Mech33

Ah, I get your point now. At the lowest level innards of most common dyno hardware (inertial dynos), they are simply measuring angular position vs. time of the roller, let's call it theta(t). The computer in the dyno can then calculate:

position vs. time:

• theta(t)

angular velocity vs. time:

• omega(t) = d(theta(t))/dt ~= [ theta(t+dt) - theta(t) ] / dt

angular acceleration vs. time:

• alpha(t) = d(omega(t))/dt ~= [ omega(t+dt) - omega(t) ] / dt

power vs. time:

• power(t) = (dyno roller inertia I) * alpha(t) * omega(t)

Technically you could say that torque is being calculated in the power calculation, since torque vs. time is:
torque(t) = (dyno roller inertia I) * alpha(t)

So:

• power(t) = torque(t) * omega(t)

Okay... but I fail to see how that is helpful or relevant to the discussion, since technically the dyno is not (at the very lowest level) measuring torque nor power (nor force)...

stout

Nail hit on head.

Signal to noise on this thread is so far off the deep end that I'm afraid the value of the thread has been significantly diminished. Some of the stuff written here makes my head hurt, and I make a living untangling sentences.

If I'm just most customers/readers, I'd just leave. If I'm Rennlist or Internet Brands—or even just a Rennlister—that's a very big deal. I don't know what the solution is...should thread starters have an edit option? Is it purely a moderator thing?

And this is the very real cost—to all of us.

I was having coffee with a Porsche friend and longtime Rennlister last weekend, and I mentioned Rennlist and he rolled his eyes. He said he hadn't been back here in a very long time. His posts were among the best on RL, imho, and he is an avid reader and has had and continues to have significant experiences to contribute here. But no more. He mentioned the signal to noise, and increasingly unfortunate tone.

Mech33

While I think much of this discussion is very relevant to the original point of this thread (discussing performance vs. gearing), perhaps this post will be more on point.

I made the attached plot from the earlier GT4 whp dyno posted. It shows the power available as a function of vehicle speed in each of the GT4's 6 gears.

To me, this says there is definitely room to shift the upper gears down to close up the big hole in the power spacing post-shift between 1st and 2nd without significant reduction in high speed performance.

Bill Verburg

For an inertial dyno the above is true
but for an absorption dyno they are actually measuring the force, yes it too is derived but it is different in that there are ways to measure it.

Why is this relevant?

I just hate to let statements that are patently false, like this one go uncommentated,
no dyno measures power, they measure either torqu or angular acceleration of a known mass

If a GT4 drive wheel pair applies 2238lbs of thrust @2200rpm to the road that can be(and is) measured and back tracked through cwp and individual gear set to appear as 223lb-ft of fly wheel torque(assuming 15% driveline loss), from that it can then be calculated that the engine power @ wot & 2200rpm is 94hp

I do have dyno charts showing hp & torque curves for the engines of various cars, similarly I can diagram out transmission plots they give a glimpse of what the actual dynamics of the car may be in a straight line on level ground w//o any air resistance, but...

When you get to questions like Pete's original one, the effects of gearing and aero are critical to understand and see, that's why I very much prefer to do the thrust curves of which I posted a couple of examples earlier in the the thread.


Another point wrt to dynos is that engine torque can be directly measured on an engine dyno, from that similar thrust curves can be calculated, or measured independently w/ data acquisition

and yes the thrust curves bear a passing resemblance to the engine torque curve(but not the engine power curve) why should this surprise anyone, torque is just circular force which is translated by gearing and tires to linear force, less aero effects

Mech33

Bill your comments are fair and agree with your point, that dynos technically do not measure power. Though by the same nit-pick, none of them measure force or torque either. Even the braking / absorption dynos do not actually measure torque... they measure some other characteristic that they calculate back to the torque or power (such as current on a motor, water pressure or flow, braking force applied with some mechanical advantage, etc). They can convert that back to torque (or to power based on the rotation speed), in the same calculation... technical semantics.

Edit: taking that further, even the current measurement in the motor applying the resistance in the dyno isn't actually measuring current... it's probably measuring a voltage drop across a resistance as that current flows across it. And technically the voltage isn't being measured directly, either... you're reading the digital output of some level comparators that taking the correlation of those 1's and 0's back to an analog voltage measurement.

So whether the actual "measurement intent" of the dyno is to measure power, or to measure torque, is really up for interpretation, no? Especially considering they are just proportional based on the angular velocity.

Manifold

Bingo, thank you! IMO, the graphs clearly show what's going on:

- Big drop in available power at the moment of shifting from 1st to 2nd, even if you wind it out. More so if you don't wind it out. And who actually winds it out in 1st?

- Not a big drop in available power at the moment of shifting from 2nd to 3rd if winding out 2nd, which is reasonable to do on the track. But if you don't wind out 2nd - which wouldn't normally be done on the road - the drop in power is to about half power.

- The drop in power from 3rd to 4th, and beyond, isn't too bad.

I'm no expert, but I'm thinking that perhaps 1st needs to be a bit longer, 2nd and 3rd shorter, and 4th and 5th a bit shorter to even out the spacing from 1st to 5th.

Would be interesting to see vertical lines showing the power drop if shifting at say 5500 rpm, for spirited road driving.

Also, notice the "sag" in the power curves below mid revs, which exacerbates the feeling of lack of power (because the sag actually causes some lack of power at those revs, so this ties in with the torque curve aspect).

Mech33

Just for Bill and Mark, I also made the attached "Force vs Vehicle Speed" plot. Note that to create this, I simply took my previous "Power vs. Vehicle Speed" plots and divided by the vehicle speed at each point, and multiplied by 375 to convert the units of "HP / mph" to "pounds". I never calculated torque directly or used the torque curve.

Neither set of plots includes aerodynamic drag or rolling resistance losses, because I am not interested in the exact absolute acceleration force at every point, and adding those forces won't change the curve intersections anyway.

If you look at where the force curves intersection gear-to-gear, they are at the EXACT same RPM and speed points as where the power curves intersect. This is expected, per all the discussion in the thread.

Personally, I find the "Power vs. Vehicle Speed" plot much easier to read and use... hence the debate.

Mech33

Agreed. I think a few different combinations would work to improve both street and track performance (generally), probably involving 2nd, 3rd, 4th, and perhaps 5th for those tracks with long straights.

Bill Verburg

Nothing wrong w/ either of those graphs they show the same thing in different ways, i've done the same for myself, but...
neither reflects aero effects, it would be easy to add that to the thrust graph though

neither reflects the effect of vehicle mass on performance potential, that could also be easily added to the thrust curves.

That's because all the pertinent information to the conversion is in the conversion factor 375.

I repeat the thing that you are most interest in is the linear force or thrust that the car can generate in pounds, Maybe it's just me, but if that's what you want to know, start w/ what is actually measured, pound-feet and the drive tire radius in feet and the engine rpm. why complicate the issue by the power curves that only obfuscate the issue at best and hide additional influencing factors at worst.

Manifold

The force curves fit driver observations as well.

If you wind out the gears on the track, the force drops progressively from 2nd and beyond, pretty much according to F = P / V.

But for more normal road driving, with medium shift points, the sudden force drops from 2nd to 3rd, and to some extent 3rd to 4th, will be quite noticeable, and you'll be hitting those dips in the force curves during those upshifts.

I think we may have QED.

Mech33

No... [HP/mph] is already a unit of force, it's just in an odd unit. The 375 is literally a scalar multiple just to convert that to pounds.

Agreed that force vs. speed is also intuitive, but only when one goes through the trouble to calculate it the correct way. If you don't, you end up with incorrect conclusions from only interpreting the torque curve by itself without fully taking into account specific gear ratios. The power curves just eliminate that common source of error.

Different strokes for different folks!

sl951

Stout - if you're looking for validation that the GT4 gearing (1,2,3 & 6) isn't great for street driving, the answer is YES. Many other editors have noted this, the majority of owners feel the same, and even key folks at Porsche know it.

Can we encourage someone to fix it (aftermarket, etc) per your 991s gearing solution?
How much real demand is there and would people purchase?

Every Porsche model has an 'achilles heel,' gearing seems to be the GT4's

I haven't researched the techs and yes it's not a street legal car - but are the ratios 'more correct' on the PDK GT4 Clubsport?

mark kibort

first of all ... great discussion bill and mech33. (great graphs)

again, as i mentioned much earlier, there is a reason porsche made the spread of 1st to 2nd. its a track car first, and because of that, 1st is for leaving the pits and 2nd though 6th is track use. 2nd does have a reduction in thrust forces (HP utilization) from 45mph to about 55mph, but that's ok, because if you are ever going that slow, you are rounding a turn, and usually cant go full throttle until 55mph anyway. since you spend much of the time in the 55 to 100mph range, the gears work exceptionally well. Matt and i might argue what is optimum, but you KNOW this is a great gear box with great spacing.

as was proved, as long as you keep the engine in the max HP range 6000rpm to 7000rpm, you are extracting the most out of the car.

in regards to street driving, you mention lower RPM shifts.. generally, you realize that most lower RPM shifts are also less throttle shifts as well. so that might not be a factor.. ... but that depends on you as a driver.

For Bill, both Mech33 and I were just showing why using HP is so effective in quickly determining shift points.. sure, your excel sheets do the same thing, no doubt. as was said, power is a rate of change of KE. its easily measured directly, but as was implied by Bill, indirectly you are still getting torque, as a answer because of the use of the one of the 3 quantities... mass /time/ distance. Its a chicken and egg question for sure, but the real answer here is knowing that with knowing power, based on power curves you can determine shift points and comparative performance.

you mention that the graphs follow Force=Hp/velocity.. yes... as you know acceleration is proportional to power at any speed, so as speed goes up acceleration always goes down directly proportional.. best case. (ignoring aero)

the aero that you bring up bill (as well as losses) , will not change the cross over points, as they will effect the same car in the same way. however with different cars, absolutely, there are net forces that need to be taken into account. those forces can easily be applied to a HP curve as well. (force x speed=power).......so you tell me the force, and we can subtract the power required to drive the aero, or what the hp equivalent is for friction, OR changes to mass, whether transnational or rotational. not that it cant be done your way Bill, it just seems much easier to use the HP curves for me, anyway, the answers should be identical.

good stuff

mark kibort

again, the ratios utiize the engine's hp almost perfectly , aside from the 1st to 2nd gear drop. it is like almost all other close geared 6speeds out there and the PDK 7 speed is different as it fills in some of the gap, which are smalll with 7 gears. as long as you are at in the max HP range, i dont care what the editors say, thats the best performance you will get. sure that little drop of power from 1st to 2nd can be annoying, in a straight line, but its a sacrifice to make all the other gears, where you spend MORE time in , to be MORE effective. changing 3rd lower, for example will hurt you at several tracks i know. 2nd isnt used that much, so the gap out of turn, from 1st to 2nd, is momentary and you are probably traction limited as well.. now , make a drag car out of it, and that 1-2nd might be something to change. Matt has already said he is not changing 2nd or 6th, and he is the authority on gear sets for the GT4. (he wants to make them just a little more perfect, by bringing down 3rd just a bit and the rest follow) might be good on the street, but its crap shoot for all tracks and street speed ranges (WOT) some will like it, some wont.
part throttle street driving is another story all together.

my street car has the same gears. Ive learned to like the 1st to 2nd drop (same as GT4). also keep in mind, almost all Porsches have this larger drop from 1st to 2nd.

mgerber

^ spot on with your thinking. I want the car set up optimally for the street, and most importantly for how it feels at legal road speeds which is where I do most of my driving. The track talk is interesting to read through, but not relevant for how I use this car. No offense meant for the track junkies here.

For my 993 box build we did a longer 1st, stock 2nd, and shortened 3rd - 6th (GTgears gear sets and mainshaft in the box). This car is a street car, and frankly the most enjoyable to drive of the 10+ 911s I have owned, and the many more I have driven.

As GTgears has pointed in the past, GT4 drivetrain is very different than the 993, so makes sense to shorten 2 - 5 given the higher redline and shorter final drive. I am in if/when the gear sets become available.

The RPM and power drop from 1-2 drives me nuts. Check out the 993 section if you want to get some insights into different re-gearing approaches for the street.