Bearing Trucks

[camcam]

Hi, New guy here.

Designing some bearing trucks and had a couple questions.

1 Does a bearing truck have to use 3 rollers, I would like to design one with 4 to ride a pipe? I was thinking one at say 12:00 oclock, one at 6:00 oclock, then just a little way back, one at 3:00 oclock and one at 9:00 oclock. 2 of the 4 would be adjustable. This would make it a lot easier to mount the bearing bolts as everything would be at right angles. This would be at each end so a total of 8 bearings for each slide.

2 What do you think about using gas pipe for the rail? I could slide a piece of rebar through the middle of the pipe and then pour fiberglass inside. No way in the world that would ever flex. I was thinking the pipe might be to soft on the outside?

Thanks in advance, and please go easy on me.

[andy_ck87028]

Hi Camcam

My preference is to have 3; (12, 3 and 9 o'clock). The reason for that is because the rail can then be supported underneath along its length. Rigidity of the guide rail is very important.

I used hollow stainless tube and bolted it down from underneath through holes along its length and a piece of inserted metal plate with threaded holes at the same point as the holes in the stainless. That has worked well.

Once the trucks are in place, I then spun them by 45 degrees placing the bearings at 7:30, 10:30 and 1:30 (ignore the minute, only consider the hour). That then provides a locking method for the trucks.

If you don't spin by 45 degrees, you will have to find a way to stop the trucks from lifting. I used the 45 degree spin on the Y axis only. On the X axis, I have dropped a plate below the side of the truck with a bearing that hits the underside of the rail support.

I explored having a non supported rail through fibreglass but was advised against it (no structural strength). I was told that a better method would be to insert two plates in the shape of a cross and weld them at either end.

I opted for the stainless tube method as I argued that almost rigid is not good enough for CNC. Nothing here is based on evidence only opinion. I am still a noobie.

For me, rigidity is the most important design consideration in CNC.

Have fun

Andy

[camcam]

Thanks Andy for the info!

Jerry

[jeffs555]

I could slide a piece of rebar through the middle of the pipe and then pour fiberglass inside. No way in the world that would ever flex.

Most of the strength and rigidity comes from the outside, so adding rebar and epoxy down the middle would have little if any effect on rigidity. If you need more rigidity go with larger diameter pipe, or else continuously supported pipe. Most people grossly underestimate the amount of flex in rails that are only supported at the ends.

[camcam]

Oh OK, Thanks Jeffs555. I was going to use two, 1/2" gas pipes for the Y (think thats the upper horizontal pipes), but now I will go 3/4" to be safe? It's my first attempt so I'm sure I will be learning a lot.

Jerry

Why is it you always find the mispelled words after you post.

[jeffs555]

Jerry,
How long are your axes? Even 3/4" pipe (1.05" od) will flex at longer lengths. An 18" length will only flex about .001 inches for each 10 lbs of force. However, deflection increases as the cube of the length, so a 38" length would flex nearly .01 inches for each 10 lbs of load.

Jeff

[camcam]

Hey Jeff

Around 3 feet long. Should I be going to maybe CSR? All ears.

Thanks,
Jerry

[jeffs555]

Don't know what CSR is, did you mean CRS? As far as deflection is concerned, it doesn't matter much what type of steel it is, or whether it is hardened or not. Deflection depends on elastic modulus, and all steel has approximately the same elastic modulus. Not sure if the pipes you are talking about are iron or steel, but the modulus for iron is only about 20 percent less than steel, so not much difference. If the deflection is too high for you, then the answer is either larger diameter pipe or continuous support. You can download the deflection program BeamBoy and see what the deflections will be for various size pipes in your design. http://www.geocities.com/richgetze/

[camcam]

Yes sorry, I meant CRS. I will check the link, sounds like I need to do a little more research.

Thanks for the help Jeff.
Jerry

[andy_ck87028]

Try:

Scalable / Adaptable Gantry Router

It will show you how I have tackled it.

Solid as a rock and 4ft between end supports.

Andy

(NB I need to get some more up to date pictures up on that thread)