hp would be to put the engine on an actual engine dyno. However, the 15% to 20% industry rule is fairly accurate based on actual engine dynos and chassis dyno runs.
MBWorld Fanatic!
Joined: Aug 2003
Posts: 8,882
Likes: 1,211
From: In my garage
E55, GLS450, GL63, GLE350
Here is some good reading for M&M. I hope you digest the information:
The fixed percentage math works out better than the fixed HP loss math.
The fixed percentage math works out better than the fixed HP loss math.
There are two main factors involved in the drive train losses:
a) Rotational mass of the entire drive train. This is a fixed number, since the weight of these parts does not change.
b) friction. This number changes with load - friction goes up as load goes up - its part of the laws (yes laws) of physics. You can not get around this. This is the larger portion of the losses.
For a simple example look here:
http://www.school-for-champions.com...ce/friction.htm
Cliff notes: simple friction calculation:
Fr = u x Fn
where Fr = the resistive force of friction, u = the coefficient of friction, and Fn = the normal or perpendicular force pushing the two objects together. Fr and Fn are measured in units of force, which are pounds or newtons.
note the = sign, both sides are equal each other. So increase Fn (HP/Torque you are applying to the drive train) and you get a corresponding increase in Fr (Frictional losses of the drivetrain).
This is obviously simplfied, since to do the actual drivetrain calculation would require much higher order math and requires details about every single part in the drivetrain, their coefficient of friction, weights, etc.
Thats why a fixed percentage is the 'rule of thumb' since nobody, including the engineers who build and design the cars, do or even care about this level of detail.
a) Rotational mass of the entire drive train. This is a fixed number, since the weight of these parts does not change.
b) friction. This number changes with load - friction goes up as load goes up - its part of the laws (yes laws) of physics. You can not get around this. This is the larger portion of the losses.
For a simple example look here:
http://www.school-for-champions.com...ce/friction.htm
Cliff notes: simple friction calculation:
Fr = u x Fn
where Fr = the resistive force of friction, u = the coefficient of friction, and Fn = the normal or perpendicular force pushing the two objects together. Fr and Fn are measured in units of force, which are pounds or newtons.
note the = sign, both sides are equal each other. So increase Fn (HP/Torque you are applying to the drive train) and you get a corresponding increase in Fr (Frictional losses of the drivetrain).
This is obviously simplfied, since to do the actual drivetrain calculation would require much higher order math and requires details about every single part in the drivetrain, their coefficient of friction, weights, etc.
Thats why a fixed percentage is the 'rule of thumb' since nobody, including the engineers who build and design the cars, do or even care about this level of detail.
