[TheForge] Bronze Bushings

Andy Vida [email protected]
Tue Apr 6 13:42:01 2004


Rick at Rafter Lazy C wrote:
> 
> OK Andy - Here we go =>
>      I am designing a Ring Roller.

	Muy cool.

>  It will be able to roll much more than the nearest competition,
> and wider as well. 

	What are the intended capacities?  How large by how large,
	according to material?  You cannot just say  1" x 6", for
	example.  You may do that in mild in the annealed state,
	but a 1/2 hard stainless steel may not only damage the roller,
	it may pose a safety hazard. Bending metal like that creates
	a spring.  The larger the spring, the greater the stored
	energy.  The greater the stored energy, the more dangerous
	a failure becomes.  So you need to know about the materials
	that you intend on designing to.  If mild is all you want to
	account for, then by all menas do so, but if this is a 
	commercial product, you'd better plaster it with stern yellow
	and red warnings all over the place that mild steel in the
	annealed state is the ONLY material that is safe to use in this
	machine.  Otherwise you need to calculate worst case for
	some high alloy steel, for example.  All you should need is the 
	worst case, as far as I can see, btu I would consult an industrial
	engineer on this because you don't want to face the liabilities
	that go with some dink putting an eye out because be tried to
	roll the wrong material and the machine disintegrated in his face.

	Mi capisce?

> I have 2 lower rolls that are
> free-wheeling, and a top roll that is attached to a keyed drive shaft.  That
> shaft will have a bronze bushing at each end and this is the one I am
> talking about.

	Why are the other shafts not bushed?  They will be under equal
	stresses, so you will probably want equal load bearing capacity
	and a good bearing makes for long machine life.  Cast iron against
	steel is, in fact, a rather good bearing material (graphite is a
	good lube) but the surface is not readily replaced.  Think of
	long term machine maintainability.  That is the consideration that
	the designers of yore made and those of today generally do not.
	I doubt there is a CNC machine out there today that will be in
	any decent operating condition in 100 years.  A will further bet
	that at that time there will still be South Bend machines and their
	ilk that will in fact stilll be up and running as the day they
	came off the assembly line.

	Anything worth doing is worth doing right.  Just my worthless
	plugged kopek's worth.

>  What I need to know is how to fit the bushing in it's
> pocket.  Does it need a press fit of .003in or a clearance of .0002in   
> I just don't know where to start looking for this, and the Machinery's
> Handbook doesn't seem to have this info. 

	Machinery's Handbook doesn't have this?  Are you certain?  That
	seems very strange to me.

	Well, given what I've seen on machinery, I would think that a
	light interference fit would be OK, but I'm not sure about this.
	You don't want the bushing slapping around in it's mounting, but
	OTOH, the bearing will be loaded in one direction at a time with
	only slight variations as the radius is progressively tightened,
	so I would imagine that a super tight fit is not necessary.  But
	you don't want to give the machine much opportunity for 
	misalignment of the shafts.  I don't know how far they can be 
	out of line before the rolls start tracking in a spiral, but 
	I'm sure it isn't very much. 0.003" is probably OK as long as 
	the shafts are all perfectly parallel.  What would concern me 
	more is the bedding in which the bearing rides.  It should be 
	a precise hole such that the bearing doesn't ride on any high 
	spot and rock, which would be disastrous for the longevity and 
	precision of the machine and may again pose a potential safety 
	hazard should it give rise to a structural failure.  The best 
	way to minimize this is to use a very beefy bushing and to 
	anchor it solidly against the shoulder with screws and a flange 
	as so in not quite so properly rendered cross section:


                 ______________                     ______________
   Flange -->___|______________|_                 _|______________|___
           |        |            |               |            |       |
           |        |            |               |            |       |
  Frame -> |        |            |               |            |       |
           |        | Bearing    |               |  Bearing   |       |
           |        |            |               |            |       |
           |        |            |               |            |       |
           |        |            |               |            |       |
           |        |____________|               |____________|       |
           |                   |                   |   ^              |
           |                   |                   |   | Shoulder     |
           |                   |                   |                  |
           |                   |                   |                  |
           |___________________|                   |__________________|


	The larger the bushing OD, within limits, the less the effect
	of slight rocking will have, but my recommendation is to have
	none.  Holding a 0.001" tolerance on the bore should be child's 
	play if you have decent machinery.  That should be plenty good 
	enough to provide the necessary stability.  What I might do is
	go to your local automotive machine shop and see if they can
	bore the holes on a Sunnen align boring machine (bores the
	main bearing seats on engine blocks.  Probably not a viable
	idea... just tossing it out to you.  But I would see that the
	the shafts were not out more than 0.001" in parallelism over
	their lengths.  It doesn't take a lot to cause spiraling, and
	therefore you'r rolls should be ground true as well.

> I don't have a Mark's, though I
> have seen the need for one for several years. 

	Get one.  I got mine for free when uncle Bengtsson died.  He
	was this big tall Swedish d00d that wore a monocle... a real
	Victorian era person.  He was so cool... and an ME to boot. :)

	I think they run about $150 new.  If you're going to use it,
	then they are worth every penny, but do a search at alibris.com
	or similar and I'm sure you can find one less costly than that.
	An older one might be your bet.  Same for Machinery's Handbook.
	Mine is from 1935 and the wealth of information there is 
	impressive.

> I already know what thickness
> I need (deduced by what is available).  ID and OD are per the catalog and
> readily available, thought I am not sure how to mount them.  Most bearing
> catalogs are full of info and engineering data concerning their bearings but
> the bronze bearing catalog I have has none of that type of info at all.
>     If you look at the tried and true methods of roller design, (look at the
> old rolling machines that are still useful, not the one that wore out after
> a measely 50 years or so) they all use bronze bushings, (or lead based
> poured bushings) rather than bearings.

	I agree.  One way is to look at a well designed and built
	machine that is close to your capacity requriements and
	double everything.  Overkill in such machinery is almost
	never a bad thing.  Will the frame be cast or fabbed?
> 
> Thanks again,
> Rick
> BTW oilite is my first choice also.

	If you're going to use very beefy bearings, you may want to
	use a regular bronze only because oilite isn't cheap and I
	am not certain what sizes it is available in.  A regular
	bronze bushing with grooves, galleries, and oil cups works
	really well and I'm so very partial to the aesthetic of
	glass oilers. :)