Has anyone thought about this? I think that it can be done, although I could be wrong. I really don't know that much about dyno's. what I do know:

1. a dyno is part computer software

2. a dyno is part computer hardware

3. a car turns some kind of large wheel with the wheels it has power to.

4. the computer part measures changes in the mechanical part.

5. when the car is mounted on the dyno, 2 things happen:

A: the weight on the flywheels changes

B: the rate at which the flywheel turns changes.

I know I am probably missing something here(for sure) But I know that measuring weight and rpm's can't be too hard. Maybe the weight of the flywheel has something to do with it.

Anyways. Homemade dyno sounds like it might be something to pursue. Just think.... free dyno whenever you want it.

 

I'd like to know what you are going to make the rollers out of. They are incredibly heavy. Plus you have to be able to put a load on them to simulate the car driving down the road

 

they have water cooled brake systems in them.. well the ones i have seen do anyway

 

But how much of that is really necessary?

I think I could make the rollers out of some tires and a very strong rod/stand

Put the front of the car on ramps, jack up the other end of the car, then let the rear wheels rest on the tires. as for weight, I am sure there are ways to add lots of it.

 

I was sriously thinking of making an engine dyno in my basement. In theory it should be easy... all a dyno is is an applied load, and a way of measuring it. All I'd need is a brake, a lever working off the brake, and a scale, that way I could directly measure torque and then calculate HP. That's all dynos do, really... the problem is the brake - wherther you use a friction brake or an electric brake or a water brake you have a LOT of energy that needs to be dissipated safely.

Dynojets work differently than the rest of the dyno world - they are inertial chassis dynos. The car itself, not just the engine, is put up on a big roller with a known mass and therefore a known polar inertia, and the computer calculates the engine's power by carefully measuring the acceleration against this known load.

Whther to easure power/torque at discrete intervals (like a most engine dynos) or under constant acceleration (like a Dynojet) is a whole nother argument. Likewise the whole chassis dyno vs. engine dyno debate. (I personally don't like chassis dynos because they are easy to trick the readings high or low, but i like the acceleration testing vs. discrete intervals)