JPrevost

This little lesson is short and simple. Transmissions don't suck up a percentage of horsepower, it's a fixed drag with the only variable being speed.
So forget all that 20% drivetrain loss, it's not a percentage!
Automatics have more drag but since the torque converter itself is a virtual gear that multiplies torque when it slips. This takes more power. Lucky for drag racers traction is the problem, not power, but with slow TBI cars it's a big problem.
Manual transmissions don't have nearly as much drag so more power gets to the wheels hence most cars that are 12 seconds or slower will actually benifit from a manual transmission. Especially if it's got a gear more than the auto and is semi performance oriented.
Auto drag at 5000rpm might be a fixed 50 horsepower while a manual might only be 30hp. Make sence?
So next time you see somebody trying to brag about how much power they're putting to the ground remind them that they can't use the % rule .
So for you TBI guys, get a stick trans, you'll be happy and faster. End of story.

dimented24x7

iTrader: (2)

It is more correct to think of it in terms of torque loss rather then power. After all, the friction is a resistive force. Whether or not its constant torque loss with speed and trans gear is too hard to say.

One thing though is there is alot of friction from the shearing of fluid in the clutchpacks. Only like .030-.060" of clearance overall in most of them. You can see this in OD with a 700-R4 . Technically its not supposed to have engine braking in OD since the overrun clutchpack isnt engaged but theres enough friction to still drag the motor along when the car is moving. Another problem with the 700-R4 is the large gap in ratio from 1st to 2nd. Even with 2.77's, first feels great and the car accelerates like hell. As soon as it hits second, it bogs down. The solution to that is to wind out first manually before shifting but that sort of defeats the purpous of having an automatic if you have to shift it yourself.

These older TCs also suck. I have a TCI saturday night special (rebuilt stock unit) and the car definatly accelerates noticably faster with the torque converter locked. Newer torque converters feel much better. This with more speeds that are more closely spaced means that gap between modern autos and manuals is smaller then ever.

Pablo

iTrader: (1)

makes senSe

Cadillac

I've always known a stick was better: power, economy, control. I've also seen the benefits of an automatic in a variety of situations as well such as heavy traffic, 1/4 Mile Launches, etc.

For me however, it's all about the control.

dimented24x7

iTrader: (2)

Id go to a manual, but any decent stick is alot of money. A T-5 would never stand up to my driving. I have a bad habit of dumping the clutch and powershifting when Im in a hurry.

Pablo

iTrader: (1)

i think the weakness of T5's is exaggerated

Or people dont know how to drive stick cars.. one or the other

My non W/C t5 has lasted me this long on a mid 13 second car, i think a WC would be about bullet proof

I think the key is knowing if the stick doesnt want to go, dont force it

90RS305

Ehh, my car is perfect proof that the borg-warner w/c ISN'T bulletproof, seeing as how my tranny is all but kicked the bucket, but one thing is for sure I will NEVER own an automatic. If I have to buy one, I'll have a stick put in it, plain and simple. I'm actually getting ready to drop a T-56 in mine right now, if I can find one through my work cheap enough

kdrolt

Too simple. Here's the detail.

A sliding friction model from statics:

drag_force = weight_load*distance

where

distance = the distance you drag something

and

weight_load is how much something weighs. (It's meu*m*g for you CE/MEs).

Drag power is the d/dt of the above.

So drag_power = wgt_load*velocity

If we assume that the above model applies to the loads in the drivetrain (trans + rear end) and we allow that the weight_load in the above is really like the force loads within the driveline (reduced from the complicated sliding friction in the gear mesh, and sliding friction of all the bearings), then the driveline power loss is proportional to velocity (or speed).

Since you can rate an engine power output in fwhp (no driveline loss) or in rwhp (with driveline loss), then the ratio of the two gives you the percentage lost at the maximum rating for the engine. I'm also assuming that these two peaks occur at the same rpm --- which seldom happens but it's close enough for this point.

The percentage found from that ratio is the one everyone throws around without spending any time thinking about what it means.

To be more precise about it, which was what prompted your note, if you want to then know what the power loss is at any other engine speed, you have to make an assumption about how power varies with speed. I used a power proportional to speed model (in the above), which is a linear model... so if I want to know what the percentage of power loss is at 2500 rpm vs 5000 rpm (where my max power occurs), then a 20% loss at 5000 rpm would be a 10% loss at 2500 rpm. Likewise at 500 rpm the loss is 1/10 of what it would be at 5000 rpm.

The torque values for every rpm can then be estimated once you have the power at every rpm.

The problem is actually tougher than what I showed above because the power loss in a drivetrain doesn't really use a force = load*distance, or power = load*speed. The actual loss depedns on the tranny design, so the formula is more like power = load*(speed^n) where n is a number greater than 1.0

I also didn't include the static friction which happens at zero rpm, and I also didn't add the loss due to lubrication and aerodynamics (inside a gearbox). And I won't.... though that stuff can be added.

So JP is right that the percentage can't be applied at every rpm ---- it's really only valid to apply it at one value for rpm and then adjust it elsewhere. FWIW.

dimented24x7

iTrader: (2)

For that matter estimating powerloss through the drivetrain in an attempt to see flywheel HP is sort of a waste of time since the power/torque at teh wheels is all that really matters. Rotational mass also comes into play and a 700-R4 has alot of that. IIRC, I estimated from data that I got that there is about a 30% peak horsepower difference between 1st and 2nd. IOW, the engine made 30% less power in first then it did in second gear. There are aftermarket planetary sets for the 700 floating around that space the ratios closer together to minimize this but theyre like $1000 or something.

JPrevost

I did over simplify kdrolt, and the reason was so EVERYBODY would understand .
If I wanted to get more intense with details we could talk about the aerodynamics inside the gear boxes. I could also have talked about the inertia of each which also requires horsepower that doesn't make it do the wheels. Heck, we could talk about a lot of things but the general idea for everybody to understand that at x speed in and x speed out there is a torque needed to rotate the transmission. Assuming the 1:1 ratio the automatics always require more torque to turn than a manual. The time this isn't true is when the manual is over it's spec input torque which could cause the case to deform and bind the shafts against the bearings. So for simplicity's sake I just wanted to make it clear that a percentage is worthless for back calculating fwhp from rwph. In other words, a high horsepower motor will put more power to the ground than a weaker one just because the drag in the transmission is BASICALLY dependent on speed only.
I've attached some graphs showing what should I'm talking about.
Keep in mind this is an oversimplification. In real life the automatic would make up for some of it's lack of peak horsepower with low end power derived from the torque converter multiplication (a cvt of sorts).
The other error in this chart is the fact that automatics have a pump. This pump would cause the drag to not be linear and in some cases have areas where it's easier to turn at a higher speed than lower! Enough babble, here's the jist.

JPrevost

Some math examples;
Lopo engine makes 200hp at 5000rpm. Hipo engine makes 400hp at 5000rpm. Both are using an automatic transmission which takes let's say 60ft-lbs of torque to rotate at 5000rpm. Knowing we have a drag of 60ft-lbs we can calculate how much horsepower is getting through the transmission.
Derive the horsepower lost through the transmission by using the formula horsepower = torque * RPM /5252 . We get 57hp lost to the transmission to keep it rotating. Take the horsepower of the Lopo which is 200hp subtract transmission lost 57hp and you're left with 143hp through the trans. The Hipo is getting (400-57) 343hp through the trans!
So the % lost with the Lopo is closer to 30 and the Hipo is like 15. See why it isn't a percentage?