I was bored one day and thought I would make a program to find the optimal shift points in my stang (t45). The only data you need is your torque curve (I had a dyno sheet) your gear ratios and final drive ratio. I used matlab software to plot a few graphs for you guys and I have had a full pi swap so my torque curve closely resembles 99-04 cars with basic bolt ons.


x axis is speed in mph
y axis is torque to the wheels (torque*gear*final ratio)

Using the intersection points, I find the rpm where shifts take place so put down the most power to the ground.

Parts that matter over stock:

253/292 hp/tq
Engine mods:
full pi swap with 10.4:1 compression
Long tubes
O/R H pipe (w/ stock Catback)
Steeda UDP
Accufab 70mm TB
sniper SF tuning software (fine tuned with wideband)

driveline:
FRPP 4.10's
315/35/17 (same OD as 245/45/17)

My torque curve is basically the same shape as a stock pi car... it peaks at ~4000 rpm and falls pretty evenly til around 5500 or so. just like a normal pi engine car. So those of you with 4.10s can use my same points.

ratios
gear 1: 3.37
gear 2: 2
gear 3: 1.333
gear 4: 1
gear 5: .667
final: 4.1

1-2
6213.76 rpm @ 34 mph

2-3
6059.8 rpm @ 56 mph

3-4
5878.3 rpm @ 82 mph

4-5
6047.5 rpm @ 112 mph

enjoy!

PS.
I would love to have a stock pi car dyno sheet to compare my dyno sheet too to see if its the same shape. Or if anyone wants me to use their data to find their personal shift points i'd be happy to help. Id need a dyno sheet, your transmission gear ratios, your final drive ratio and tire size.

 

To me it seems as if you would want to shift just about half way between the peak and the intersection. If you would shift where you are suggesting, where the lines intersect, your torque is dropping off soon after you shift. Sure, if you shiftin between the peak and the intersection, you'll be going back below peak torque. But youre also spending a lot more time at less than peak torque if you dont shift until they intersect. Also, here is cliffyk's dyno sheet from his website...

http://paladinmicro.com/PalMFrame00.htm?wks=dno

What I suggested has always resulted in the most felt shifts and the ones that put me back in my seat the most. It seems like even though you are using torque in your setup, that you are shifting based off of a horsepower curve.

Im curious to hear other's thoughts on this. How did it go over on modded mustangs?

 

Not necessarily, If you shift anywhere but the intersection areas follow straight down to the next gear and you are putting a lot less torque to the rear wheels.

EX:
1-2 gear is the most important, if you shift at 4krpm (peak) then you are dropped straight down to 2nd gear where you have a lot less torque to the rear wheels untill you meet the intersection where 2nd gear becomes optimal.

 

Cool graph, but wtf am I supposed to be looking at

 

I edited the post after you must have hit quote, because I wanted to change what I had and justify it a little more.

 

it went over well at modded mustangs (test area :P)

THe shifts mentioned are at the intersections, sorry I should have explained the graph a bit better. I only showed engine rpms up to 6500 rpm to show how the power drops off past the shift points.

THe graph is a comparison between torque at the rear wheels in each gear and speed of the car. The shift points are the intersection areas where torque at the rear wheels (rwtq*gear*final) is max at all times when using the intersection points as shift points.

 

What I think you need to do is find where the average torque in each gear is and place the shift at that point past the curve. Honestly, for a stock longblock car, the shift points should not be anywhere near 6000rpm. They should be in the mid5000's. Peak power is at around 5000rpm, peak torque at around 4250rpm. If you would look at a dyno graph, like the one linked to cliffy's, the curve, if shifted at 5200rpm, maintains over 250rwtq after the shift. If you hold it to 6200rpm for the shift, you're throwing it down towards where peak torque is, but you're also spending a lot of time not pulling as hard as you could from spinning it up to where torque is not made. If you shift at 6200rpm, the torque drops to around 200rwtq at that point, which means its dropping off power and torque. You never want to shift that far past your peaks. you always want to shift a little past your peaks, but not 1200rpm past, no matter what gear youre in.

This is just what makes sense to me. Hopefully i'm wrong and your calculator kicks my butt.

 

Yours engine torque drops down at 6200 to under 200rwtq yes, but remember that that is not what the wheels or the driving force is acting through. Basically you shift when your engine torque is 3.37/2=1.685 or 68.5% more engine torque from where you shift to were it ends up. The 1st to 2nd is high mostly because the gears are far apart thus the window of were you initially shift to were it lands.

1-2
1.685

2-3
1.5

3-4
1.333

4-5
1.4925

so shifting at 6200 when the torque is 190 or something the shift will land before your peak torque ~3680 rpm where your torque at the engine is higher by 68.5% (1st gear/2nd gear)

tq*gear1*final=tq*gear2*final

 

I dont have a dyno sheet, but i have 4:10 gears, headers, exhaust, Diablo sport tune, tires are 255/35/17 and Im trying to figure out what would be good shift points for my 2007 v6 so im not loosing any torq.

 

Nice work, although at my age my butt does a pretty good job of letting me know when to shift.

I haven't used MatLab for years, looked in to purchasing it a while back but it was too rich for my semi-retired budget...