Well I have read many thoughts on how to determine optimum shift points, most of which I am sure are wrong. I recently had my car dynoed & I am confident the following procedure is correct. It is based on the fact you want to have maximum accelerating force at the rear wheels at all times. It is further based on the fact if you use for example a 2:1 gear ratio you double the accelerating force.

Step 1 is to normalize the data from your torque curve. Normalize means to take the torque reading at various rpms and divide it by the maximum torque. I did mine for every 500 rpm, that is 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6200, & 6400.

Step 2 is to determine the relative torque for each gear. So for my 6 speed I multipled the result in STep 1 by 2.68 for 1st gear, 1.80 for 2nd gear, 1.29 for 3rd gear, and 1.00 for 4th gear.

Step 3 is to convert the "torque" vs RPM to "torque" vs MPH for each gear. I did this by first calculating tire rpm from motor rpm. So in my case for 1st gear I took motor rpm & divided it by 2.68(transmission ratio) and took this result and divided it by 4.10(rear end ratio) which gives you tire revolutions per minute. Then take this result and multiply it by 60 to get tire revolutions per hour. For my tire size it takes 801 revolutions to go 1 mile, so I took the tire revolutions per hour and divided it by 801 to get mph. I did this for the 1st four gears.

Step 4 is to graph the results by plotting accelerating force vs mph for each gear.

Step 5 is to determine shift points by looking at which gear gives the maximum accelerating force "torque" at each speed.

So the shift points are determined by the rear wheel torque curve AND the ratios in the transmission.

This might sound complicated but it really isn't, especially with the help of a spread sheet. I wished I knew how to post my results since they clear things.

Well this is a long post, but I thought it would be of interest. I would be glad to help anyone that is interested in using this method to evaluate there car.

 

Example of the method (simplified a little)

When you shift a 6-speed from 1st(2.68:1) to 2nd(1.80:1) two important things happen. First the engine rpm will drop by ratio of 1.80/2.68 (0.67). Second the torque multiplication will drop by the same ratio.

1. For each rpm in 1st gear look at the torque from the torque curve.
2. Multiply the RPM by 1.8/2.68 to get the RPM you will be at in 2nd gear.
3. Determine the torque for this rpm from the torque curve
4. Multiply the new torque by 1.8/2.68, to get the relative torque in 2nd gear.
5. Determine which torque is higher. If it is higher in 2nd then shift.

Example:

1. Torque at 6300 is 235 ft-lbs.
2. If I shift at 6300 RPM then engine speed becomes 6300*1.8/2.68=4231
3. Torque at 4231 rpm is about 305 ft-lbs from torque curve
4. Equivalent torque in 2nd is 305*1.8/2.68=204.9
5. Would be better not to shift since 235 is greater than 204.9.

So for my LT4 the best 1-2 shift for performance is actually above redline of 6300!!

 

We give this calculator out for free to HeaderDesign.com members (Membership is free, too). It is quite revealing, and will help you argue against the BS. If you want a copy of the spreadsheet, just send me an Email and tell me what version of Excel you are using.

 

The gist of the matter is to shift when torque of the new gear exceeds torque of the old gear. Maximizing the area under the composite torque curve of the gears. Basic Integral Calculus.

For an engine with a peaky torque curve, shifting so the new gear is at the torque peak gets fairly close if the gear ratios are not too wide. You could use this method on C3 Vettes. Problem is that with a wide torque curve, like on the C4 Vette, and wider gear ratios, you often need to shift at a higher RPM to maximize the area under the curve.

Curious to know, on the strip, what's the typical difference on a stock Vette?

Nice calculations!

 

tkrussell
You have got the right idea when you say maximize area under the rear wheel torque curve. Also the concept of shifting when torque of the new gear equals torque of old gear.

For my LT4 the torque was nearly constant at 300-310 ft-lbs up to 5000 rpm and then fell off. So another way to look at my shift points using your description is:

1. Shift from 1st to 2nd at the rpm when torque falls to 305*1.80/2.68=204.8
2. Shift from 2nd to 3rd at the rpm when torque falls to 305*1.29/1.80=218.6
3. Shift from 3rd to 4th at the rpm when torque falls to 305*1.00/1.29=236.4

My torque fell to 235 at 6300 rpm, so my best shift points are at redline unless I want to over rev the engine. Actually right at redline is optimum for the 3-4 shift since 236.4>235.

Also notice that the shift point is different for each of the 3 shifts!!!!!!

Peaky torque curves don't change the concept of maximizing rear wheel accelerating force at all speeds. If you have a dyno torque curve and know your transmission ratios, then you should still use the basic method of comparing torque in the new gear vs old gear at every mph to determine shift point. I like the method of first graphing or charting relative torque of each gear vs mph. Shift points just fall in place then.

I would be glad to try an assist anyone with torque curve to determine shift points.

Thanks for you interest & reply!!

 

I use my "Mr Dyno" accelerometer for this. I download data from a couple of 0-60 runs. Then I just look at acceleration, not torque. I plot 2 runs - an early shift, and a late shift run. As soon as my 1st gear acceleration curve drops to below the second gear acceleration curve, the precise optimum shift point is defined. It is the greatest toy (and tool) I ever bought! Check the web site:

http://www.mrdyno.com/

George

 

Bud, for what it is worth, I have tried shifting my car at differing points to get the best ETs and beleive I have found that it I get a slightly better ET making the 2-3 shift at a little higher RPM than the 1-2 shift. Car seems like it wants to be shifted from 1-2 at about 6,100+- and from 2-3 at 6,300+-. I do not hit the limiter at 6,300 so the tach is either not keeping up with the actual RPM or (if I recall) the custom program raised the limiter to 6,500.

By the way, all of this is somewhat guesswork on my part as I make no claims to being an incredibly consistent driver with the 6-speed. I had literally 1,000s of runs in my old car with a manual valve body auto, but the 6-speed is a different animal all together!




[Modified by LT-4 CE, 9:03 PM 10/29/2003]

 

This doesn't sound quite right for some reason. If torque at 6300 is 235 ft-lbs, then torque to the wheels would be 235*2.68*4.10=2582. If you then shift to 2nd, torque to the wheels becomes 305*1.8*4.10=2251, which is less than 2582. Same conclusion, though.

 

85DAVE

You are right, but...

What is most important is to determine which gear gives the most accelerating torque at the rear wheels. The actual value isn't necessary to do this.

I have not attempted to calculate the actual acceleration etc. So I was just trying to simplify things as much as possible, keep the numbers small & as easy to work with and interpret.

That is why in my initial description of this method I "normalized" the torque values. Makes it a little easier to see on a percentage basis what is going on.

Appreciate the time you spent evaluating the method. I hope it helped you.

 

George West

That Mr Dyno looks real good. Not sure how it compares to a Gtech. But you have got the right idea, which is to maximize acceleration.

If I had a dyno curve I would still go thru the method I have suggested and then fine tune your shift points with Mr Dyno. I am not sure how wheel spin might affect you interpretation of the g meter readings. I am thinking very seriously of either a Mr Dyno or a Gtech.

LT4 CE

The red line is 6300, but the factory rev limiter kicks in at 6400.

I would expect you to get your best times shifting at 6300 or higher based on the dyno curve, gear ratios etc. I am not sure how to explain your better performance by shifting at 6100, perhaps as you said your tach is lagging a little. Or maybe when you tried shifting at 6300 you did bump the limiter a little. In retrospect, I am almost sure that hitting the rev limiter was one of my many problems at the track.