14 replies to this topic

[seryt]

This may be a bit off topic for modern F1, but I'd like to try understand how some of the early limited slip diffs worked, including the ZF used by Merc on their 30's and 50's F1 cars. I've haven't been able to find any more useful technical info than is what in Pomeroy's "The Grand Prix Car" - that seems to indicate a friction type arrangement. What's not immediately clear however, is whether the difference in torque at which some relative motion could occur was fixed for a given configuration (although presumably this could be adjusted in some way) or whether the difference in torque for relative motion to occur varied according to the total amount of torque being transferred. Can anyone provide some further info on this? Does anyone know of a source of drawings of the diff arrangement? Also, was the diff such that it was effectively open under power off conditions?

[McGuire]

The original ZF limited-slip was a cam-and-pawl type, configurable but not really adjustable. Yes, it is effectively open in coasting (or as it rapidly wears out har). You can find explanations and diagrams in Mercedes, Porsche etc service manuals and literature. Some Hewland boxes used the ZF... a variation of the design was also employed in some Dana Spicer truck axles.

[Ray Bell]

Hewland also made the ZF-style units for BMC A-series differentials... or I believe that Hewland did... there were plenty of these around for people racing Clubmans and Spridgets.

[phantom II]

A friend of mine restores Ferraris. He had a brand new replacement differential for a 250 Lusso for a SWB that he was restoring. (Ferrari will make any part for any Ferrari. Not cheap.) It is a magnificent piece of machinery. It can be dismantled or assembled completely by hand. Ferraris very rarely require special tools, an Italian engineering trait. It has 5 tapering slots on each side of the casing with corresponding slots in the core which is splined to the axle. In each slot is a 3/4" ball bearing. It seems that if torque is lost on the shaft the ball is jammed into this slot and the axle locks on this side. The rest of the time, the diff is open.

Originally posted by McGuire
The original ZF limited-slip was a cam-and-pawl type, configurable but not really adjustable. Yes, it is effectively open in coasting (or as it rapidly wears out har). You can find explanations and diagrams in Mercedes, Porsche etc service manuals and literature. Some Hewland boxes used the ZF... a variation of the design was also employed in some Dana Spicer truck axles.

[McGuire]

Originally posted by phantom II
Ferraris very rarely require special tools, an Italian engineering trait.

Right, no special tools were employed in their manufacture so there's no good reason they would be required thereafter. A lead hammer and a crescent wrench and you are good to go.

[seryt]

Originally posted by McGuire
The original ZF limited-slip was a cam-and-pawl type, configurable but not really adjustable. Yes, it is effectively open in coasting (or as it rapidly wears out har). You can find explanations and diagrams in Mercedes, Porsche etc service manuals and literature. Some Hewland boxes used the ZF... a variation of the design was also employed in some Dana Spicer truck axles.

Thanks for the info McGuire. Would the rate of wear have been so high that in a normal GP length race there would be a significant change in behaviour?

[McGuire]

Originally posted by seryt


Thanks for the info McGuire. Would the rate of wear have been so high that in a normal GP length race there would be a significant change in behaviour?

In the era of skinny tires I would tend to doubt it.

Of the various limited-slips the ZF cam-and-pawl is the smoothest and easiest to drive...it never totally locks up.

[Ray Bell]

Originally posted by seryt
Thanks for the info McGuire. Would the rate of wear have been so high that in a normal GP length race there would be a significant change in behaviour?

I can give a rather different answer to that...

Originally posted by McGuire
In the era of skinny tires I would tend to doubt it.....

Winding the clock back to 1961 (still the skinny tyre era, but getting into the Dunlop soft-tread research era), Bib Stillwell arrived at the Australian Grand Prix in October.

From 'The Official 50-race History of the Australian Grand Prix' - Graham Howard
Second to get the flag was Stillwell, with the bafflingly slow lowline (Cooper). The team had tried to do everything right, and, apart from having the best car, had also imported the best rubber, the first D-12 Dunlop R5s to be used in Australia. Somehow the Cooper had run not only the Monday-morning practice, but the race itself, with one D-12 and one D-9 on the rear (possibly because the team had carefully buffed off all identification marking to preserve the secret of their intended advantage). The result was very difficult handling, testified to by the fact that the Cooper's ZF differential, with a normal life of at least 1000 miles between overhauls, was completely worn out at the end of the 105-mile race. And with the fading of the ZF went Stillwell's chance to win the race.

That might give some insight into things, I think. I will add, however, that the circuit (Mallala) was very tight and twisty.

[Greg Locock]

Interesting - i wonder if the two tyres had different rolling radii?

[Ray Bell]

Doubtful... I can post photos so you can measure if you like?

One reason he got away with buffing off the numbers was because they were the same size, that would be logical, wouldn't it?

Grip was in the process of starting to improve dramatically in those times, and if he had more grip on the left than the right, he'd have been in trouble at Mallala.

[seryt]

Thanks for sharing that info Mr Bell. I guess if the two tyres had different longitudinal stiffness then to run at a common slip ratio the diff would be loaded even when accelerating in a straight line. The differences must have been pretty substantial to wear it out in such a short time compared to normal!

[Ray Bell]

Stillwell had been in some pretty good cars for about four or five years at this time, and raced fairly regularly. He had formed some kind of close alliance with Brabham too, I think, so he would have become capable of being reasonably analytical...

But like Moss with the Colotti boxes in the Coopers... he may have come up with the wrong solution to solve the problem.

You might recall that Moss, faced with a gearbox that frequently failed, short changed a lot to try and save the box. But, in fact, doing that exposed the improperly cut gears to more torque and worsened the problem.

Stillwell, not aware of the true nature of the problem, might have been trying things that made it harder on the diff.

Of course, we'll never be able to ask him now...

[McGuire]

If the reporter Mr. Graham Howard has it right, running two different tire compounds on LR and RR caused the limited slip to wear out in less than 100 miles. I am very skeptical about that but I cannot say it's impossible. I am skeptical beause the ZF has considerable slip to begin with.

[Ray Bell]

There is always the possibility that this was a conclusion they drew erroniously...

But the circuit is very tight. Tightest in the country. That puts the greatest strain on the differential action and calls for a lot of high-torque loadings as they accelerate out of those tight corners.

[David Beard]

Have a look at this thread on TNF from about post 90
http://forums.atlasf...y=&pagenumber=1