Re: VWvortex R32 Stock Dyno Numbers (Daemon42)
| Quote, originally posted by Daemon42 » |
So there's two ways to look at this.
1. It's a quirk, perhaps even a flaw of the Autopilot mode of the dyno because of the way
it tries to match speed at all 4 wheels at once. Might see if one of the other modes
would avoid the issue. They'd undoubtably see the same thing running say.. an
AWD Porsche with its viscous coupling to the front (no center diff), and it could
be even worse because a VC always allows *some* significant slippage (has to, to keep
the fluid hot and viscous). There may be a setting to allow a larger rpm difference.
Make sure the R32 uses it.
2. Or one could argue that this test represents a more accurate representation
of what Haldex is really doing in the real world switching between AWD and FWD
as it speeds up. Except that it most certainly is not running a 50/50 split in 4th gear
on the highway if you roll on the throttle at 2000 rpms. Too much traction at the front
for that, no rpm difference between input and output shafts at the Haldex unit, thus
no significant rearward torque transfer. 1st and 2nd gear.. yes.. 4th.. no
That's my take at least. Might want to run it by the dyno operators, or even Dynopack
and see what they think. Just tell em that Haldex needs a small rpm difference
between input and output to maintain lock, and see what they say. |
Ian, thanks for the insight and here are a few thoughts.
Number 2 of the above is whole heartedly more realistic in terms of measuring real-world horsepower to the wheels wouldn't you say? On the street you're not going to have a difference in rotational speed unless you were turning or on a slippery surface (wet, ice, snow, etc., etc.). So unless we're looking to try and force the haldex system to stay at a 50/50 split (assuming we can do that) and do a dyno run to determine what kind of power the car puts down when there is no traction front and rear (a "snow" dyno run so-to-speak), our current scenario more accurately captures what happens in real life. Even then it isn't exactly capturing that either as most of us that might accelerate hard on a dry surface *might* get some power transfer to the rears as the fronts break traction, but as the car continues to accelerate it won't have a speed difference between front and rear and will move power back to the front wheels. Essentially that's what happens on this dyno. At the start of the run, the Haldex electronics see a "slip" condition of the front wheels spinning at a different rate than the rears and immediately locks up the clutch pack so that all four wheels are now spinning. From there it is like you said, the amount of difference in rotational speeds reduces and the Haldex correspondingly reduces the amount of lockup on the clutch pack.
So the new questions are:
1. Does the amount of driveline loss through a Haldex system change as the amount of rotational speed differences decrease or increase? Short answer is yes, as there will obviously be more work involved in turning a full load rear wheels vs. a partial load rear wheels. So...
2. What is the amount of parasitic loss (or what is the scale percentage) when the Haldex system is not seeing a difference in rotational speeds ("at rest" if you will).
I have some contacts at Haldex and will try and see if there is any way to clarify this but I don't expect answers to be easy to get as I'm sure the individual programming for VW and Audi is proprietary and different than say, Volvo which uses different programming. We'll see.
Either way, we have some plausible explaination for what is going on and in the end the numbers are just what they are. Again, trying to extrapolate driveline loss would be near impossible and quite frankly not important to us here as we want to now measure differences we see when we make modifications to the engine.
And who says you can't find a good discussion on Vortex anymore?
-jamie
