DIYEFI.org Forum

Not sure if this goes here or if this has been brought up before. I saw a brief mention of it in my search of "dyno" in the forum.

I've seen a lot of talk about building a dyno...some about a steady state and about an inertial type. I think it would be pretty easy to simulate an inertial dyno using a flat road and logging of RPM and speed data. You have a fixed mass for the driver + car. The change in speed/time is the acceleration which is proportional to the torque at that RPM. From the torque and RPM you can get the HP at that RPM.

Caveats: I would not use this to calculate an absolute HP ("My car just made 3000 HP!!") because you'd have to accurately weigh the car and drive on a perfectly flat road. Also, it would be a bit different that a regular dyno since you'd also have to overcome the rotational and frictional inertia of the non powered wheels....and the wind resistance.

You could, however, compare how changes in settings/mapping affected the RPM-torque curve. You'd have to drive the same section of road in the same gear.

I think it would be of benefit. I asked at one place and the guy wanted $200 for 3 dyno pulls. Ouch.

Comments?

do a run in both directions and null out wind ?

Good idea. You could do a run in both directions to null out any slope of the road and wind speed. You'd still have wind resistance, but that would just mean the higher RPMs are producing relatively more torque than calculated compared with lower RPMS. The lower the gear used, the lower the speed and less wind resistance you'd have. You'd also have a shorter sample period.

Firstly, for tuning, the road or a track is fine. Dynos are convenient, but not necessary.

Secondly you can get a gtech-pro to do this for you OOTB. It's not difficult math to do regardless, but those aren't that expensive.

Thirdly, you can get phone apps that do this too these days. If you have a smart phone or tablet with accelerometer, you're covered.

Mind if I move this somewhere more generic?

Fred.

I'll have to look into how you tune it on the road. If dynos are convenient for tuning, it still seems like a good idea to make tuning more efficient.

G-tech looks pretty nice, but looks like they are doing the RPM/torque/HP curve with just RPM/weight and GPS position (could just just RPM/gear/tire size to get distance). FreeEMS could certainly do it too if not just from data logging. G-tech does a bunch of other cool stuff too especially if you want to race.

I don't know if the apps have RPM sensor data, but they are a very cheap way to go if you have a GPS smart phone.

Yes, feel free to move it.

Andreas wrote:I'll have to look into how you tune it on the road. If dynos are convenient for tuning, it still seems like a good idea to make tuning more efficient.

Depends on whether it's a chore or a pleasure ;-) Most find it the latter. A labour of love.

Good point re RPM. The Gtech gets its RPM from ripple in the power supply via calibration. Not hugely accurate, but good enough, and pretty realistic.

Andreas wrote:If dynos are convenient for tuning

A dyno is useful for holding the engine at a certain load/speed. The numbers it produces are for masturbatory purposes.

Just do a datalog and compare the slope of the RPM data ...