Ratio Selection/Torque curves

[Magpie]

Following on from a post in a build thread, wherein the merits of ratios was being discuused I thought it would be appropriate to start a new thread to collect thoughts/knowledge...

An engine produces a torque curve.

The gearbox/diff are mutipliers of that torque curve.

Tire radius has an effect on torque to the ground as well as acceleration

Gearbox/diff selection have an effect on torque to the ground, acceleration and speed.

Mass of the vehicle effects acceleration.

An OEM gearbox is (generally) designed so that when you change up a gear when accelerating, the next gear will be at the start of the engine's power band. This design provides maximum acceleration all the way up to the top speed of the car.

However, once a car is modified its torque curve could make the OEM gearbox a bottleneck. Therefore if we do not want to get a custom gear set our options are limited to changing the mass of the vehicle, changing wheel size or reshaping the engines torque curve.

If the car is used on the track then, we should shift when you will be at an RPM where the car will make more torque AFTER you shift than in your current gear. You cannot use speed for this calculation, speed is a result not an input!

The question is when is the best point to shift. To help I have updated an excel file that gives Torque, Speed and Acceleration for the chosen gearbox ratios and up to 3 different dif ratios. This way you can see the change for your self. However you will need to have a RPM/torque curve to make it work (mine is in there). Pink cells are inputs.

In my exmaple shifting by speed, to ensure that the revs don't drop below 5,500 (4.3 dif, 6 speed) means that I'm shifting before the car develops peak acceleration as well as gives me less torque in the next gear!

Now there are other factors that effect the car's acceleration like grip, resistance however this would further complicate this post. This grip and resistance needs their own threads.

[rascal]

What constitutes a cars 'power band'? Is it the rpm range from max torque through to max power?

To determine shift points, as you stated you would look at the figure to see where it crosses over such that the next gear has a bigger figure than the current gear. But do you use torque or power for the figure to use?

For example, I have a turbo golf as my DD. It's torque peak is at 2100rpm but its power peak is at 4900rpm.
The car accelerates much faster if you rev to just beyond 5000 and change than if you change just beyond 2000.
That would imply that you should use kw to determine shift points and not Nm?

Or is it just because I don't have a 20 speed gearbox, and if I did then I could stay close to the 2100rpm peak at all times and it would accelerate faster?

My feel it is the former (ie hp not torque for change point) cos after a gear change if the revs drop to 2100 it is slower to accel than if the revs only drop to 3500, which suggests being closest to peak power and not peak torque is the desired.

[oztrackdays]

My paraphrased definition is when trying to go fast ...

"You must change into the next gear up eg. 3rd - 4th when the car will now accelerate quicker in fourth than third."

what RPM range this falls into will vary per Engine Power chart and "Gearbox/Final Drive/Tyre" combination.

So it's a combination of DYNO chart and Ratios.

You can also measure acceleration very easily now with GPS data loggers so the answer can be fully defined.

What complicates matters is on a tight circuit sometimes the ultimate lap time is completed by holding a
car in gear despite it not accelerating well anymore as the next corner is close/tight.

[Dan]

what RPM range this falls into will vary per Engine Power chart and "Gearbox/Final Drive/Tyre" combination.

You are close but something that is important to understand is that the optimum shift point from an RPM perspective to maximise torque does not change by changing the Final Drive ratio or Tyre size, those things only change the speed you will be traveling when you shift.

The reason these things do not affect the shift point is because they are a constant.

Changing the Gear Ratio however will change the optimum shift point to maximise torque as gear ratio's change the RPM you will end up in on the next gear and therefore the torque the engine is making when you shift will change.

[MattR]

Attached is what I use to help with choosing gear ratios and speeds based on the variables that I would have at my disposal, gear ratio, final drive and tyre size.

As a start point it should have the standard MX5 ratios in it, and you can input custom ratios different final drive and different tyre sizes to see what effects this has on speed and rpm. It also allows input for maximum rpm used and rpm at max power to show the speeds at change points and max rpm as well as gives you the rpm in the next gear at the change point.

I find it very useful in deciding on obviously gear ratios but also the type of engine I want, ie lazy torque monster truck motor or a peaky screamer. you can also work backwards from the speed you want to achieve or work out best ratios for a given rpm for each track you race on.

I now have a useless databank for the Zed with the different tyres I used, slicks and wets, as well as for the different ratios I used and also experimenting with if it was worth spending big dollars on gearsets from Japan for it.

I also have a useful data bank for the S14 and plenty of options as the dollars stack up for providing the engine builder with some targets for the motor and what i need in terms of gearsets as the diff ratio is pretty much decided on.

[Magpie]

Thanks MattR, it is neater than my excel file

What I'm probably also trying to show is that there are a lot of other factors involved and it is not a simple case of selecting a dif ratio because it is what everybody else has. I did exactly that for the strut braces (which are for sale very soon).

The best lesson I have learnt from this is that weight plays a critical role (obvious) and if you can reduce this your acceleration will improve, with no change to the torque curve or ratios required! A shift point change calculator does not show this nuance, then again maybe some people don't really care or want to understand.

Acceleration (6,250 rpm, 1.646 gearbox, 4.1 dif)
1100 kg = 3.83 m/s2
1000 kg = 4.21 m/s2
_900 kg = 4.68 m/s2

The extra 200kg equates to 4.25m over 5 secs.

Acceleration (6,250 rpm, 1.646 gearbox, 4.3 dif)
1100 kg = 4.02 m/s2
1000 kg = 4.42 m/s2
_900 kg = 4.91 m/s2

The extra 200kg equates to 4.45m over 5 secs.

Weight does not change your top speed (at least for this discussion), but less weight means that you get to the top speed faster!

I had to be a nerd to prove the answer to the statement that it corners as quick as any other car but has no straight line speed. Simply it cannot accelerate out of a corner well because it is too 'heavy' for the selected gear and the RPM at the corner apex is not optimal for the torque curve. So at the corner apex I need to lighten the car or change the RPM point for that speed or select a different gear or a combination.