From: jobst.brandt@stanfordalumni.org
Subject: Re: loud freehub
Newsgroups: rec.bicycles.tech
Date: Mon, 03 Sep 2001 04:56:23 GMT

Kevin Kolse writes:

> I see on the Police bikes they have a silent freewheel.  Is this
> avaialable to the public. Any disadvantages?  What is the cost as
> opposed to one with the pawls?

I'm not sure what you are seeing but there are sprague clutches as
freewheels that have NO backlash, something I once pursued and found
impractical at the time.  When the torque required is given the
sprague manufacturers offered a diameter larger than the mid sprockets
on a freehub.  The sprague clutch is generally made of concentric
cylinders with a series of s-shaped rollers that have springs between
them like a roller bearing so that they slide in one direction but
through their gradual s-ramp in the other direction, lock up with
enormous radial force.  It is this force that makes the unit large in
diameter.  They are used as overrunning clutches in most automatic
transmissions.

I have not had one from a bicycle to disassemble but I think it would
be interesting.

Jobst Brandt    <jobst.brandt@stanfordalumni.org>

From: jobst.brandt@stanfordalumni.org
Subject: Re: American Classic hub design problem
Newsgroups: rec.bicycles.tech
Message-ID: <pkjld.5175$_3.63388@typhoon.sonic.net>
Date: Sat, 13 Nov 2004 08:11:01 GMT

David Newman writes:

>> However, various escapements have been invented over many years and
>> the silent pawl used by Regina, SunTour and others has practically
>> no drag because the large "heel" has enough viscous friction in the
>> body to not snap back at rotational speeds of interest.  That is
>> also why they cannot be heard.

> Are any currently-available hubs equipped with the silent pawl you
> describe? If so, which ones? Up to now, I've found Shimano to be
> quieter than the other hubs with which I'm familiar, but if there's
> a hub even quieter than that, I'd like to know about it. I am
> appalled by noisy hubs like Chris King, which sounds to me something
> like a swarm of angry bees.

I haven't disassembled a Shimano hub so I don't know what they use.
The problem with ratchets arose when freehubs reduced the operating
diameter of the ratchet and at the same time MTB's began using less
than 20t on chainrings with as much as a 1:2 ratio... four times any
gears that were encountered with previous freewheels.  With Pawl
failure, most designs went to what they considered failsafe, using
double or triple engagement and unusual pawls.

In freewheels with ball bearings, that invariably are not perfectly
adjusted, eccentric rotation is probable and in that case only one
pawl carries the entire load.  This is something engineers of the past
were aware of and therefopre never attempted multiple engagements.
Regina for instance had two pawls 180 degrees apart and 21 ratchet
teeth to give fine 42 engagements per rotation.

Campagnolo made an aluminum freewheel that didn't work even though
they thought they had triple engagement (of three pawls).  Since they
were singly carrying the entire load at some point, they went into
yield under high pedaling torque.  Hugi used the face spline that has
all teeth engaged at once but these also suffer from slightest
eccentricity that is inherent even elastically with the chain pull as
great as occurs at maximum torque.

As far as I could see, superficially, Shimano bit the bullet and made
their single pawls wide enough to hold the load one at a time.  I
think their large size makes them noisier than classic freewheels.
Take one apart and see if there are an even or odd number of teeth in
the ratchet and the number of pawls.

Jobst Brandt
jobst.brandt@stanfordalumni.org

From: jobst.brandt@stanfordalumni.org
Subject: Re: American Classic hub design problem
Newsgroups: rec.bicycles.tech
Message-ID: <Jmsld.5204$_3.65239@typhoon.sonic.net>
Date: Sat, 13 Nov 2004 18:27:53 GMT

Marten Gerritsen writes:

>> Take one apart and see if there are an even or odd number of teeth
>> in the ratchet and the number of pawls.

> 2 pawls @ 90 degrees, 16 teeth

Oh how crude.  16 clicks per revolution isn't a low backlash device
when compared to old Regina and Atom freewheels with 42 clicks per
revolution.  At least they gave it some thought.  At 90 degrees, both
pawls (if perfectly spaced and identical in size and shape) can be
load bearing at the same time as they crowd the rotor against the ball
bearings on the opposite side.  Still, the greatest load is held
primarily by one pawl that is at the top (where the chain pulls).  At
other positions the load is divided between them variably.

 www.m-gineering.nl

That's a nice collection of failures in the "Oops" site.  Most of
those should not have occurred.

Jobst Brandt
jobst.brandt@stanfordalumni.org

From: jobst.brandt@stanfordalumni.org
Subject: Re: American Classic hub design problem
Newsgroups: rec.bicycles.tech
Message-ID: <83tld.5220$_3.65376@typhoon.sonic.net>
Date: Sat, 13 Nov 2004 19:15:16 GMT

Carl Fogel writes:

 www.m-gineering.nl

>> That's a nice collection of failures in the "Oops" site.  Most of
>> those should not have occurred.

> Why should "most of those not have occurred"?

Because these are critical parts that usually cause a crash with
failure if the rider is moving at a reasonable speed.  Fork crowns,
for one, are parts that should not fail.  Therefore, they should have
a greater safety margin than other parts.  Handlebars and brake parts
fit that description as well.

> Do you mean that they could have somehow been prevented by better
> design?  Or by better assembly and maintenance?  Or that the
> incipient failures would not have progressed?  Or what?

I'll take "Or what."

> For those interested, here's a direct link to Marten's lurid
> gallery of failing bicycle parts:

 http://www.m-gineering.nl/oopsg.htm

Jobst Brandt
jobst.brandt@stanfordalumni.org

From: jobst.brandt@stanfordalumni.org
Subject: Re: American Classic hub design problem
Newsgroups: rec.bicycles.tech
Message-ID: <Ussld.5207$_3.65347@typhoon.sonic.net>
Date: Sat, 13 Nov 2004 18:34:28 GMT

Phil Lee writes:

>> Are any currently-available hubs equipped with the silent pawl you
>> describe? If so, which ones? Up to now, I've found Shimano to be
>> quieter than the other hubs with which I'm familiar, but if there's
>> a hub even quieter than that, I'd like to know about it. I am
>> appalled by noisy hubs like Chris King, which sounds to me
>> something like a swarm of angry bees.

These guys are all scrambling to make a ratchet that can take the load
of overweight riders on 1:2 chain ratios and within the smaller
confines of current slide-on sprocket carriers.  I'm not sure most of
them understand the problem with ratchets and which tooth is receiving
what loads.

> Old LX silent clutch hubs are still available.

I'm not thrilled with roller ramp 'clutches', commonly used in
automatic car transmissions.  They are silent and continuous but large
for their ability to transmit torque.  They can be essentially
no-backlash devices.

Jobst Brandt
jobst.brandt@stanfordalumni.org