Melbourne Dyno Day - Saturday April 9th Results:

[DNA Tuning]

The new GTI is very well tuned from the factory to give an amazing linear power delivery - aftermarket tunes will remove some of this linearity and it will feel more like a turbo car, but obviously this is a small sacrifice in order to achieve a significantly higher level of performance. It all depends on what you're after I guess - if it were me, I'd keep it stock until I got bored of it, ha.

[Tim]

The Golf Rs that ran were in line with other dynos. Also the RS4 that ran pulled an almost identical figure to another RS4 at the previous dyno day on a different dyno. So im not sure all the awd figures were low.
I was quite surprised at the consistency between the two different dynos to be honest.

[kaanage]

And paul_d's R32 pulled just over 140kW which was a fair bit more than he was expecting

Gambit, your Polo and the stock Golf TSi had torque "curves" that were so flat they looked like electric motors - I couldn't see any trace of the changeover from the supercharger to the turbo.
No wonder some people think they feel boring since the power delivery is so linear.

[kaanage]

nice power figure Scotty... Im looking forward to getting my thing tuned again

Wait until you see his plot vs mine - especially the torque curve. It explains why my PoD pulled slowly but steadily away from his when we did a rolling 4th gear drag from 80 => 110 km/h (he reckons he weighs more but I've got more crap in my car)
It ain't all about peak power

[DNA Tuning]

Tim, that's fair enough I guess; what I meant was the AWD numbers seemed low relative to the front drives, the absolute figure itself doesn't matter too much I guess, but essentially I don't think the values we saw of the AWD cars are indicative of their actual performance; ie. as amazing as deancorp's gti pirelli is - I don't think it could match, let alone beat a B7 RS4, even from a rolling start (so as to remove the AWD grip advantage from the equation), and yet obtained 30kW higher numbers at the wheels.

Greg, as for the VW haldex cars, I believe they tend to produce higher numbers than torsen quattro cars, not only because there is less mechanical drivetrain losses, but also because on a rolling road situation, more torque is being sent to the front than the rear - as such you'll find the figures are closer to that of the front drives. Of course this is still speculation, but considering the B7 RS4 which has a default torque distribution of 60% rear and 40% front under non-slip situations (like on a rolling road) and quite frequently only pulls 190's on dyno dynamics rolling roads shows that this hypothesis may be true to some extent!

[Tim]

That makes perfect sense but then thats normal. Drivetrain losses through some systems are greater no doubt. Thats just the way of the world. At the end of the day we are there to compare our cars against similar one and to see if the modifications we are making are heading in the right direction.
If it was all about who has the quickest car I can think of a certain R35 GTR that put us all to shame!

[kaanage]

Possibly but paul_d has a haldex with rear torque bias, not the stock one. If the dyno is summing the power from the front and rear rollers, I don't see how the bias makes much difference, though.
I'd still lean towards the Torsen center diff losses being the culprit - they way that they (and the Quaife) work has to eat up a fair bit of power.

But peak power isn't everything - the delivery makes more difference unless doing top speed comparisons
The rolling drag I did with Scotty, in our identical apart from tune PoDs, showed that - he would have rolled over me as we got nearer top speed but I got the jump due to the initial torque (and hence power) advantage.

[VW Central]

Here is a video of the beast of the day - Nissan GTR 35

[YOUTUBE]uchtLa1AWOE[/YOUTUBE]

[DNA Tuning]

Possibly but paul_d has a haldex with rear torque bias, not the stock one. If the dyno is summing the power from the front and rear rollers, I don't see how the bias makes much difference, though.
I'd still lean towards the Torsen center diff losses being the culprit - they way that they (and the Quaife) work has to eat up a fair bit of power.

Dyno Dynamics - Specifications

Having hands on experience with implementing and controlling electromagnetic dynamometers at at uni; I can say it would be very difficult to "sum" up torque between front and rear rollers on an AWD car. Essentially, the front and rear rollers both have adjustable loads; you could pretty much load up the rollers hard enough to gradually slow down a car from peak power down to stalling - this is why electromagnetic dynamometers are great for tuning as you can hold steady-state points (ie. you can set the rotational speed of the dyno, and this means while the car is on the dyno, no matter how hard you floor it, it will not rise in RPM; this allows tuners to accurately monitor the AFR, timing, etc. at a steady state condition and make changes as appropriate), etc.

That said, I am sure many tuners still just make some changes and then do a full run again rather than focusing on steady state point tuning.

Anyway, before I get too carried away with explaining the physics/engineering of the system, the point I'm trying to make is, the dyno needs to be swept through an entire range of speeds in order to give the torque/power readout. What the controllers will attempt to do is adjust the loads on the 2 rollers so that the speed of the front and rear rollers can be kept as close as possible.

If for instance you take an old school 4x4 with a locked centre diff and chuck it on the dyno - load balancing from the dyno controller would not need to take place as the speed is automatically matched between the front and rear rollers.

On a conventional haldex system (ie. prior to the new golf R), the front rollers start rotating first; immediately, depending on the configuration of the dyno, the dyno's controller will increase the magnetic loads on the front rollers to provide resistance (otherwise the wheels will just freely rotate, almost as though you are revving in neutral), whilst keeping the rear roller load as low as possible as it has not begun rotating yet. The haldex controller will sense the "slip" and begin to transfer some of the speed-differential (which acts as a hydraulic pump) to the rear, allowing the rear wheels to start applying torque. However, not much torque needs to be transferred to the rear in order to keep the rear rollers rotating as fast as the front rollers because the dyno controller is working in tandem with the haldex controller; only enough torque to overcome the rotating inertia of the rear rollers is required. The end result is that very little torque is being transferred to the rear.

Now, if we take the new Golf R, which has an electric hydraulic pump to constantly provide some torque transfer to the rear, both rollers should begin rotating immediately. The balance between the front and rear roller loads throughout the rest of the run will then depend on the haldex controller settings, and, even then, will be ultimately limited by how powerful the electric hydraulic pump is.

Yes, a haldex system can provide 100% torque to rear, but only when sufficient hydraulic pressure is achieved; in traditional haldex systems this is only possible when the front wheels spin faster than the rear; the haldex controllers simply change the how much of this hydraulic pressure is applied to the rear - but without hydraulic pressure in the first place, it still can't magically apply 100% of torque to the rear.

On the other hand, a torsen quattro system works very differently as it operates purely off torque, and not speed. This means that both rollers will start spinning at the same time, and the torque ratio should not change throughout the run. At the end of the day, the more torque that needs to be diverted to the rear, the more drivetrain losses will occur.

I'm not very good at explaining things, which is probably why I'm an engineer and not a teacher, but I'd be more than happy to have a chat about it if you'd like!

[kaanage]

For sure. Engineer to engineer - it'll be a real nerd session

[Rsome]

Here is a video of the beast of the day - Nissan GTR 35

[YOUTUBE]uchtLa1AWOE[/YOUTUBE]

B***dy hell, the sound that thing makes when it's angry!

Makes my R sound positively anemic

[VW Central]

yup, if only i could afford one of these...

[Rsome]

yup, if only i could afford one of these...

Mate, one consolation is that we can buy three Rs for the price of one of them

[Gambit]

And paul_d's R32 pulled just over 140kW which was a fair bit more than he was expecting

Gambit, your Polo and the stock Golf TSi had torque "curves" that were so flat they looked like electric motors - I couldn't see any trace of the changeover from the supercharger to the turbo.
No wonder some people think they feel boring since the power delivery is so linear.

Yeah i can imagine that some people would feel that way, however for me the linear power delivery of these cars is a revelation and one in which makes the car and engine combination a fantastic package.

[Water Boy]

Remember kiddies your'e lucky I didnt put the fire breathing CL on the rollers again. (Though I do think id have to turn up my confibulator to pull the numbers of the r35)