Filled aluminium extrusion frame, good or bad idea?

[Xenofon9]

Hi everyone,

I have build myself a CNC router (see image).
The frame is constructed from aluminium extrusions.
I mostly mill aluminium parts on this machine.

The frame is quite stiff, however vibrations are an issue sometimes, especially on heavy cuts.
Basically the frame acts as a giant tuning fork, promoting chatter a lot.

I am moving apartments in a few weeks so I have to take the machine to parts (again....)
So I thought this is a good time (after dismantling) to maybe try and improve the frame.

I was thinking to fill the extrusions with either sand, or epoxy granite.
Do you think that's a good idea? and if so which of the 2 options would you suggest?
Oh and any more suggestions are very welcome

Thanks in advance!

Similar Threads:

• Retro-Fit... Good idea or bad idea?
• Aluminum plate or extrusion for frame??
• Need Help!- Good or bad idea? Concreate!
• Using X-Channel For Rails Good or Bad Idea?

[JohnZ]

First I want to say, very nice clean construction.

For advice, I'd ask a question... how is both your gantry and table built. From the view provided, we cannot see behind the aluminum plate on the gantry, or the bellows protecting the structure under the table.

If it is indeed vibrations and not overall flexing of the structure, you may be able to reduce the vibrations with constrained layer dampening. For CLD, you would sandwich a second layer of aluminum with epoxy between the two. This would be where you have an area of aluminum spanning between extrusions or other supports. Basically where you have an aluminum plate acting as a drum skin. I suspect the area under the bellows may be the source of your vibrations. To locate where the vibrations are occurring, get a mechanics stethoscope, about $10, and probe around your structure. It could be one of your aluminum plates, the gantry, table, or even the large flat plexiglass plates in the frame. But probe around and find the source of the most noise, and probe the milling table and your Z axis to see if the root source is the spindle mount, or the table under your vice. Then tackle both the source of the vibrations, and the area making it the most audible.

Next, I would take a look at where you could stiffen up the overall design of your machine. Consider adding a plate to the bottom of the gantry that is deeper than your extrusions behind the existing vertical plate across the gantry. Then wide triangular plates supporting the vertical side supports of the gantry, and bolted to the horizontal and vertical extrusions. Other areas would be whatever is under the bellows that we cannot see. I suspect ample areas of opportunity to increase stiffness there as well.

I hope that this helps.

John Z

Sent from my SM-G991U using Tapatalk

[Xenofon9]

Thanks for your compliment!
Yeah I realize there's somewhat limited view of the construction.
I attached some images of the design to help.
The backbone of the construction is all aluminium extrusion. The aluminium plates are bolted on the backbone.
They are tooling plates so fairly flat and constant thickness so I also use them as reference surfaces for some of the machine alignment steps.

To come back to the point, no doubt the frame can (and perhaps should) be stiffened.
Your idea of using a mechanics stethoscope is a brilliant one by the way!

Stiffening the frame by adding or adjusting the design is one thing.
Filling the extrusions with epoxy granite (or sand) is another thing that will require a complete strip down. Which I dont wanna have to do more than once
So it's gonna be now or never I think.

[JohnZ]

As I understand it, filling the central open cavity in the aluminum extrusion will help reduce vibrations, but very little for strength or stiffness.

Do you have an aluminum plate against the back side of the gantry?

Looking at your design, I can see an area with potential for creating very audible vibrations.

The plate on the face of the gantry is bolted close together along the outside edges of the top and bottom extrusions, but I see very few bolts along the inside edges of the extrusions, or on the center vertical member. Vibrations, especially if they hit a resonant frequency can cause the gantry plate to vibrate against those "unsupported" edges of the extrusion. Bolting the plate to the extrusion all along the inner slots can eliminate this from occurring and overall tighten up the assembly along here.

I would also consider side gussets that butt up against the bottom of the side plates on the gantry. This would be triangular shaped with the tip cut flat to match the existing side plates, then going down to the base Frame at a 30-45 degree angle, and is bolted along the side slots of the gantry upright, and along the base frame slots. This should significantly reduce the front to back flex of the gantry uprights.

Next would be to stiffen up the transition between the gantry uprights and the supports under the table. Perhaps a horizontal plate under the entire frame, spanning under both gantry uprights and the supports under the table.

It would be really interesting to see what you find out with a mechanics stethoscope and where the loudest noise is coming from.

I hope that this helps.

John Z

Sent from my SM-G991U using Tapatalk

[peteeng]

Hello Xeno - Before you do anything you need to identify actually what is vibrating & why. Filling things arbitrarily is a hit and miss approach. Plus are you using correct feeds and speeds so there is no skipping or sliding at the tool edges? I would check out your typical chip thicknesses and understand what is happening at the tool well before filling things with epoxy. Peter

[JohnZ]

Xeno, you really want to listen to Peter. He has done more deep dives into machine design, FEA analysis etc. Than I ever will.

Once you locate the area of your machine that is screaming with the mechanics stethoscope, let us know as that more than anything else can focus advice for what can be done to reduce or eliminate the issue.

Sent from my SM-G991U using Tapatalk

[peteeng]

Hi John - thks for the comment. Xeno I use a long screwdriver for listening to machines. Or a steel rod is good. I place the handle in my ear or on my jaw and the end on the thing you want to listen to. The machine is not "vibrating" per se. Its being exited by some input. Remove the input and the machine won't vibrate (roughly speaking)

Cutting aluminium with a high speed spindle means you have to cut very fast and clean otherwise the tool is skipping over its cutting edge. The tool supplier will give you a starting tooth load or chip thickness range.



If you know your chip load is correct and the chips are good and there is no rubbing then you can move onto the machine issues (unless the machine stops vibrating at this point) as you have eliminated this from the causes.

In another thread we spent a lot of time looking at construction extrusions and their free edges where the culprit in making a lot of noise. So putting screws into the ends to tie up the free edges helped. Peter

[Xenofon9]

John you are right, the gantry does not have a plate against the back of the extrusions, just the front plate you see.
I'm not really sure what you mean by the "inside edges of the extrusions".
The bolting strategy of the plate to the extrusions was a tricky one for me to be honest.

During construction I noticed that if the plate is bolted to multiple different extrusions (and the extrusions are pre-attached together) then there is a tendency for the aluminum plate to be conforming to the extrusions and therefore warping.
...and since I use it to mount rails/ball-srews that was no good news. So I started ditching bolts to have it as least "overconstrained" as possible.

Your suggestions for stiffening the frame are great! I will do as much of them as possible during the next rebuild

[Xenofon9]

Pete, thank you very much for the input. I appreciate taking the time.
You are right. I shall first try and diagnose than blindly shooting from the hip :P
I think I'll get that mechanics stethoscope as they are dirt cheap.

With regards to cutting parameters I have found success with a chip load of 0.025mm (1 thou) on aluminum for roughing. [6mm tool, 2 flutes, 16krpm, 1200mm/min feed, 6mm DOC ~15% WOC]
I haven't done a good experimentation on that yet. My tool vendor recommends ~0.1mm chip load. So maybe I should try that haha!

[peteeng]

Hi Xeno - 1200mm/min 16k rpm and 2F is 0.0375mm chip thickness . Try a single flute bit for roughing its easier to get to the target thickness with the high rpm spindles and has more gullet space so clears easier (the space between the teeth) CL=feed/(RPMxT) 1200/(16000*2)= 0.0375mm if the tool deflects even half the CL then it will slid not cut. This creates chatter. The thinner the CL the stiffer the machine must be to cut it. Publish some photos of your edge quality that will give us some clues. 1200/(12000*1) = 0.1mm a thick chip load has less chance of skipping. The tool must be sharp in all cases! Peter

Have a look at your 2F cutters with a magnifying glass and see if only one flute is cutting or one is wearing more or has pick up. You may find the 2F is behaving as a 1F....

you have a screw driver or piece of rod in your shop they are "free" I've found them to be more sensitive then a stethoscope.... up 2 U....

[peteeng]

Hi Xeno -starting at the tool. 1) check your runout as if its high then you are only cutting on one tooth 2) looking at your saddle its thin and the spindle is mounted below the bearings. This part of the saddle is a cantilever or overhang. It can vibrate very easily. The sides of the saddle need to have flanges added to stiffen it. The flanges need to be deep and go along the full length of the saddle. That's a good starting point. Peter