Issue re Dyno test for 4WD

Shane Hodges · Dec 5, 2005

Unfortunatley I (and a few other people) had a wasted journey to Weltmeisters a week ago where we were hoping to get our BHP measured. Their "state of the art" dyno doesn't do 4WD cars very well.

Their solution was for us to all come back another day when their Porsche technical chap would disconnect the prop to the front wheels so that a "proper" measurement could be made.

Now, I am no expert, but surely the ECU would get a little confused if it is getting no drive info from the front wheels? Wouldn't the warning lights start lighting up like a Christmas tree?

My feeling is not to go back and look for a proper 4WD rolling road - does anyone know of one?

Cheers,

Sundeep · Dec 5, 2005

ORIGINAL: Shane Hodges

My feeling is not to go back and look for a proper 4WD rolling road - does anyone know of one?

AmD down the road in Bicester have one.. works fine for 4wd et al....

oceancarrera · Dec 5, 2005

I just returned from a brilliant dyno day at Weltmeisters (aka WRC Technolgies as used by the bods at EVO) where they dyno'd my Carrera 4 together with 3 or 4 other 993 TTs and even a 996 C4S.

Because the 4wd system used in the 964/993/996 all use viscous couplings, power is not evenly split between the front/rear. Depending on your car, the 4wd typically sends anywhere between only 5% - 40% of the power to the front axle depending on the amount of slip that the system can sense. One of the key sensors the 4wd system uses is a G-force accelerometer to detect forward motion in the car. Being stuck motionless on rollers means that the G-force sensor never picks up forward motion, but the other sensors detect wheel movement. As a result, the 4wd shuffles torque between the front and rear axles in an inconsistent manner, thereby making accurate measurement of power split between the front and rear axles impossible.

Because a rolling roads rollers are stationary unless moved by the force of the wheels... the rollers of a 4 wheel rolling road would never be consistently accelerated by the wheels if the car constantly shifted the amount of torque between the axles. 4 wheel rolling roads work for the likes of Quattros and the Scoobarishis because the amount of torque split between the front and rear axles is fixed. As a result, these rolling roads can be pre-programmed to accurately account for the consistent split of torque between the front/rear of these cars.

The "intelligence" of the Porsche system, however, means that the only way to accurately ensure an accurate reading at either axle is to decouple the 4wd and let it run in 2wd mode.

Imho, walk away from any 4 wheel rolling road that allows you to put your 4wd Porsche on without first decoupling the drive shaft as the operators clearly do not know what they are talking about when it comes to dyno'ing Porsches.

oceancarrera · Dec 5, 2005

... and having had my car dyno'd at other highly "reputable" garages/outfits (no need to name names...) , and I can confirm that the guys at Weltmeister/WRC know what they are talking about when it comes to dyno'ing. Their equipment is state-of-the-art, and they have a reasonable reputation to upkeep/maintain as their site results are published every month in EVO!

Shane - don't get put off by them - they were saving your car and saving your money if their resident Porsche guy wasn't in.

And no, I don't work for them! Just a happy customer.

968er · Dec 5, 2005

Can back up what Adrien has said, you must uncouple the front and run it in 2WD.

Dave Wilkinson · Dec 6, 2005

Just to add a little clarification here, mentions above of 964/993/996 4WD grouped together is a little misleading since the 964 4WD system is TOTALLY different to the 993 viscous coupled system. The 964 C4 uses a purely mechanical centre diff and if you disconnect the driveshaft to the front you lose ALL motive force and you cannot even reverse the car off the ramp!!! Ask me how I know!!!

The 964 C4 can be tested perfectly well on the Weltmeister dyno but the viscous coupling in the 993 (and, I assume, the 996) causes a very inconsistent variation during the run and it becomes impossible to get an accurate reading. Disconnecting the front drive does eliminate that problem.

Regards

Dave

oceancarrera · Dec 6, 2005

ORIGINAL: Computamedic

Just to add a little clarification here, mentions above of 964/993/996 4WD grouped together is a little misleading since the 964 4WD system is TOTALLY different to the 993 viscous coupled system. The 964 C4 uses a purely mechanical centre diff and if you disconnect the driveshaft to the front you lose ALL motive force and you cannot even reverse the car off the ramp!!! Ask me how I know!!!

Thanks Dave for the clarification! Now I know where that extra 50kg in the 964 system went to...

Cheers -
Adrien

uktrucks · Dec 7, 2005

Thank you to Adrien for his concise description, saved me a long answer. The Porsche 993/996 are not 4wd in the truest sense, hence the potential for misbehaviour and differential damage.
We have now run over 550 4wd cars with up to 930 BHP on one example, with no issues at all. Please do not lay blame at the Dyno, certain cars just cannot be accomodated.

Cheers

Allan

Dave Wilkinson · Dec 7, 2005

ORIGINAL: MoC2S

So with the 964 C4, Dave, do you just run in normal road mode or some other setting? I know you can lock your diffs, but I thought that was a low speed only thing ...

cheers, Maurice [8|]

Maurice,

Just regular normal road mode - no changes, no messing!! The diff lock is, indeed, a low speed thing and it self cancels above 30mph or there abouts.

Just to pinch a little of Allan's reply above, does this mean that the 964 IS a real 4WD if the 993/996 aren't?? My instincts tell me to head for cover now!!!

Regards

Dave

Guest · Dec 7, 2005

I found the same thing at G-Force.
They tried the 996 C4S I used to have, but on the first run concluded the variable apportioning of torque meant the figures were meaningless.
To be fair they were not sure it would work, and stopped as soon as it was clear.
They were not sure about the C4S but knew the TTs could not be figured.

They explained that for all tuning work they do, the propshaft is disconnected.

I don't think you can say "real" or "not real" 4-wd.
The system is setup to give pretty much rear wheel drive handling when the wheels are not slipping, which is what you want, but to have the advantage of extra traction when you occaisionally need it.
It is 4-wd since all 4 wheels are driven, just that the amount of torque varies according to conditions.
Permanent fixed 4-wd is less flexible and changes the handling character more.

Shane Hodges · Dec 8, 2005

So disconnecting the prop seems to be the way to go for 993/996, but

1) why doesn't this confuse the electronics if they are registering that the front wheels are not turning? and

2) Is the "disconnection" done before the 95/5 split. ie how does the ECU know it is to deliver 100% or a fixed 95% or some other % to the rear wheels only and completely ignore the front?

Cheers,

Guest · Dec 8, 2005

The system is mechanical - it is a viscous coupling.
There is no electrickery going on to apportion the torque.
The 5% is arranged just through slightly different rolling circumference of the front/rear tyres.
The ECU does not know or care where the power is going.

The stability control does not care, since there is no yaw coming from the sensors.

I am not sure why the traction control does not care though, since it uses the ABS sensors and must notice there is some difference in the rotation of the various wheels. This is normally managed by applying the brakes to the wheel that is slipping (although it is setup to not do it too much to avoid overheating the brakes).
It can cut the power too, but to make it less intrusive the brakes are used first.

I cannot remember what you do on the rolling road, so it is possible you have to turn PSM off.

Guest · Dec 9, 2005

I had mine done at AMD in Bicester (pictures of car rolling on the 'road' at 150 mph can be found at www.porschepictures.net - go to Pictures and select 'TT')

Guest · Dec 9, 2005

all four wheels driven!

uktrucks · Dec 9, 2005

Hi All,

I do not post on here very often, but I just need to clarify a couple of points. The dyno used at Weltmeister/WRC can run Porsche vehicles in 4wd. There is a specific traction stability parameter in the software that means there is no need to remove the propshaft.

The problem is one of the transmission protection parameters built into the dyno has to be turned off. In Dyno Dynamics' words, " this means the Dyno can no longer protect the transmission from damage so you need to be carefull".

The dyno is run in various transmission configurations. If there is a problem with a front, rear, or centre differential in any configuration the dyno will not let the operator perform a power test, or apply load to the transmission - which will save the transmission system from damage if there is a faulty component. The electronic torque spilt, based on acceleration and load, that the Porsche delivers does not fit easily within one of the 8 dyno transmission configurations - as it is constantly altering.

Until it is possible to run 4wd cars without risk of transmission damage Weltmeister will continue to remove the vehicle's propshaft. I would rather add 30 minutes to the time taken for a power test - than destroy a customer's drivetrain on the dyno. If you search hard enough you will find stories of Porsche gearboxes (tiptronic) and front diferentials (996 Turbo) being blown up on 4wd chassis dynos. This is not an issue to be taken lightly.

I hope this makes things a little clearer.

Chris Davies
Technical Director
Weltmeister Ltd.

oceancarrera · Dec 10, 2005

ORIGINAL: MoC2S
Hi Adrien

I let this ride at the time, but have since tried to find evidence to back this up without success ... have you any documentation for this statement ?

TTBoMK the only sensors involved here are the ABS sensors detecting wheel movement. This inputs to the ABD system, but again this system is for low speed only.

To my mind the root of the problem will be the viscous coupling driving to the front. It is designed to drive only a small percentage to the front under normal condition using the rolling diameter ratios F <=> R. Presumably on the 4WD setup of the dyno each axle is left to its own devices (roller). So as we accelerate on the dyno to get the runs, the viscous coupling will transfer maximum torque to the front, because the front wheels are not being accelerated by the road as they would in normal driving. If the front drive ever catches up to speed the viscous coupling will pulse the drive on and off, hence difficulty getting a stable reading.

The obvious solution (patent pending) would be to turn the dyno into a 'rolling road' by linking front <=> rear rollers ... then the viscous will not boil itself, just run at its low % torque transfer.

So fiendish electronics are not part of this equation .... IMHO at least.

As always, happy to be proved wrong ... as long as the evidence is there !

cheers, Maurice

Hi Maurice,
I don't have any documented knowledge on this I'm afraid! Just quoting what an experienced Porsche technician recently told me.
Cheers -
Adrien

Shane Hodges · Dec 12, 2005

Thanks Jeff. This looks as though it could be my answer. I'm still not too keen on my pride and joy being taken apart simply to get a BHP figure, so if JM 79 went on in one piece and nothing burnt out, AMD appear to be the solution as I guess the front and rear rollers are connected as Maurice discusses above.

.... unless Maurice you still percieve there to be a risk of destroying diffs, even if the front rollers are connected to the rear?

By the way, it looks as though there were some serious straps holding your car in place ... at the front!

oceancarrera · Dec 12, 2005

ORIGINAL: Shane Hodges

Thanks Jeff. This looks as though it could be my answer. I'm still not too keen on my pride and joy being taken apart simply to get a BHP figure, so if JM 79 went on in one piece and nothing burnt out, AMD appear to be the solution as I guess the front and rear rollers are connected as Maurice discusses above.

.... unless Maurice you still percieve there to be a risk of destroying diffs, even if the front rollers are connected to the rear?

By the way, it looks as though there were some serious straps holding your car in place ... at the front!

Shane,
Just my humble £0.01 worth here... and with all due respect to Jeff and Maurice, but considering you have direct advice from the technical director of the country's most reputable dyno (and a Porsche specialist at that) advising against running with the prop connected, you may want to consider getting your propshaft disconnected for the run especially since it is a simple 30min (that's being generous) job on the 993. FWIW, Chris' advice is the same as I received from Steve at JZM. I wouldn't normally say this as your choice is your decision, but I would hate to hear of your "pride and joy" going bang when someone like Chris has taken the time to explain.
Whatever you choose, I hope to hear that your dyno goes well.
Cheers -
Adrien

Shane Hodges · Dec 17, 2005

I did not see Chris's posting before my post ... I didn't see that this had run to a second page!

Thank you all for all the information. What started out as a interest in how 10 years and 57,000 miles had affected the engine has turned into an enlightening experience.

However, to be brutally honest, despite a number of you with C4's having successfully tested your cars in different places, I think I will go on without knowing the latest BHP.

What I am reading verses what I was told in person continues to be inconsistent, and what has been posted here on other issues also continues to change (latest eg it appears the 993 C4 F<>R drive is purely mechanically but above Chris talks about electronic torque spilt that the Porsche delivers).

I think what really did it for me was when the C4 was referred to as not a 4WD in the truest sense .... I think Porsche would have had a few law suits by now if that had been the case!?!

Anyway, if the car gets towards rebuild time, I might look again.

38F3E1 · Dec 18, 2005

In the Driver's Manual you get with the car there is the following statement regarding dynos

"

Performance Testing

Performance test may only be carried out on 4-roller dynamometers with speed coupling. When testing on 2-roller dynamometers, the front to rear axle connection must be seperated at the central shaft.

"

Page 32 of the UK version.

Ian.