Finagling the Gantry

[samualt]

I have a router (not built yet) that uses a straight beam that slides along the two top-rails (X-Axis), as opposed to the donut shape of many gantries. I wish to mount the router in the most economical way. I was wondering about the pros and cons of each way. What I am worried about is the Y-Axis, or side-to-side movement.

1. In the first picture below you see the router mounted so that it uses two rails to slide from side-to-side (Y-Axis). However, the rails are one behind the other on different beams.
2. In the second picture it uses two rails but only one beam to slide from side-to-side.
3. The third picture shows the most common format. One rail on top of the other.


Blue= THK Rails.
Green = Bearing Blocks.
Purple = Router.
Grey = Metal Beams.


1. The first picture looks ok, but it seems to me most of the weight would be on the bearings from the first beam, with the rear bearings just keeping it from pivoting.
2. The second picture I won't use because all the weight is on one metal beam. I want the gantry weight distributed over two beams.
3. The third picture looks best but it would mean adding a third beam (for the top most Y-Axis bearings). Thats allot of added weight. I'm trying to purposely make this router light. It's going to be portable.

Is there some reason I should only use the third method instead of the first? Would the first method be stable at all? Any comments? Are there other methods I'm missing?

Similar Threads:

• Fixed gantry vs moving gantry
• Gantry
• New XYZ Gantry Kit
• Gantry: planning to design moving open gantry
• Diffrences between bulltear cnc gantry kits and torchmate2 gantry kits

[azsigns]

You're really making me think. Now I can't go to sleep until I know myself.

Gotta dust off the old static mechanics text book!

It does not really make much difference. The most significant thing is the torque on the bearings. The torque or the moment are the forces on the tip of the cutter times the distance to a bearing.

To make things simple say you have an even 10 lb force on the bit and it's 10 inches from the bearing.
The moment is 10lbs * 10" = 100 lb-in

If you reduce that distance to 5" you get only (10 lbs * 5" = 50lb-in) half the moment on the bearing for the same 10 lb force on the bit.

Another way to look at it is like when you put a pipe on the end of a wrench to take out a stuck bolt. You get a whole lot more torque on a long wrench.

This was just a simple illustration. In real life things are a little more complicated with forces & moments on multiple planes & vectors not to mention the dynamics when its all in motion.

On multiple bearings the moment on the bit exerts forces on the two rails in opposite direction of each other.

Here, I'll just draw a diagram I'm not good with words.
A whole essay can be written about this alone.

Note that the diagrams to the right are the equivalent of your different layouts.

Hope this helps.

Anyone, if I'm wrong let me know!
But give me a break it's 3:00am & I can't sleep.

Chris

[azsigns]

I just realized that the THK bearings may have different loading & moment capacities on different planes.

It may not be significant but you might want to also keep this in mind.

[High Seas]

Both of you guys have presented nice analysis and pics to get the brain juice flowing! Thanks for that!

I offer a method/option to add to your choices. I note you have not considered the router between the rails option. You can place the rails on top of the Gantry (Y Axis componets) that way too.

By placing the router between the Gantry, the loads are better "shared" by the two rails and bearing sets. Less cantilever between the rails and router. Some slight complication on bit changing perhaps, and you'd need to work out the load diagrams to your satisfaction.

I'll post a pic of my System2 that is set up that way later today - or tomorrow. I'm using 80/20 so configure/reconfigure was not to hard. Loss in X axis was minimal - the same nearly as you show in your diagrams.
Jim

[azsigns]

High Seas is right. The best way is to put the router between the bearings without the cantilevering.

Makes it a little more complicated but I've seen some plasma cutters set up this way.

I would just try and keep the router high and close to the Y axis plate with the bearings spread.

BTW, I just realized that my old Dahlgren Engraver is set up like your first drawing with Y bearings in the horizontal plane.

Chris

[arvidb]

Hi!

I'd say the first option is the most stable one (minimum flex). I'll get back to this soon. You might have a very difficult time building it though, with the more complicated shape of the Z "sled", and not the least, aligning the THK rails! The ones I bought (HSR25) need to be parallell within a few tens of microns!

As Chris' diagrams shows very well, #1 and #3 is about equal in load on the bearings. However, in the 3rd example, the whole moment is concentrated in the lower right corner of your gantry beam, while it is spread out over the whole gantry in #1.

But, the good part is, why not combine the pros of #1 and #3? Just go with #3, but add a supporting "lid" between the two vertical parts of the gantry beam.

Arvid

[Hobbiest]

IMO, the third choice would be the most dimensionally stable of the three. I am not an engineer, but it seems that keeping the axis of rotation close to the Z axis is essential to lightweight, strong design. If you so want to do option 2, you can always set up the rails on the opposite sides of the beam, then just switch one to be inside. As Jim stated though, the cantilevered design may pose some strength and deflection problems.

[High Seas]

samualt -
I wasn't clever enuf to get the images hung on this post - but go to the Photo Gallery (members) - High Seas and check out the System2 images.
You'l l note there is little difference in the overal width this set up as compared to the others you've been evaluating - but there is no off center loading (moment) from the router. There is a slight complication in bit changing - but I deal withit. I intentionally kept the rails "tall" so I can drive the router up above the x axis for bit changing.
System2 is all recycled bits form the System1 where I played around with the more "typical" router mounting. Felt the movement potential was to much. Plus I gained a few inches by running the cars off the end of the x axis - but not the rails,
Hope this helps stew some brain juice!
Jim

[cnczane]

Here's what I built to try to "sling" the router weight between the X-axis rails. Does it work? Heck, I don't know! But it rolls, and the lead screw pulls it with a 3/8" drill attached!

I'm still trying to get electronics/paint/stand/"it all put together".

[samualt]

Hmmm, I think I have decided to try to put the router in-between the Y rails. That way the weight is better distributed.


Blue= THK Rails.
Green = Bearing Blocks.
Purple = Router.
Grey = Metal Beams.

Somewhat like the ShopBot PRT96. Only mine will be a little more heavy duty.
Thanks for all the suggestions!