HORSEPOWER LOSS IN TRANSMISSSION 5

My wild guess is that in a manual transmission at nominal RPM (max power) loss is about 5% and in an automatic loss is let say doubled, 10%.

Let us hear the comments.

I have not checked it. But in a manual transmission, you would have a higher pumping loss than a modern automatic. Because automatics now use variable displacement pumps.
The gear sets in automatics are more efficient.
Also in a manual trans usually the countershaft is always being turned and thus is turning all the gears it is messed with. I would say the case is reversed. And the manuals present more drag than an automatic.

Speaking strictly from DragRacing experience with computer modeling going from SuperFlow SF-901 dyno data
to actual DragStrip testing

Automatic= 85.0 % approx drivetrain efficiency
Std= 90.0 % approx ""
(not including losses from rotational inertia )

when including rotational inertia losses it still very closely same results ;

Auto Trans= 88.0 to 91.5 % Eff
Std Trans = 93.0 to 96.0 % Eff

if you reverse eff , actual track results will not correlate to dyno data


Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

The dip in the automatic is most likely parasitic losses of the torque converter.
Drag strip is not the same as cruising down the road. In drag strip you would be into the TC much more. Down the road you would be in lock-up. And that is where the automatic will pass the manual. I think the pumping and rotational frictional losses in the manual will be greater.
What would not help the automatic is the drag from frictions, and steels in unlocked clutch packs. Unless it has the clutch separator rubbers or?

The extra performance of an auto in drag racing is not because of more efficent transmission of the power produced, but because the convertor slip allows the engine to stay in the rpm range where it produces more power. This more than compensates for lower efficiency of the transmission.

I would think that a good guide to transmission efficiencies could be found by comparing published fuel economy data for cars that were otherwise identical, except for the transmission.

This might be skewed very slightly to favor the manual, because the auto is normally a few pounds heavier, which would cause a very slight increase in consumption even if the transmission losses were identical.

I think the variation in consumption is much more than can be xplained by weight differences

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

Any extra losses in an automatic would be because of the TC.
The only way to compair would be to, remove the TC and use a clutch just like the Clutchflites and Turboclutches, back in the old days. Well unless that is still being used today.

I think the losses are less in a manual, esp. w/ATF in it. The countershaft isn't spinning that fast, about 2.5 times slower than the engine, so fluid-drag losses are lower.

An auto must run a pump to apply the clutches, and I think that's where the biggest loss is.

The best way to measure the two is on a dyno, where driver technique/car differences are removed. A non-TCC auto wouldn't come close to a good manual.

Hopefully someone that has dynoed both will post here.
In the manual trans the counter shaft and the other gears in constant mesh with it. Become the pumps. Not only is there massive windage, and the lifting of the oil out of the sump and coating the gears, but the squeezing effect between the gear teeth. The basic arrangement in the automatic is much more efficient. The variable displacement pump is putting out just enough volume to maintain the holdig pressure and lube requirements. At a lower speed the manual may have a less drag. But as speed increases the drag is going to increase. Where in an automatic the drag should stay pretty linear.

I have been wondering about tranny power losses....wouldn't the loss be better defined as an actual number as opposed to a percentage.

Lets say motor X makes 100hp (at the crank) and looses 15hp with an auto and only 10hp with a manual. Now through the magic of the aftermarket the same engine now makes 200hp at the crank...will it loose 15 or 30hp with the auto and 10 or 20 with the manual? Oh yea...the transmission is the same.

Thanks guys!

I'd think more in terms of a fairly constant portion (for the parasitic losses) plus a portion that's a percentage of the amount of power being transmitted. With the above numbers, I'd guess a little less than 20 vs slightly under 25.

Norm

AT's are more efficient at and just after launch because the TC multiplies torque through the stator, providing more input shaft torque. Once the input shaft starts rotating and torque multiplication falls, it becomes more inefficient. Also in an AT, there a multiple multi-plate clutch packs, and some have bands, that add considerable drag loss to the transmission compared to synchros in the manual.

When is the last time someone saw a tranny cooler as standard equipment on a manual tranny?? Qualifier, I refer to run of the mill, straight forward standard vehicles, nothing non standard!!!!

rmw

"AT's are more efficient at and just after launch because the TC multiplies torque through the stator, providing more input shaft torque. "

Care to explain that? Efficiency and torque multiplication are separate concepts in this context.

Cheers

Greg Locock

What are we trying to measure here guys ?

Transmission "loss" is all going to show up as heat. But under what conditions are we measuring this loss? In a road car it would probably be constant road speed low torque cruise conditions perhaps.

A lot will also depend on torque transmitted and shaft RPM. Oil churning loss is going to be insignificant at very low operating RPM, and heating from oil shear minimal at high RPM and low transmitted torque.

Oil viscosity is going to have a big effect on both these under different operating conditions.

There will be a whole range of operating conditions from maximum torque in first gear to constant speed top gear at small throttle opening.

As others have also pointed out, there is a lot more to quick quarter mile times than gearbox oil heating. Choice and number of ratios, rotating inertia, just to name a couple.

Estimates for a Lepelletier ZF 6HP26 :-

Drag in clutches, 3% losses
Gear inefficiencies 9% losses
Oil pump at 3000 RPM 12 bar, 5 litres per minute 5% losses

Total estimated losses 17%

=============================================

As I understand it, clutch losses seem to be about ½% per dragging clutch. Planetary gears seem to be about 96% efficiency (ie 4% losses) when not in direct drive, eg 92% losses when two planetary gearset are compounded. For the transmission above, the losses are so high due to the compounding effects in higher gears like 5th or 6th, compared to 3rd which is say 96% efficient (4% losses) but only used for half as long as 6th on average.

=======================

I don't know if you count the dual clutch transmissions as automatics. Once something is drive-by-wire, the difference between manuals and automatics is eroded: (1) software prevents stupid gearchanges in manual modes, eg changing to first at Autobahn speeds, so real manual operation has been removed (2) paddle shifts or +/- controls can be added, so automatic operation has or can been extended.

These are more efficient than a manual if you count fuel economy or 0-60 times. This can be seen by comparing the figures for manual and dual clutch versions. But maybe its not fair to do that. Fuel economy is improved by better gear selection, and 0-60 times are improved by the lack of torque interruption.

You could estimate the inefficiency of automatics by looking at 0-60 times for manual and automatic version and by looking at top speeds.

When people talk about a car and horsepower, they are not talking about how much horsepower is used driving to the supermarket. They are talking about maximum horsepower.

Let's have a look at an example, albeit just a ZF CVT, not a real automatic.

=============

BMW Mini CVT.

PERFORMANCE, FUEL CONSUMPTION
AND CO2 EMISSIONS
MINI One: 1,598cc manual: 0-60: 10.6 seconds; top speed: 115mph; combined mpg 43.5; CO2 emissions 158g/km; Emissions Class EU4; Insurance Group 5

MINI One: 1,598cc CVT auto: 0-60: 12.4 seconds; top speed: 106mph; combined mpg 36.7; CO2 emissions 187g/km; Emissions Class EU4; Insurance Group 5

MINI One: 1,598cc chain cam petrol: 66kW (90bhp) at 5,500rpm/140Nm (103 lb ft) torque at 3,000rpm.

=================

106/115 = 0.92 = 92%

So the CVT's top speed is 8% less than than the manual's. The CVT presumably has an identical gear ratio available to the manual, so the lack of top speed is due to the CVT's inefficiency.

Top speed is governed by air resistance which is a square law, so an 8% drop in top speed means about a 16% drop in horsepower.

16% * 90bhp = 14.4.

So the CVT automatic mini loses about 14 horse power more than the loses on the manual, out of 90 BHP.

(The CVT 0-60 time is also about 15% slower than the manual despite the mini CVT always having the ideal ratio.)

===

If you have a torque convertor on your automatic rather than a friction launch clutch, then you can add on losses for that too, if looking at fuel consumption type losses, but if looking at numbers to impress the public: horsepower, top speed, etc, the convertor would be locked up and you can forget about it.

SAAB 9-3
2.0t 5-speed manual: 0-60 8.2 seconds; top speed 135 mph; combined mpg: 34.0; CO2 emissions: N/A g/km.
2.0t 5-speed automatic: 0-60 9.4 seconds; top speed 132 mph; combined mpg: 30.1; CO2 emissions: N/A g/km.

Difference in top speeds 3mph, 2 1/4%, or (doubling because wind resistance is a square law) Manual is 4½% more efficient when flat out.

MPG difference 3MPG in about 33 or 9 in 100, ie 9%.
Manual is 9% more efficient in general.

0-60 just over one second in 9 difference, ie bit more than 10% difference in efficiency. Probably bigger than the other figures due to convertor slip.

PERFORMANCE (MANUFACTURER'S ESTIMATES)

Acceleration 0-60 mph (0-100 km/h) - sec Manual: 8.9 (9.4)
Automatic: 10.4 (10.8)
Top speed - mph (km/h) Manual: 130 (210)
Automatic: 127 (205)

FUEL CONSUMPTION (1)
Urban - mpg (l/100km) Manual: 22.2 (12.7)
Automatic: 19.8 (14.3)
Extra urban - mpg (l/100km) Manual: 39.6 (7.1)
Automatic: 37.9 (7.4)
Combined - mpg (l/100km) Manual: 30.7 (9.2)
Automatic: 28.4 (10.0)
Carbon dioxide emissions g/km Manual: 219
Automatic: 239

==============================

Acceleration 0.5 in 10, about 5% difference

Top speed 5 in 200, 2½% difference in speed ie 5% difference in efficiency

Urban Fuel 2.5 in 22, about 12%
(slipping torque convertor?)

Non Urban 1.7 in 37. Call it 5% again.

Carbon dioxide 20 in over 200, bit under 10% difference.
That'll be 5% plus some more for urban driving.

oops, accelaration in that last example,

It should be 1.5 in 10,

15% more losses than the manual
that's more like it,

I said 15% for the mini CVT
and over 10% for the SAAB,
so 15% worse accelearation for the Jaguar automatic would be about right.

Thank you crystalclear. At last some real data in this thread.

I expected the original poster might have done what you did after my original post. Maybe he did, but didn't think to share it with those that helped him along the way

One point I never thought of earlier, is that various torque converters are matched to the engine, vehicle and overall gearing. How well this is done effects efficiency, as does the number of ratios and the ratios selected for the box and the final drive. Final drive ratios are inevitably different on auto vs manual, but I would presume that on the cars you mention, considerable time was spent optimising both transmissions, but I wonder if marketing identified different market sizes and preferences between auto and manual buyers and optimised to a different set of parameters, re economy, noise levels, performance, manufacturing cost, development time etc.

I think my above concerns would only possibly account for a very small efficiency variation compared to the figures quoted.

I know, a very disjointed post. I wrote as it came to mind.

I guess what I am saying is that comparative fuel efficiency and acceleration and speed figures gives the efficiency of the vehicle, not the transmission per-say. Even if the transmission differences are in otherwise supposedly similar models, there are always other small variations

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.