Engine Ramp Time on Dyno

All I seem to read about is the head line HP/KW that a engine hits when in fact its more about area under the curve. To explain my point, a number of years ago Honda with a LMP engine decided that what counted more then hp was the time it took the engine to rev up when under load. It was a V8 LMP2 engine in 2007.

They had the engine loaded up on the dyno part throttle running at 6.500 rpm they then opened the throttle fully letting the engine rev as the dyno loading was constant. They then timed how long it took for the engine to get from 6,500 rpm to 10,000 rpm, in this test they played with intake diameter etc. During the test the fastest time to go from 6,500 to 10,000 rpm was 3.9 seconds the slowest was 4.9 sec (best as we can see). If you think how much of a effect this would have on a cars lap time it adds up.

So my question Andre is have you seen testing like this done?

But on the track (particularly an Prototype), the load isn't constant, as the aero drag varies with the square of the airspeed. That almost looks like the results you would see from an inertial dyno, so the package that makes more average torque, will have a shorter run. It would be interesting to see the same configurations in the traditional dyno output graph showing torque vs. RPM.

Notice the time scale starts at 2.4s @ 6500 RPM, so it really took (3.9-2.4) 1.5s to go from 6500 to 10000 RPM. I think the smaller intake plenum diameter moved the torque peak to a lower RPM resulting in faster acceleration at that RPM.

Hi David,

Good point I didn't ever look at the start point I had assumed it started at Zero dumb me. So that is even more important then I had assumed. I agree the load is not constant but I think the way Honda do it is pretty smart. We just changed the the ITB on a Toyota 1.8l from the ones it came with to ATPower and different ram pipes. The result were amazing in that we got a 15% average increase in power from 5,000 rpm to 8,000 rpm with no increase in total KW. On the track the car was a totally different car, pulling out of the corners faster then cars with 30% more power and a similar weight.

Do you have a more detailed source document on the procedure? As far as I know, Honda racing uses very expensive AVL electric engine dynos and AVL Puma automation systems which is way beyond the resources of most people. You need to write an automation script to do that kind of test.

For example, you can simulate a transmission, driver, and road load with AVL Puma using their ISAC system if you enter the gear ratios & shift schedule, weight of the vehicle and the road load coefficients (a quadratic function that takes into account friction and drag). So their test automation script might be set to ramp the engine to 6500 and some target torque with a PID loop, or just open loop to a fixed accelerator pedal position. Then maybe it changes into vehicle simulation mode simulating a fixed gear ratio and a vehicle speed ramp. It is literally millions of dollars worth of resources (test cell, dyno, automation system) to do something that sophisticated.

I agree that is an interesting way to present the data since you want a racecar to accelerate quickly. Perhaps their team wanted to show they had made improvements without sharing exact power figures, or perhaps they were trying to show that more mid-range power might be more useful than more high-RPM power for their application. I'd also be interested to know what those different engine configurations looked like on a traditional dyno chart showing power (and torque) vs RPM.

I'm not surprised to hear that your engine feels faster when making more power (and torque) between 5000-8000RPM compared to the old setup with the same max power but less mid-range power (and torque). Somehow this reminds me of the 'Tractive Force' article by Alexi from OnPoint Dyno. https://www.onpointdyno.com/understanding-torque-and-horsepower-lets-try-this-again-with-tractive-force/ , it's worth a read if you haven't seen it.

Hi Guys I have attached the doc,

Yes I have read OnPoint Dyno, I also emailed him to discuss the R&D I have done on intakes and why his 350Z car failed to develop the power he expect but he never replied. So I don't waste my time trying a second time. AS you can see in the video he doesn't know why he lost the power after all the work.

You don't need a special rig to do the Honda test you just need a clever dyno that has a mod and some smarts. Even set up the engine ECU to do a fair bit of the work for you.

https://www.youtube.com/watch?v=9u4hrFG4N70&t=481s