Bill,
This was a topic of some discussion a while ago which you can find in
the list archives (see the trailer to this message.
I have included below some correspondence from Jeff Boatright to Timken
engineers on the subject to save you some time.
Mike Myers
-----------ORIGINAL MESSAGE FOLLOWS-----------
Dear Sir or Madam,
I am in a bit of a rush for your answer on this as my only car and
daily driver is down until I fix my bearing problem. After getting a
lot of contradictory advice about bearings, I am trying to educate
myself at the source. I have read your online applications
information for my car (you have a great web site!), but I have more
questions.
I own a 1965 Austin-Healey Sprite. These cars originally used thrust
ball bearings separated by a spacer. There is a lively discussion on
several Sprite/Midget internet newsgroups about how to convert to
tapered roller bearings. Basically, there are two schools of thought.
First is to discard the spacer, torque the axle nut to about 10 ft-lb
and then back off to insert cotter pin. The other school of thought
is to use the spacer and torque to stock settings (46 ft-lb). This
may require some trimming or shimming of the spacer to achieve 46
ft-lbs torque and still allow the hub to spin freely.
The one UNIVERSALLY agreed upon point is to use Timken bearings and
accept no substitute.
I am wondering if a lot of the controversy is due to two different
bearing solutions being applied to the same problem. Basically, could
one set of Timken bearings require the spacer and the other not
require the spacer?
I recently purchased a set of Timken bearings from Mini-Mania, a
popular performance reseller for small British cars. The Timken
cone/cup numbers are: 07097/07204 (inner) and 05066/05185 (outer).
Don Racine, owner of Mini-Mania, long-time racer and fabricator, was
adamant that the spacer must be used and that no shims are needed (I
find that I cannot torque to spec 46 ft-lb without the hub locking
up, indicating the need for shims).
On the other hand, your web site (and several newsgroup members) say
to use Timken part numbers 30205 (inner) and 30303 (outer) AND to not
use the spacer. The torque setting is very low (about 10 ft-lb).
So, which is correct? As I have already bought the 07097/07204
(inner) and 05066/05185 (outer) bearing set from Mini-Mania, I am of
course very interested in getting them to work.
Thanks much for any advice that you can give. If it is amenable to
you, I will inform the spridgets email list of your comments as this
is a topic that has been discussed since 1998!
Finally, where can I buy Timken bearings locally (Atlanta) or online
for home delivery?
Sincerely,
Jeff Boatright
-----------TIMKEN MESSAGE 1 FOLLOWS-----------
Jeff
As an ex MG Midget owner I know what you are going through and
hopefully can help you.
Both of the bearing sets listed are suitable for use. The bores and
ODs of both sets are the same and are the same as the ball bearings
they are replacing. There are differences in the widths and this is
why the spacer becomes a key issue.
I am not familiar with how the OE build was done on the Spridgets but
on similar arrangements, using Timken bearings and a spacer, it is
common practice to have a range of spacers with different widths.
The actual spacer width selected would be dependant upon the actual
bearing widths and the width of the shoulder in the hub. This is
fine for an OE build, but since this requires a range of spacers and
some measuring equipment this is not suitable for the changeover to
Timken bearings you are talking about.
As you say trimming or shimming can be done on the spacer to achieve
the required setting. However the benefits you would gain would be
outweighed by the problems of achieving the correct width.
We would suggest that you rebuild the hub with the Timken bearings
you have and not use the spacer. The bearing inner races should be
greased before they are fitted on the shaft. The hub should be
rotated as the axle nut is tightened. This is to ensure that the
rollers are correctly seated. You should tighten the nut until you
can no longer rotate the hub by hand. The nut should then be backed
of until you can feel a slight axial movement of the hub. The cotter
pin should then be fitted. You should be looking to have an axial
endfloat in the bearings of between 0.001" and 0.003".
This is similar to the method you describe but you should rotate the
hub and use more than 10 ft-lb torque before you back off the nut.
Let me know if this helps, or if you would like further clarification.
--------MY FOLLOW-UP MESSAGE WITH TIMKEN ANSWERS EMBEDDED--------
Jeff
Sorry for the delay in responding, I took a couple of days vacation.
Questions to your latest questions are embedded in the attached notes.
-----Original Message-----
Dear Sir,
Thanks much for your rapid and thorough response; it certainly helped
and is in keeping with everything I've heard about Timken. I will
take you up on your offer of further clarification, though.
1. Given that they are of different widths, is there any strength
difference between the 07097/07204 and 05066/05185 combination versus
the 30205 and 30303 combination Timken suggests for this application
on your web site?
In terms of strength we assume you are referring to the load carrying
capability of the bearings. Yes, there is a difference, the 30205 &
30303 combination have a high load carrying capability. That said
both selections should give better life than the ball bearings.
2. If I wanted to buy the 30205 and 30303 combination, is there an
online reseller or a local (Atlanta, GA) reseller that you suggest?
There should be a local Timken Authorised Distributor local to
Atlanta. They are probably listed in the Yellow Pages.
3. Is it that the hub nut is loosened back to where there is just
slight wobble in the wheel? The problem here is that I'm not that
familiar with some of the terms. Is axial endfloat where I grab the
wheel at 12 o'clock and 6 o'clock positions and wobble it back and
forth, towards and away from me (i.e., not rotational motion, but
rather motion in the plane of the wheel)?
The key here is "Axial Endfloat", this is measured along the axis of
the stub axle. You should push the hub inboard whilst oscillating
the hub. This gives you one extreme. You then pull the hub
outboard, again whilst oscillating the hub. This gives you the other
extreme The push / pull should be along the axis of the stub axle,
not the plane of the wheel. You should be able to just feel the
0.003" endfloat as you stop pushing and start pulling.
4. One of the arguments for using the spacer has been that it somehow
adds strength to the stub axle assembly as a whole. I have never been
able to understand this theory (not saying it's wrong, just saying I
don't get it). Do you have any thoughts on that?
In theory the spacer is clamped between two bearing inner races and
effectively creates a shaft of a larger diameter. The ability for
this theory to work you need a high clamp load. In the Spridget
design the ability to have a high clamp is limited by the small
thread size. Basically the threads start to deform before you have a
high enough clamp load applied. This theory is used on some Heavy
Duty truck axles.
5. This is my last question: I've installed the roller bearings on
the driver side and they work great. The hub on that side seems to
run a little warmer than the other side, which still has the old ball
bearings. However, if I loosen the hub nut one notch back (to allow
installation of the split pin), there is way too much end float. Is
the "extra warmth" normal and acceptable?
The bearings will go through a running in period. This period is
characterised by higher running temperatures. It will typically last
for a few hundred miles.
---
Thanks again for your time and thoughts. It is really pleasant
dealing with knowledgeable and responsive people.
Jeff
________________________________________________________________
Sign Up for Juno Platinum Internet Access Today
Visit www.juno.com
/// unsubscribe/change address requests to majordomo@autox.team.net or try
/// http://www.team.net/mailman/listinfo
/// Archives at http://www.team.net/archive/spridgets
|