From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.misc,rec.bicycles.tech
Subject: Re: Avocet tire wear question
Date: 3 May 2001 15:48:28 GMT

Doug Milliken writes:

>> I don't know what you mean by one wheel per axle driving.
>> Differentials are employed for exactly that reason and divide drive
>> to left and right equally unless there is a large rotational
>> disparity and a limited slip feature is included.

> An aside on automotive diffs:
> Many automotive differentials that we have seen data on (and in some
> cases measured) have enough internal friction that they act like
> limited-slip units (to some extent)...without any special limited
> slip clutches or other obvious mechanism.

That became apparent when NSU and some other small cars used a spur
gear differential, one that has no differential friction drag.  Bevel
gears of common differentials have a plain bearing thrust face with
substantial drag, besides which they force the bevel gears of the
output shafts apart by their large pressure angle.  By placing stator
and rotor friction plates on these "axle" shafts is how limited slip
is introduced to increase a greater cross-locking effect.  The higher
the torque, the more solidly the two axles are locked together.

Jobst Brandt      <jbrandt@hpl.hp.com>

Newsgroups: rec.bicycles.misc,rec.bicycles.tech
From: Doug Milliken <bd427@freenet.buffalo.edu>
Subject: Re: Avocet tire wear question
Date: Thu, 3 May 2001 18:19:00 GMT

On 3 May 2001, Jobst Brandt wrote:

> Doug Milliken writes:
>
> >> I don't know what you mean by one wheel per axle driving.
> >> Differentials are employed for exactly that reason and divide drive
> >> to left and right equally unless there is a large rotational
> >> disparity and a limited slip feature is included.
>
> > An aside on automotive diffs:
> > Many automotive differentials that we have seen data on (and in some
> > cases measured) have enough internal friction that they act like
> > limited-slip units (to some extent)...without any special limited
> > slip clutches or other obvious mechanism.
>
> That became apparent when NSU and some other small cars used a spur
> gear differential, one that has no differential friction drag.  Bevel
> gears of common differentials have a plain bearing thrust face with
> substantial drag, besides which they force the bevel gears of the
> output shafts apart by their large pressure angle.  By placing stator
> and rotor friction plates on these "axle" shafts is how limited slip
> is introduced to increase a greater cross-locking effect.  The higher
> the torque, the more solidly the two axles are locked together.

Exactly.  Some designs like the Dana Trac-Loc (iirc) use bevel-gear tooth
forms that give the desired pressure on the clutch packs (in conjunction
with a Bellville spring that supplies the initial clutch preload).  Other
designs squeeze the clutch plates with other mechanisms (i.e., pin-on-ramp
in the Salisbury).

It's possible to make a bevel-gear diff with rolling bearings but I don't
know if anyone does these days.  I've seen prototypes but I don't think
they went to production (too expensive).  This would probably work like the
spur-gear diff that you mentioned above - low internal friction and very
nearly equal torque out each axle, independent of input torque.

-- Doug
                Doug Milliken  <bd427@bfn.org>