gearbox experts comments?

[dougn1]

I am re-gearing my 91 C2 for street use. I have dynoed my car and used the torque numbers and gear ratios to generate these plots. They represent the force in pounds at the contact patch. I think they are called acceleration curves. Each plot has five curves….one for each gear. The curves start at 2700 rpms and stop at 6700 rpms.

My goal is to come up with a sporting street gearbox and to do so I have imposed a couple conditions. These conditions are as follows:

1) Retain stock 1’st gear because of light weight flywheel

2) 5th gear cruse at 80 mph = 3700 rpms (sets top speed at 145 also)

The first plot is using the stock gear ratios. Note the huge gap between 1’st and 2nd. The second plot is my proposed gearbox (3.5 2.4 1.68 1.29 1).

I’d like any comments about my proposed ratios (are the jumps from 3 to 4 and 4 to 5 to great) . This is not meant to be a track gearbox…it’s a backroad gearbox

[Bill Verburg]

The curves seem somewhat optimistic, what peak hp and torque #s, what tires, and stock gears?

I found that sort of plot to be useful for comparing tires and what effect they might have at a given track

here's one I did



Far more useful is something like this

[dougn1]

stock gears are 3.5, 2.059, 1.407, 1.086, .868
tires are 25 inch diameter

[randywebb]

what hp & torque?

I like BV's thrust and rpm drop analysis methods.

Also, it would be useful to know the torque spread of the motor -- if the rpm's drop by 2k what torque are you left with?

[Bill Verburg]

Here's where you are now


That's really not that bad for a street trans.

[dougn1]

the gap between 1 and 2 is huge and between 2 and 3 is a little too much IMO. also it seems like 2nd is too tall for alot of the tight turns i encounter...and 3rd is too tall for the short straight sections i normally drive. so i'm thinking it's going to be a huge improvement. what am i missing?

also those force numbers are torque(ft-lbs) x gear ratio x final drive ratio / tire radius(ft). is that correct?

[Bill Verburg]

thrust = (.85*TG*FG*torque)/(DTR)

thrust -lbs
TG- transgear
FG-final drive gear
torque- measured on a dyno in lb-ft
DTR-dynamic rolling radius of the tire in ft.
.85 is needed to change lb-ft to lb of thrust

[dougn1]

i don't understand the .85 factor???

[randywebb]

mech & heat losses in the drivetrain (tranny)

think of it as the 15% number...

[Bill Verburg]

Ok, the ROW 964 Carrera RS g50/10 fixed those problems


now you could lower 4,5 or 3,4,5

remember ~$800/gearset

[dougn1]

ok...these are already taken into account in the torque measurement (at the wheels)....in this case

[Bill Verburg]

If you believe that the dynos we have access to are precision measuring instruments

[dougn1]

actually there's no reason (that i can see) they shouldn't be. the one i used just spun up a big drum. if you know the I of the drum the rest is at least easy to measure and compute.

it was certainly precise based on its repeatability.

as far a accuracy....i here people say one thing or another but untill someone comes up with a good (physical) reason why they are not accurate......

[Bill Verburg]

just as a for instance
factory rating on a '91 964 is 250hp@6100, 228lb-ft@4800

according to the dyno your 964/02 is putting out 245hp@6100, ~231LB-FT@4700 to the rear wheels

pretty impressive for a 15 yrr. old stock 964
accuracy and repeatability are not the same and should never be confused w/ each other.

[randywebb]

"accuracy and repeatability are not the same and should never be confused w/ each other."

- very true


Bill, do you have spreadsheets already loaded with data & formulae that you want to share? If so, for which cars?

[Bill Verburg]

I have templates that are modified w/ info from various sources.

one of these days i'll get it organized the ones above are useful but can be improved, they are not simple to do

[randywebb]

re the above - here is what I came up with quickly... We should starta new thread if there are responses, so this thread stays on topic...


bias -- a consistent tendency to overestimate or underestimate the true value

imprecision -- lack of repeatability of series of measurements

accuracy -- extent to which measured values approach true values; consists of two components: imprecision and systematic error (= bias)

systematic error -- bias

resolution -- smallest measurement than can be made

repeatability -- extent to which repeated measurements agree with each other

reliability -- precision of a measurement as determined by the variance of repeated measurements


REFERENCES


McGraw-Hill Dictionary of Scientific and Technical Terms. 1974.

Chu article

Vogel 1981, p. 8

EPA manual - got from Ed @ Nat'l Water Quality Lab, USGS


Miller, J.C., and Miller, J.N., 1993, Statistics for Analytical Chemistry (3rd Edition): Analytical Chemistry Series: New York, Ellis Horwood PTR Prentice Hall, 233 pages. -- commonly used as a class text

[randywebb]

Originally posted by Bill Verburg
they are not simple to do

... hence my query


sounds like they are not yet ready for prime time

[dougn1]

"just as a for instance
factory rating on a '91 964 is 250hp@6100, 228lb-ft@4800

according to the dyno your 964/02 is putting out 245hp@6100, ~231LB-FT@4700 to the rear wheels

pretty impressive for a 15 yrr. old stock 964"

on the same dyno, several stock boxster s's put out 220 rwhp...in line with 250 claimed fwhp

"accuracy and repeatability are not the same and should never be confused w/ each other."

you called it precision....i was pointing out it is in fact precise. it's probably accurate too based on the above comparison.

my car has headers and a re-map. i think it's not uncommon for a stock internals 964 to put out 240+ rwhp