[multivactubes]

Hello,

I intend to build a gantry mill/router using a granite surface plate as the base. To this base I will bolt two hsr25 rails 10" apart. To the carriages that ride on the rails I plan on using a 1” thick cast aluminum tooling plate for my table.

My issue is this: the difference between the thermal coefficients of aluminum and granite is about 12*10^-6 inches per inch per degree c. If there is a 10 degree c change in temperature then I am looking at about a 1.2 thou change between the width of the table and the distance between the rails. I suspect this could cause binding. Or maybe the screws that hold the carages to the table will slide the 1.2 thou needed to make up the difference?

Thoughts?

[awerby]

I don't think the screws will slide that much (they shouldn't), but there will probably be enough flex in your gantry assembly to accommodate that degree of shrinkage. I doubt that's enough to cause binding, but maybe others here will disagree.

[Eldon_Joh]

I don't think the screws will slide that much (they shouldn't), but there will probably be enough flex in your gantry assembly to accommodate that degree of shrinkage. I doubt that's enough to cause binding, but maybe others here will disagree.

I think he is talking about 2 rails and 4 blocks under a 1" thick aluminum table.

I think that the rails will slide on the granite depending on how tight the bolts are, and how stiff the rails and the bearings are. otherwise the bearing blocks are probably going to elastically take up the .0006" each.

[multivactubes]

In my design the gantry is fixed and does not move. I have attached a simple drawing to illustrate the area I am concerned about. I am worried about the difference in expansion between the table and the granite base causing the linear rails to bind or warping the table.

[he1957]

That's an awesome design for consideration. What are you planning on making with such precision?

The data sheet for the rails: https://docs-apac.rs-online.com/webd...6b80f448f1.pdf

Just a thought: If you need a certain amount of "give" and depending on the forces involved in your use perhaps nylon (or some other material) screws will give you the "flex" you need?

What about a similar granite table on the top of the bearing blocks with a tooling plate mounted on top; that would undergo the same expansion/contraction conditions as the underlying sliding surface. The Al tooling plate is attached on top and would not be involved with the mechanical movement.

[multivactubes]

Thanks for your comments. I primarily want to be able to machine plastic, wood, and aluminum and hold around a 1 thou tolerance. I think you may be on to something with the idea that there should be some provision for the dissimilar materials to slide past each other. One Idea I have is to use belleville disc springs between the cast aluminum table and the bolt heads that hold the table to the carriages on one side only. I could set the torque on the bolts to produce around 50-100 lbs of spring force per bolt. The idea would be that when the aluminum expands more than the granite the disc springs allow the aluminum plate the slip the 1.5 thou or so needed so that the rails don't bind. It would take 500 pounds of so of force to make the table slip on the carriages so the machine should still be rigid when cutting.

Another possible problem is that for the Z axis of the machine I had planned on mounting linear rails directly to another 1" thick cast aluminum plate 15 inches long. I wonder if this bimetallic strip will bend excessively with temperature changes?

[he1957]

So if 0.001 (Inch I gather) is 0.0254 mm that accuracy is claimed by many machines today using just typical materials. The material you are planning on using will suffer tolerance variations greater than the machine, if while cutting temperatures vary you'll have to consider those into the machine cycle.

You've got me curious, what "product" in those materials need that tolerance (if not for space craft)?

[he1957]

Depending on how large your parts, how about a custom cut of granite in an "L" shape from which you can use as a base for table and column vertical rise (doubt it would be inexpensive).

You could consider a cast reinforced concrete then machined approach too. Not being familiar with concrete's expansion properties mind you :-)

[extent]

I'm building a slightly smaller machine the same way ( 9.5" separation, 20mm rail, 3/4" 6061 table ) and while I haven't checked it for deflection I certainly haven't run into any binding with it yet. I've been planning on mounting a cast iron sub plate to the trucks with an aluminum fixture plate on top of that, but mainly for vibration mitigation than worrying about thermal expansion.

[RCaffin]

what "product" in those materials need that tolerance?
Ha! Why specify a product? Why not just say 'because'?

Cheers
Roger

[ninefinger]

First, 10 deg C is a large temperature change if you are trying to machine high accuracy parts. The material you are machining will expand and contract and the thermal changes you noted for your machine will also apply to the part. I work for a firm that manufactures aircraft parts and our machine shop and inspection dept have to be climate controlled to the same temperature to prevent scrap due to thermal expansion causing mismatch in the dimensions between when it was machine and when it was measured.
That said, also think about keeping the machine at the middle temperature, then you are only deviating +5 or -5 degrees. Your total change from your nominal unstressed position is only 0.0006, which would likely be taken up by the bearing blocks, slight deflection etc.

[mactec54]

Thanks for your comments. I primarily want to be able to machine plastic, wood, and aluminum and hold around a 1 thou tolerance. I think you may be on to something with the idea that there should be some provision for the dissimilar materials to slide past each other. One Idea I have is to use belleville disc springs between the cast aluminum table and the bolt heads that hold the table to the carriages on one side only. I could set the torque on the bolts to produce around 50-100 lbs of spring force per bolt. The idea would be that when the aluminum expands more than the granite the disc springs allow the aluminum plate the slip the 1.5 thou or so needed so that the rails don't bind. It would take 500 pounds of so of force to make the table slip on the carriages so the machine should still be rigid when cutting.

Another possible problem is that for the Z axis of the machine I had planned on mounting linear rails directly to another 1" thick cast aluminum plate 15 inches long. I wonder if this bimetallic strip will bend excessively with temperature changes?

Don't go there Build your machine and you won't have a thermal problem, your Linear Rails unless they are ultra precession can compensate for any mismatch in expansion between different materials like this, then you won't have any problems

Datron have been doing this for a long time take a look at there machines Granite and aluminum and everything on there machines are just fastened direct these are from $100,000.00 to $260,000.00 Plus machines

[multivactubes]

Here is an image of what I plan on building. The base (24x36x4) and gantry (18x24x3) are both granite surface plates. I will have a U shape cut out of the gantry surface plate by waterjet. Thank you all for the feedback. I think I will go ahead and use the mix of aluminum and granite in this design. I am not looking forward to drilling the 60 or so holes that are needed in the granite.

[awerby]

That certainly seems like a solid design. Maybe your waterjet place can put the holes in the granite for you?

[extent]

Even if it's just punching pilot holes that will massively, massively reduce the amount of work you need to spend drilling

[mactec54]

Here is an image of what I plan on building. The base (24x36x4) and gantry (18x24x3) are both granite surface plates. I will have a U shape cut out of the gantry surface plate by waterjet. Thank you all for the feedback. I think I will go ahead and use the mix of aluminum and granite in this design. I am not looking forward to drilling the 60 or so holes that are needed in the granite.

Holes are a normal thing that everyone has to do when building a machine with good diamond drills it should not take long at all