20 replies to this topic

[Wishbone]

There have been recent reports of Ferrari writing to the FIA to confirm their interpretation of the rules regarding differentials. This is apparently related to Ferrari having suspicions that the macs are using some kind of torque steer differential.

My question is:-

What does a torque steer differential do?

[moog101]

Torque steer is the unpleasant sensation (often felt in front wheel drive cars) that makes the twisting of the cars wheel (it's torque, generated by the engine) want to pull the car's steering away from where the driver has directed it.

This would be MUCH less apparent (maybe not at all)in a RWD car - unless the Mac had a system that varied drive to the wheels in a more sophistcated fashion than a conventional differential (which allows outside wheels to rotate at a higher speed than inside)

Anyone with the necessary qualifications to comment??

BTW, I take it no-one is interested in helmet aero, as the thread posted earlier this week. Sad lonely old Moog

[desmo]

While electronically controlled differentials are allowed and used in F1, they must ape the behavior of what are deemed to be the parameters of a mechanical analog. The teams appear to use a Salisbury-type diff which hydraulically loads the clutch pack to alter the internal friction between the two outlet shafts. This is accomplished by means of a Moog valve which is controlled by the chassis' ECU.

The specifics of what defines the parameters that are presumed to be mechanically possible are somewhat vague, and are, apparently, decided by the scrutineers on a case-by-case basis. Basically, to mimic a mechanical diff two types of algorithms are allowed for torque splitting: schemes that apportion torque as a function of total torque, or those that use the difference between the wheel's rotational speeds. As well, the diff may not allow a greater ratio of torque distribution than the ratio of grip between the driven wheels. There appear to be a lot of potential grey areas to be mined here.

Peter Wright, in the May 2000 issue of Racecar Engineering, described a "tank steer" twin-clutch diff design that employes a pure mechanical means to allow the diff to react to varying lateral and longitudinal accelerations. This could allow the diff to impart a turning moment similar to Mac's outlawed "fiddle brake" system while, arguably, staying within the technical regs.

[PDA]

Wright's article also stated that both Ferrari and Mac might well be developing something very similar to the twin clutch system, and even suggested that Mac may already have installed such a system. I presume that the "request for clarification" relates to: a) amount by which torque may be varied between inside and outside wheels; or b) the mechanical means by which such control is achieved.

Might we see Minardi or Sauber protesting the Mac during first practice next week?

[desmo]

And isn't it ironic that the inputs to allow the diff to adapt to varying lateral and longitudinal accelerations in this system would have to be implemented by a ridiculous system of Rube Goldberg antedeluvian-tech pendulum-type accelerometers acting upon a hydromechanical system that would be far more expensive, complicated and fragile than simply allowing a more purely electronic control? This is the sort of idiocy that results from trying to "uninvent" technologies.

[moog101]

Moog valve huh? Now that is cool. Looks like dad did more than just make Amps

[desmo]

It's a very clever hydraulic servo that is a development of designs used in aircraft controls for decades.

[MrAerodynamicist]

Originally posted by desmo
This could allow the diff to impart a turning moment similar to Mac's outlawed "fiddle brake" system while, arguably, staying within the technical regs.

However, following the previous "clarification" is should therefore be automatically disallowed. The brake system was banned under the rule that anything that produces a turning moment other than thought steering the front wheels is disallowed. Not that I think it should be DSQ on technical merit.

[Darren]

Originally posted by moog101
Moog valve huh? Now that is cool. Looks like dad did more than just make Amps

And synths, voltage-controlled oscillators, envelope filters and effects modules. He's done some fine work with Theremins and guitar effects pedals as well, has yer dad. Perhaps you could get him to build a modular based on a Supertec v10.

[moog101]

I dunno - I think a Cosworth DFV would be more his era

[tak]

The Gleason 'Torsen' differential is a very clever contraption that uses worm gears instead of the typical bevel gears in differentials. The advantage is that a worm gear cannot be driven by it's mating gear, so a torsen diff will never light up the inside tire. According to Gleason, the Torsen is capaple of up to a 90/10 torque split, biasing up to 90% of the torque to the tire with the most traction. Sounds great.

The disadvantage, is that it is very heavy, and rather inefficient (because worm gears have sliding contact opposed to the rolling contact of an involute bevel gear). The higher the bias ratio, the less efficient it becomes.

Now then, if F1 teams are allowed to emulate a mechanical device with electronics and hydraulics, one (namely Maclaren) could emulate the action of a torsion without paying the weight or efficiency penalty. And from what you've all posted about the regs, it sounds legal.

Comments?

[desmo]

Yes, if that is approved as an analog for a torque-splitting algorithm, that might expand the allowable torque split. Might look like TC out of low-speed corners to someone using a less versatile algorithm.

[Ray Bell]

If a worm gear cannot be driven by its opposing gear, how come people can drive Peugeot 404s without locking the wheels when they back off?
Ratios vary from 4.2 to 6.75:1 in my experience, and this has never happened to me. Even though an engineer friend of mine reckons it should...

[Damop]

In the other thread I posted the following:

IndyIan - the concept would be very similar in execution to Honda's Active Torque Transfer System (ATTS). ATTS is quite the system - it can feel somewhat unnatural and get fooled when hammering the throttle during lane changes, but I have never driven a front-drive car that can power out of corners with minimal understeer like the Prelude SH. The SH has sensors that measure yaw, steering inputs, and lateral and longitudinal acceleration. It will feed up to 80 percent of the power to the outside front wheel, literally pulling the car around the corner. It is very effective, and it requires far less steering inputs, especially for a front-drive car - I was amazed how much you can change the line around corners just with the throttle, much more so than normal. Based on tests between ATTS and non-ATTS Preludes (which are identical except for the ATTS), the system is worth about 1 second a lap on a 1-minute lap. That is significant.

Extending this to the McLarens, if they have such a system it would be a tremendous benefit. Not only would it allow more power to be applied earlier in the corner (since the outside rear wheel has more grip), but it would "artificially" lower the car's polar moment, and it would turn much better since steering inputs would be applied at both ends almost simultaneously, rather than the rear always being a half-step behind the front end's moves. This system would most certainly be illegal as it would involve using electronics to transfer power to the wheel with more grip (outside wheel) in an optimal manner for each corner compared to a standard mechanical LSD - it would be damn close to TC.
__________________
Ah, alcohol! The cause of, and solution to, all of life's problems.

[desmo]

You can say that again Homer.

[tak]

Ray-
Leave it to the French to get a worm gear to work backwards.

Actually, it is not unreasonable--as the ratio of the worm gear gets lower and lower (closer to 1:1), the gear starts looking more and more like a bevel gear.

Most worm gears have ratios on the order of 20:1 and up, so the units you describe have very low ratios indeed...

[Ray Bell]

I've seen block and tackles with numbers around the same as the lower Peugeot diffs... They don't slip.
And earlier model cars and trucks saw a proliferation of the worms... Rover in the early part of the century, lots of pommie trucks, Fodens and so on, I think there may even still be some, they are a bit slack in keeping up with the technology. They must be lower ratios, too.

[swoopp]

However, following the previous "clarification" is should therefore be automatically disallowed. The brake system was banned under the rule that anything that produces a turning moment other than thought steering the front wheels is disallowed. Not that I think it should be DSQ on technical merit. [/B][/QUOTE]


Mr.A, the brake system was disallowed because the rules said that the brake circuit must operate all the brakes on the car. In case one or more of the brakes failed.

That was the point Ferrari argued...

[BT52]

[QUOTE]Originally posted by swoopp
However, following the previous "clarification" is should therefore be automatically disallowed. The brake system was banned under the rule that anything that produces a turning moment other than thought steering the front wheels is disallowed. Not that I think it should be DSQ on technical merit. [/QUOTE]


Mr.A, the brake system was disallowed because the rules said that the brake circuit must operate all the brakes on the car. In case one or more of the brakes failed.

That was the point Ferrari argued... [/B][/QUOTE]

If only the steering wheel is allowed to cause a turning moment then how does that deal with power oversteer...?

And why should the additional brake be classified as "THE" brake circuit if it is in addition to THE brake circuit.

I'm not comfortable with a system which is entirely down to the driver's control being banned.
Seperate braking controls could have added a whole new dimension to a driver's art.

[moog101]

I agree with that point. It is hardly as though it would be an easy system to use, and it was completely uner driver control. The diff system is clever, but I think if it is allowed then the second brake pedal should be too.

[PDA]

After the Brazil fiasco, the F1 tech rules were changed, and now, brake pressure must be equal form side to side, so diff. braking is just plain illegal, no argument about turning moments and being under drivers control. That is what makes twin clutch "tamk steer" type diffs so important (at least, potentially).