Intake Testing

[BeaverGuy]

I was wondering how the teams out there test the throttle response of their engine. I was thinking that setting the load on the dyno then opening the throttle and checking how long it takes for the engine to reach max speed.

Also, how do you test to make sure that airflow is even to all the cylinders. I was thinking of using airflow meters or pressure transducers in the intake runners.

Thanks for any suggestions.

[BeaverGuy]

I was wondering how the teams out there test the throttle response of their engine. I was thinking that setting the load on the dyno then opening the throttle and checking how long it takes for the engine to reach max speed.

Also, how do you test to make sure that airflow is even to all the cylinders. I was thinking of using airflow meters or pressure transducers in the intake runners.

Thanks for any suggestions.

[BryanH]

The 2nd one is easier! Motec test inlet manifold air distribution by doing a dyno run with a PLM in each exhaust primary pipe.
An 18x1.5 bung is needed in each primary.
If you don't have 4 lambda meters, do 4 power runs and compare each lambda graph. The difference is more or less the exactly the diff in air flow btwn each runner. Mark told me that they have seen large (25%) variations with custom made inlets, even ones that looked "good". I personally have had trouble convincing people to mess up their headers to prove that their all singing/dancing custom inlet manifold is for the bin.
Transient acceleration Q is big can of worms.
Bryan H

[John Bucknell]

If you have a pressure transducer, you can get an excellent look at throttle response. Ultimately, the torque response is a function of the manifold pressure. If you can run the engine speed up, close the throttle and open it again when speed drops to an appropriate rpm (for non-motoring dynos) and record time it takes to get to a steady state pressure - you've established throttle response. What's a good number is also debatable.

Manifold distribution is a toughy on the dyno. Better to set your head up on a flow bench and measure each runner. A span of 5% is about as good as anything with a single throttle.

[BeaverGuy]

The problem I see with testing the intake manifold on a flow bench is it neglects the pulsed nature of the flow. It also neglects the interference of cylinders firing sequentialy from neighboring runners.

An intersting idea on the last bit is that it might be possible to equalize the flow in a log type manifold by rearagning runner order.

[John Bucknell]

Keeping in mind that the momentum effects and pressure wave effects are discrete in a single throttle multi-cylinder. You can isolate the momentum effects on a flow bench - and my experience is that momentum is the dominating factor (given proper attention to the plenum so that all the runners have a decent enough cross-section change to initiate the pressure wave reflection). Your idea to use MAF meters in the runner will impact both the momentum and pressure wave effects - they are extremely intrusive.

My above statement about measuring distribution on the dyno is based on the fact that I have access to big dollars worth of test equipment and it works by sampling exhaust gas in every runner - which you can do with a Wide Range O2 in every runner, but even that is too expensive for us OEMs to do regularly. EGTs via a thermocouple aren't sufficiently accurate....