DIY Bottom Bracket Spindle

AIR CONDITIONERS !

https://www.google.com/search?q=CART...w=1067&bih=472

On Wed, 05 Feb 2014 13:28:06 GMT, Ralph Barone
wrote:

wrote:
? use a cartridge bearing ! or oak and tallow with leather seals.

Now THAT would be an interesting design exercise.

In the 1960's I made a lignum Vitae bearing for a 10 inch propeller
shaft. It does work :-)
--
Cheers,

John B.

John B. wrote:
On Wed, 05 Feb 2014 13:28:06 GMT, Ralph Barone
wrote:

wrote:
? use a cartridge bearing ! or oak and tallow with leather seals.

Now THAT would be an interesting design exercise.

In the 1960's I made a lignum Vitae bearing for a 10 inch propeller
shaft. It does work :-)

I was imagining oak half shafts in a bamboo axle tube and trying to imagine
the bearing design for that which would withstand radial and axial loads.

James wrote:
On 06/02/14 00:45, Ralph Barone wrote:
Frank Krygowski wrote:
On Tuesday, February 4, 2014 11:31:45 PM UTC-5, Ralph Barone wrote:

The big problem I have here is estimating the loads. Assuming 150 lbs on
the back end, static radial load is approx 75 lbs each on the outer
bearings and half of max chain tension on each of the inners. Assuming 0.5
g max cornering and popping up on one wheel, the max axial load is 75 lbs
(or less if you're an optimist and assume that both bearings on an axle
share the axial load. However, I don't have a handle on dynamic loading.
What's the multiplier for going off a curb (either up or down)? Hitting a
big rock in mid-turn?

I think the only type of person who could reliably give you that
multiplier would be a person who had successful experience designing
similar machines.

Maybe your best bet would be to examine prior art. Find a trike that's
similar to what you have in mind, and find out what they're using for
bearings. They may be plain old cup & cone bearings, but if you get
information on the number and size of balls, plus the diameter of the
races, you can probably find an angular contact bearing with similar
features.

It's also worth remembering that if you under-size a ball bearing,
it rarely causes a catastrophe. Usually, it just gives you a lower
than expected life. For most unusual bikes, I don't think that
would be a big deal, unless you're way, way off!

- Frank Krygowski

Well, if this paper is to be believed, 40 g's is a reasonable number to
use. It seems high, though. I suspect when converting that acceleration to
a force, I would have to be somewhat selective in what mass I choose to use
(ie: the bike by itself vs. bike plus rider).

http://ir.library.oregonstate.edu/xmlui/bitstream/handle/1957/34273/OrendurffMichael1997.pdf?sequence=1


Why not use a bearing that's a bit larger than regular bike wheel bearings?

My wheel hubs have a 6001 bearings...

http://www.vxb.com/page/bearings/PRO...FYnjpAodz0AAxA

So go for something a bit bigger than that perhaps?

It's looking like a type 1623 bearing might be as good as I can get. 5/8"
ID, 1 3/8" OD, about 700 lb static radial load rating (so maybe 175 to 350
lb axial load rating). Interestingly enough, I can't seem to find a 3/4"
ID, 1 3/8 OD bearing. I guess there's not enough room left to stuff the
balls in.

On Thu, 06 Feb 2014 02:33:52 GMT, Ralph Barone
wrote:

John B. wrote:
On Wed, 05 Feb 2014 13:28:06 GMT, Ralph Barone
wrote:

wrote:
? use a cartridge bearing ! or oak and tallow with leather seals.

Now THAT would be an interesting design exercise.

In the 1960's I made a lignum Vitae bearing for a 10 inch propeller
shaft. It does work :-)

I was imagining oak half shafts in a bamboo axle tube and trying to imagine
the bearing design for that which would withstand radial and axial loads.

My God! Be modern! By the 1800's most wagons had shifted to steel
axles and wooden bearings :-) No reason to stick with the wooden
axles.
--
Cheers,

John B.

bamboo isnot wood. bamboo fibers are bundled without much cellulose binding. more weedy without the water content.

A large diameter oak axle for a single wheel under light load would last 2-300 miles if the grease was clean.

How far could you go?

Ralph Barone wrote:
James wrote:
On 06/02/14 00:45, Ralph Barone wrote:
Frank Krygowski wrote:
On Tuesday, February 4, 2014 11:31:45 PM UTC-5, Ralph Barone wrote:

The big problem I have here is estimating the loads. Assuming 150 lbs on
the back end, static radial load is approx 75 lbs each on the outer
bearings and half of max chain tension on each of the inners. Assuming 0.5
g max cornering and popping up on one wheel, the max axial load is 75 lbs
(or less if you're an optimist and assume that both bearings on an axle
share the axial load. However, I don't have a handle on dynamic loading.
What's the multiplier for going off a curb (either up or down)? Hitting a
big rock in mid-turn?

I think the only type of person who could reliably give you that
multiplier would be a person who had successful experience designing
similar machines.

Maybe your best bet would be to examine prior art. Find a trike that's
similar to what you have in mind, and find out what they're using for
bearings. They may be plain old cup & cone bearings, but if you get
information on the number and size of balls, plus the diameter of the
races, you can probably find an angular contact bearing with similar
features.

It's also worth remembering that if you under-size a ball bearing,
it rarely causes a catastrophe. Usually, it just gives you a lower
than expected life. For most unusual bikes, I don't think that
would be a big deal, unless you're way, way off!

- Frank Krygowski

Well, if this paper is to be believed, 40 g's is a reasonable number to
use. It seems high, though. I suspect when converting that acceleration to
a force, I would have to be somewhat selective in what mass I choose to use
(ie: the bike by itself vs. bike plus rider).

http://ir.library.oregonstate.edu/xmlui/bitstream/handle/1957/34273/OrendurffMichael1997.pdf?sequence=1


Why not use a bearing that's a bit larger than regular bike wheel bearings?

My wheel hubs have a 6001 bearings...

http://www.vxb.com/page/bearings/PRO...FYnjpAodz0AAxA

So go for something a bit bigger than that perhaps?

It's looking like a type 1623 bearing might be as good as I can get. 5/8"
ID, 1 3/8" OD, about 700 lb static radial load rating (so maybe 175 to 350
lb axial load rating). Interestingly enough, I can't seem to find a 3/4"
ID, 1 3/8 OD bearing. I guess there's not enough room left to stuff the
balls in.

Well, it seems that all that remains is to reconcile the fact that you
can't cut a 1.375x24 tpi thread in a tube with a 1.375" ID and I might have
a design. Thank you everyone.

On 2/6/2014 7:27 AM, Ralph Barone wrote:
Ralph Barone wrote:
James wrote:
On 06/02/14 00:45, Ralph Barone wrote:
Frank Krygowski wrote:
On Tuesday, February 4, 2014 11:31:45 PM UTC-5, Ralph Barone wrote:

The big problem I have here is estimating the loads. Assuming 150 lbs on
the back end, static radial load is approx 75 lbs each on the outer
bearings and half of max chain tension on each of the inners. Assuming 0.5
g max cornering and popping up on one wheel, the max axial load is 75 lbs
(or less if you're an optimist and assume that both bearings on an axle
share the axial load. However, I don't have a handle on dynamic loading.
What's the multiplier for going off a curb (either up or down)? Hitting a
big rock in mid-turn?

I think the only type of person who could reliably give you that
multiplier would be a person who had successful experience designing
similar machines.

Maybe your best bet would be to examine prior art. Find a trike that's
similar to what you have in mind, and find out what they're using for
bearings. They may be plain old cup & cone bearings, but if you get
information on the number and size of balls, plus the diameter of the
races, you can probably find an angular contact bearing with similar
features.

It's also worth remembering that if you under-size a ball bearing,
it rarely causes a catastrophe. Usually, it just gives you a lower
than expected life. For most unusual bikes, I don't think that
would be a big deal, unless you're way, way off!

- Frank Krygowski

Well, if this paper is to be believed, 40 g's is a reasonable number to
use. It seems high, though. I suspect when converting that acceleration to
a force, I would have to be somewhat selective in what mass I choose to use
(ie: the bike by itself vs. bike plus rider).

http://ir.library.oregonstate.edu/xmlui/bitstream/handle/1957/34273/OrendurffMichael1997.pdf?sequence=1


Why not use a bearing that's a bit larger than regular bike wheel bearings?

My wheel hubs have a 6001 bearings...

http://www.vxb.com/page/bearings/PRO...FYnjpAodz0AAxA

So go for something a bit bigger than that perhaps?

It's looking like a type 1623 bearing might be as good as I can get. 5/8"
ID, 1 3/8" OD, about 700 lb static radial load rating (so maybe 175 to 350
lb axial load rating). Interestingly enough, I can't seem to find a 3/4"
ID, 1 3/8 OD bearing. I guess there's not enough room left to stuff the
balls in.

Well, it seems that all that remains is to reconcile the fact that you
can't cut a 1.375x24 tpi thread in a tube with a 1.375" ID and I might have
a design. Thank you everyone.


Right but if everything else is suitable you could turn the
thread off and just press the cup into the tube.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971

In article , AMuzi
wrote:

On 2/6/2014 7:27 AM, Ralph Barone wrote:
Ralph Barone wrote:
James wrote:
On 06/02/14 00:45, Ralph Barone wrote:
Frank Krygowski wrote:
On Tuesday, February 4, 2014 11:31:45 PM UTC-5, Ralph Barone wrote:

The big problem I have here is estimating the loads. Assuming 150 lbs
on
the back end, static radial load is approx 75 lbs each on the outer
bearings and half of max chain tension on each of the inners.
Assuming 0.5
g max cornering and popping up on one wheel, the max axial load is 75
lbs
(or less if you're an optimist and assume that both bearings on an
axle
share the axial load. However, I don't have a handle on dynamic
loading.
What's the multiplier for going off a curb (either up or down)?
Hitting a
big rock in mid-turn?

I think the only type of person who could reliably give you that
multiplier would be a person who had successful experience designing
similar machines.

Maybe your best bet would be to examine prior art. Find a trike that's
similar to what you have in mind, and find out what they're using for
bearings. They may be plain old cup & cone bearings, but if you get
information on the number and size of balls, plus the diameter of the
races, you can probably find an angular contact bearing with similar
features.

It's also worth remembering that if you under-size a ball bearing,
it rarely causes a catastrophe. Usually, it just gives you a lower
than expected life. For most unusual bikes, I don't think that
would be a big deal, unless you're way, way off!

- Frank Krygowski

Well, if this paper is to be believed, 40 g's is a reasonable number to
use. It seems high, though. I suspect when converting that acceleration
to
a force, I would have to be somewhat selective in what mass I choose to
use
(ie: the bike by itself vs. bike plus rider).

http://ir.library.oregonstate.edu/xm...957/34273/Oren
durffMichael1997.pdf?sequence=1


Why not use a bearing that's a bit larger than regular bike wheel
bearings?

My wheel hubs have a 6001 bearings...

http://www.vxb.com/page/bearings/PRO...rwCFYnjpAodz0A
AxA

So go for something a bit bigger than that perhaps?

It's looking like a type 1623 bearing might be as good as I can get. 5/8"
ID, 1 3/8" OD, about 700 lb static radial load rating (so maybe 175 to 350
lb axial load rating). Interestingly enough, I can't seem to find a 3/4"
ID, 1 3/8 OD bearing. I guess there's not enough room left to stuff the
balls in.

Well, it seems that all that remains is to reconcile the fact that you
can't cut a 1.375x24 tpi thread in a tube with a 1.375" ID and I might have
a design. Thank you everyone.


Right but if everything else is suitable you could turn the
thread off and just press the cup into the tube.

The reason I had chosen to put bottom bracket cups in the design was not
to hold the bearings radially (ie: put the bearing inside the cup), but
to give me a means to adjust the axial slop in the system and allow me
to fine tune the axle position wrt the diff. It turned out that old
style bottom bracket cups were the cheapest part that came readily to
mind to do that, so the thread is the important bit for me. I really
should try and put together an ASCII-CAD drawing of what I have/had in
mind.

In article invalid-D9B8AE.06353206022014@shawnews,
Ralph Barone wrote:

In article , AMuzi
wrote:

On 2/6/2014 7:27 AM, Ralph Barone wrote:
Ralph Barone wrote:
James wrote:
On 06/02/14 00:45, Ralph Barone wrote:
Frank Krygowski wrote:
On Tuesday, February 4, 2014 11:31:45 PM UTC-5, Ralph Barone wrote:

The big problem I have here is estimating the loads. Assuming 150
lbs
on
the back end, static radial load is approx 75 lbs each on the outer
bearings and half of max chain tension on each of the inners.
Assuming 0.5
g max cornering and popping up on one wheel, the max axial load is
75
lbs
(or less if you're an optimist and assume that both bearings on an
axle
share the axial load. However, I don't have a handle on dynamic
loading.
What's the multiplier for going off a curb (either up or down)?
Hitting a
big rock in mid-turn?

I think the only type of person who could reliably give you that
multiplier would be a person who had successful experience designing
similar machines.

Maybe your best bet would be to examine prior art. Find a trike
that's
similar to what you have in mind, and find out what they're using for
bearings. They may be plain old cup & cone bearings, but if you get
information on the number and size of balls, plus the diameter of the
races, you can probably find an angular contact bearing with similar
features.

It's also worth remembering that if you under-size a ball bearing,
it rarely causes a catastrophe. Usually, it just gives you a lower
than expected life. For most unusual bikes, I don't think that
would be a big deal, unless you're way, way off!

- Frank Krygowski

Well, if this paper is to be believed, 40 g's is a reasonable number
to
use. It seems high, though. I suspect when converting that
acceleration
to
a force, I would have to be somewhat selective in what mass I choose
to
use
(ie: the bike by itself vs. bike plus rider).

http://ir.library.oregonstate.edu/xm.../1957/34273/Or
en
durffMichael1997.pdf?sequence=1


Why not use a bearing that's a bit larger than regular bike wheel
bearings?

My wheel hubs have a 6001 bearings...

http://www.vxb.com/page/bearings/PRO...HtrwCFYnjpAodz
0A
AxA

So go for something a bit bigger than that perhaps?

It's looking like a type 1623 bearing might be as good as I can get.
5/8"
ID, 1 3/8" OD, about 700 lb static radial load rating (so maybe 175 to
350
lb axial load rating). Interestingly enough, I can't seem to find a 3/4"
ID, 1 3/8 OD bearing. I guess there's not enough room left to stuff the
balls in.

Well, it seems that all that remains is to reconcile the fact that you
can't cut a 1.375x24 tpi thread in a tube with a 1.375" ID and I might
have
a design. Thank you everyone.


Right but if everything else is suitable you could turn the
thread off and just press the cup into the tube.

The reason I had chosen to put bottom bracket cups in the design was not
to hold the bearings radially (ie: put the bearing inside the cup), but
to give me a means to adjust the axial slop in the system and allow me
to fine tune the axle position wrt the diff. It turned out that old
style bottom bracket cups were the cheapest part that came readily to
mind to do that, so the thread is the important bit for me. I really
should try and put together an ASCII-CAD drawing of what I have/had in
mind.

There... You just can't draw stuff like this on an iPhone. Imagine the
exact same hardware on the other side of the axle tube.


----------------------------------------------+
AXLE TUBE |
--------------------------------/\/\/\/\/\/\/\|
| |/\/\/\/\/\/\/\/\/\/\
| || BB CUP |
| |+----------------+ |
+------+|BEARING | | |
|COLLET|| | +__+
| || |
--------------+------+--------------------------------------------+
|
-- TO DIFF AXLE TO HUB-- |
|
--------------+------+--------------------------------------------+
| || |
| || | +--+
+------+| | | |
| |+----------------+ |
| || |
| |\/\/\/\/\/\/\/\/\/\/
--------------------------------\/\/\/\/\/\/\/|
|
----------------------------------------------+

The collets lock to the axle shaft. I figure this is a minimal parts
design that allows fine adjustment of axle position as well as bearing
preload but doesn't use any weird parts or require any custom machining.
Any suggestions on improvements or simplifications are certainly welcome.