Thread: Optimum shift points from dyno graph

I heard a rumor somewhere along my racing career that reading a dyno graph will tell you the optimum time to shift your car. Do you want to shift at the peak hp/rpm value before the line falls off or the plain standard red line does the best trick for fastest quarter times? I never used a dyno sheet before just my butt dyno.

I had an entire explanation of this with pictures I made and everything. Let me see if I can find it. It boils down to area under the curve and maximizing integrated HP. Best shift point is always past the power peak. Hang on...

I found the pics but not the explanation.
I'll post them and explain later.

Short answer is yes. Real answer is longer. Basically you try to shift a the point of diminishing returns.

My guy/guru coached me with dyno graphs and 1/4 runs and with the shift light it helped a lot.

On a real track with turns it gets much more complicated.

Explain away I'll try to keep up with you guys. Be prepared to be harrased with questions. lol

Just get the dyno graph and ask the operator this; "Where should I shift"?

Then experiment and see what gets the best times. A dyno chart only tells you so much but its a great tool. BenFenner will fill you in on the theoretical side and then marry it up with the practical.

On a twisty course you might choose a different gear because it will allow you to position yourself better for the next turn or series of turns. That's introducing more variables. For now let's just look at this assuming you are traveling in a straight line... like you said in the op.

In my understanding, You want to shift at a point after the peak that will bring you to the same WHP rating.

In the pic i just made a hypothetical line to where you should shift. This makes the most sense to me so you are not going to far under the curve after you shift to where you are not falling off more than you need to.

Here is a dyno graph for a typical 4 cylinder engine with a high redline and a peaky output (it could be an SR20, or some Honda motor, whatever).


Peak horsepower is at 7,000 rpm.

In this example when shifted, the transmission loses 2,000 rpm. Meaning, when you shift at redline (8,000 rpm) from 2nd you will end up at 6,000 rpm in 3rd.

If you shift at peak horsepower you will drop down about 2,000 rpm and end up at 5,000 rpm for the next gear. The amount of power you have to take advantage of is highlighted in pink.


The average amount of power you're using is also shown, and is about 200 HP.

If the engine is held to redline and then shifted, you're spending your time from 6,000 rpm to 8,000 rpm. Here is what that looks like in pink.


As you can see, the area under the curve is larger, and on average you're outputting more horsepower throughout the gear. The average power is marked and is about 220 HP.
Shifting at redline in this example is the best shift point. Certainly better than shifting at power peak. You want the power peak to be somewhere between your shift points if possible.


Shifting at redline isn't always the best thing to do. Here is a dynp graph for a typical 4 cylinder engine that puts torque output as a priority over power. Think about maybe an N/A Volkswagen engine.


Assuming the same 2,000 rpm drop between shifts, shifting at redline in this engine is a bad idea. You will be utilizing the power band from 4,000 rpm to 6,000 rpm and the area under that curve. The average power there is about 200 HP. If instead you shift at 4,500 rpm you will fall back to 2,500 rpm (right before power peak). This will include the power peak in your acceleration run and the power average will be about 220 HP causing greater acceleration throughout the gear.

Basically you need to find out how much of an rpm drop you get in each gear, and pick shift points that maximize the area under the curve. The real way to do this is with calculus and integrating the curve to get the proper area, ut you can usually do a great job by eye.

I know I didn't do a great job of explaining it, but ask any questions and I'll have the answers.

Oh dang I hate integrals. Now I'm extra glad that my mechie tells me where to shift. Nice explanation.

Don't worry, those are straightforward if you just fit a polynomial to the dyno data.