Roller Mill
Roller Mill
(OP)
I am designing a two-roller mill press (imagine an old washing machine that wrings out clothes), and I need to make the rollers out of tube rather than keep them solid (for a variety of reasons). My question is: how do I calculate how thin the walls of the tube that I make the rollers out of can be so that I can make the design as robust as necessary without the possibility of collapsing the tube while minimizing material in the roller?
I have all of the necessary parameters: roller OD, material properties, anticipated forces, etc. I just need some general guidance on what calc. to use. Thanks.
I have all of the necessary parameters: roller OD, material properties, anticipated forces, etc. I just need some general guidance on what calc. to use. Thanks.





RE: Roller Mill
If you design to minimize the deflection in the tube due to the lateral forces exerted during the rolling (like a beam analysis), it should override the torsional forces, so the tube can handle it as long as the bearing/bushing attachment at the ends can handle it.
How much lateral deflection can you tolerate?
Ron
RE: Roller Mill
Dan
www.eltronresearch.com
Dan's Blog
RE: Roller Mill
Many designs use backup rollers to help the main rollers.
What are you rolling and what forces do you expect??
RE: Roller Mill
Dan
www.eltronresearch.com
Dan's Blog
RE: Roller Mill
What is the thickness of pulp (downstream) and the clearance of the rollers? (Yes, should be the same; however, the load required in the pressing will exert a lateral force...thus slight deflection).
Ron
RE: Roller Mill
Is it the roller itself, or something else?
RE: Roller Mill
Dan
www.eltronresearch.com
Dan's Blog
RE: Roller Mill
Are you squeezing the water out of a extruded shape of pulp to end up with a fairly dry, and firm 14" wide strip of paper that's .03 - .06" thick. How do you form the edges of this strip or prevent the extruded shape from expanding laterally when going through the rollers? What is the compressive strength of the .06" paper strip material, right out of the rollers? Where did you get or how did you arrive at ~200psi?
The rollers are both driven, and at the same surface speed, right? So you don't introduce a shearing force into the paper. You say 16" long and 8" O.D. with thinnest wall possible. How do you know the roller material before you do the design? Why 8" dia. and as light as possible? Why not 6" dia. and much thicker wall? You get a much better roller surface and a stiffer roller. Then how do you put ends on this mechanical tubing, which are fatigue resistant, and provide for end bearings and a drive system, and won't defect to much? I venture to guess that you will end up better off buying two pieces of a high quality 7" round stock, and machining your rollers out of them. Or, go for thinnest tube wall possible and start a fatigue testing program, and get paper which is .06" at the center and .04 at each edge. What's the reason for the thinnest possible pipe walls, because this fatigue testing program could get pricy.
RE: Roller Mill
Why not give us the roller OD, material properties, anticipated forces, etc. And, while you are at it, provide a sketch of the rollers and forces. The calculations may be relatively straightforward, but we need a little more information. Thanks.
BA
RE: Roller Mill
ex-corus (semi-detached)
RE: Roller Mill
Anyway, if anyone is curious this set of rollers is part of a retro-fit of a test rig in a partner's facility. The rollers aren't driven, and the force of 200lb. is their industry standard. I'm making the roller out of tubing for several reasons which I don't really want to get into, plus the current rollers are made similarly. There's nothing too top secret or too bizarre about the materials, but I didn't add the whole of my proposed design because I don't want to turn a quick question into a Phase I project. I did some FEA on my models, and it seems to match what experience tells me should work. However, I just wanted to see for my own benefit what calcs. folks would use.
Thanks.
Dan
www.eltronresearch.com
Dan's Blog
RE: Roller Mill
I think any hand calculation you do though will just give you a 'feel' on the true solution, which is more accurately obtained with finite elements (if used properly).
ex-corus (semi-detached)