Machine base for manual bench mill: epoxy granite vs concrete

[maxki]

I'm building a steel tube cabinet who's main function will be to support a manual bench mill.

I'm trying to decide which material I will use for the top plate on the top of the cabinet, on which I will bolt the mill base.

My choices of material have narrowed down to:

1. epoxy granite
2. "specialized" grout or concrete (a kind that is available at driving distance)
3. "hardware store" / cheap concrete

The slab will be 18"x24", and the thickness will depend on the material chosen, to obtain sufficient stiffness, massm, and vibration dampening.

Inside the slab/plate will be a rectangular steel frame of 2" x 4" steel C-beams ( more exactly 4mm x 10mm, with 5 mm thickness), and also a rebar matrix.


I believe all of the above candidate materials can easily satisfy the "mass contribution" strength and stiffness criteria, but in terms of vibration dampening, I suspect that epoxy granite is a far superior option.

I have read enough about EG on this site to know that it's quite a challenge for a DiY to fine tune the mix and to get results anywhere close to "professional" EG.

But I suspect that my application is forgiving: stiffness and strength is already provided by the steel frame, the mill only weights 250kg, I can exagerate on the rebar for cheap.

And I can also take "shortcuts" to make things easier, like putting a bit more epoxy than the optimal, to ease the mixing, and air purging, vibraiting, etc.

From what I understand, the downside of a higher than optimal epoxy %, is loss of compression strength, and lower stiffness, but is better at vibration dampening.

For my application, perhaps an off the shelf "specialized grout", might be close enough, cheaper and easier.

I have found this grout, specialy designef for machine bases: https://can.sika.com/en/construction...routpakle.html

But it costs an arm and a leg ($648 for 28.32 Liters)

There is this other grout, also easily available, with a much more reasonable price:

https://can.sika.com/en/construction...grout-212.html

But the technical sheet says nothing of the vibration dampening, I suspect it's very close to "normal" concrete, and the only added benefit is that it doesn't shrink while curing.

Another reason I'm tempted to go through the trouble of Epoxy granite, is to gain experience with the material, and eventually use it for other projects: filling the mill column, making a CNC mill with it, etc.

So, even if EG is a bit overkill for a mill base, I will gain valuable knowledge and experience.

Similar Threads:

• Epoxy-Granite machine bases (was Polymer concrete frame?)
• Newbie- Do it yourself or buy factory machine bases made of polymer concrete / epoxy granite?
• Epoxy Granite In Practice (Mineral Casting, Polymer Concrete)
• Epoxy granite or UHPFRC Concrete 5 axis CNC
• Custom made precision mill with epoxy-granite base

[pippin88]

Bolting a manual mill to it?

Just build a steel bench

Or maybe regular concrete. Do not waste money on specialist grout or EG.

The base will not contribute strength/stiffness to an existing mill. To do that would require very rigidly bolting the mill. You would need precision mounting surfaces on your base. The mill base mounting areas are unlikely to be precise as they are not usually made for precision mounting. You would likely distort the mill base.

[maxki]

The base of the mill will be shimmed (if required) against the slab, before bolting, in order to prevent distortion of the base, and to distribute load evenly.

A slab of epoxy granite, or even concrete will have greater dampening than a steel plate, and also cheaper, so why not benefit from greater vibration dampening ?

Bolting a manual mill to it?

Just build a steel bench

Or maybe regular concrete. Do not waste money on specialist grout or EG.

The base will not contribute strength/stiffness to an existing mill. To do that would require very rigidly bolting the mill. You would need precision mounting surfaces on your base. The mill base mounting areas are unlikely to be precise as they are not usually made for precision mounting. You would likely distort the mill base.

[joeavaerage]

Hi,

A slab of epoxy granite, or even concrete will have greater dampening than a steel plate,

Vibration damping???....but the table on which the mill sits is not in the load path. The load path is from the tool, to the spindle, to the Z axis, through the column
to the base, through the Y axis , saddle , X axis and vice. Any component in that load path that has some damping capacity might be useful, but the table on which its sits
has absolutely no bearing on it at all.

For all that its worth the table could be made out of wood.

Craig

[maxki]

Hi,

Vibration damping???....but the table on which the mill sits is not in the load path. The load path is from the tool, to the spindle, to the Z axis, through the column
to the base, through the Y axis , saddle , X axis and vice. Any component in that load path that has some damping capacity might be useful, but the table on which its sits
has absolutely no bearing on it at all.

For all that its worth the table could be made out of wood.

Craig

My understanding is that if a component can receive vibrations (and vibrate as a result), it will also transmit back the vibrations, following the inverse path.

All the vibration energy dampened by such component, will not be transmitted back, and the overall vibration will be reduced.


If it weren't so, filling cavities of a machine tool would have no effect, because the cavities are "not in the path" as you say.

The epoxy granite in cavities receive energy, dampens a fraction of it, and that fraction is subtracted from the overall vibration.


I haven't done any simulation or even calculations, perhaps the vibration reduction is insignificant, but I need a plate for the top of the cabinet, and if I was subconsiously looking for an excuse to mold some epoxy granite, I found one ! ;-)

[joeavaerage]

Hi,

If it weren't so, filling cavities of a machine tool would have no effect, because the cavities are "not in the path" as you say.

BS. The filling material must adhere to the inside of the section, and then any flexure in the tube is transmitted to the filler material which in turn dissapates energy. This is the problem with filling tubes
is that the filler material seldom adheres sufficiently to the inner of the tube to effectively participate in any damping. If you want stiffer.......then make the walls of the tube
thicker. When all said and done if a structure is stiff enough is does not vibrate worth a damn anyway.

and if I was subconsiously looking for an excuse to mold some epoxy granite, I found one ! ;-)

If you want to try epoxy granite then try it, trying to suggest that it can dampen the vibration of a machine sitting on top of it is not correct. You could just as well make a cube of epoxy granite
and sit it on the other side of the workshop for all the vibration damping it will do on your machine.

Craig

[peteeng]

Hi Max - The top of the cabinet as you describe it, although it could vibrate it probably won't. It is also not in the critical path so even if it did vibrate it would not affect the tool very much at all as the entire cabinet and machine would vibrate in unison. Its the vibrations between the tool and the work piece that you have to minimise. EG is an expensive material. Spend the $$$ elsewhere on the project.

Materials do not absorb that much energy when you look at the big picture. What you really need to concentrate on is any structure that is a cantilever & then make it as stiff as possible. The machine column, the machine Z axis etc. These are critical elements prone to vibration. A large flat plate holding the machine up is not. It may add mass but that's an expensive way to add mass. Peter

The sika 212 is double compensated and will be as good as you can get out of a bag... unfortunately the data sheet does not spec its modulus. I went thru my correspondence with Sika and the 212 has a modulus of E=25GPa if that's helpful. If you mix blue metal with it you will lift the E to around 35Gpa with a good blue metal loading.. Peter

Looking at the epoxy grout modulus its 15GPa so go with the concrete grout...plus it will be considerably cheaper...

[peteeng]

Hi Max - Re reading your #1 I think you are conceptualising "vibration absorption" incorrectly. If you want to "absorb" vibrations at the contact between the machine and its base then you need to use rubber feet designed for machine mounting. These are readily available. This isolates the machine vibrations from the floor so you don't annoy the neighbours or upset sensitive machines next to the mill. Using damp materials in the machine parts means that if the part does vibrate (and all things vibrate) its motion instead of taking say 30-50 cycles to disappear it only takes 4-5 cycles. The vibration (motion or deflection say like a tuning fork) gets turned into heat as the internal parts of the component rub against each other. If you grab a paper clip and bend it back and forth a few times then feel it you will feel the heat generated by the motion. That's what damp materials do. Its called hysteresis. Both EG and grout being made up of multi materials having lots of interfaces generate a lot of internal friction vs metals that have very intimate grain boundaries that transfer load a bit more effectively. The EG and grout do not have an initial linear modulus. Its sort of a gentle curve vs metals which are very straight until they yield. This creates difficulties in how the modulus is determined. If you measure it over a wide strain then the modulus is less than if you measure it close to zero. The dynamic modulus as its called by concrete people (me being a mech eng call this something else) is measured by ultrasound techniques. If you read the notes prior you will see that measuring it using ultrasound results in a slightly higher value then by mechanical means....

[maxki]

Thanks peteeng, those are great answers, and the article is exactly the kind I'm wanting to read these days.

I'll go with the Sika 212.

Now I'm trying to decide on the slab thickness, 2 bags will yield a 3.8" thick slab, while 3 bags will give a 5.6" slab (in reality both slabs will be a bit thicker if I account the rebar volume).

The slab dimension is only 18" x 24", and it will have plenty of steel in it, so strength wise, I think 3.8" is enough, but modeling the thing is above my expertise, and using my own "guesstimates" I'd be tempted to go with 3 bags just to be on the safe side, but if 2 is enough, it has a more "friendly" weight of 50kg instead of 75 kg, I will be thankful when I need to move it.

I also suspect that a 3.8" slab weighting 50kg is plenty enough in terms of added mass, especially with the steel structure underneath it.


I can also go with 2.5 bags (and have left overs for another project), it will give a 4.735" slab.

Hi Max - The top of the cabinet as you describe it, although it could vibrate it probably won't. It is also not in the critical path so even if it did vibrate it would not affect the tool very much at all as the entire cabinet and machine would vibrate in unison. Its the vibrations between the tool and the work piece that you have to minimise. EG is an expensive material. Spend the $$$ elsewhere on the project.

Materials do not absorb that much energy when you look at the big picture. What you really need to concentrate on is any structure that is a cantilever & then make it as stiff as possible. The machine column, the machine Z axis etc. These are critical elements prone to vibration. A large flat plate holding the machine up is not. It may add mass but that's an expensive way to add mass. Peter

The sika 212 is double compensated and will be as good as you can get out of a bag... unfortunately the data sheet does not spec its modulus. I went thru my correspondence with Sika and the 212 has a modulus of E=25GPa if that's helpful. If you mix blue metal with it you will lift the E to around 35Gpa with a good blue metal loading.. Peter

Looking at the epoxy grout modulus its 15GPa so go with the concrete grout...plus it will be considerably cheaper...

[peteeng]

Hi Max - Be careful using traditional steel reinforcement. I would not use steel in it at all... If it is connected to the frame in such a way that it is in the loadpath then you defeat the purpose of the "damp" slab. Thats because loads take the path of least resistance which in structures is the stiffest path. So the load will preferentially travel down the steel. If your going to use grout for machine builds use a steel fibre reinforcement or a glass fibre reinforcement that creates a more homogenous mass with no direct load paths (except thru the mass). Counter tops use glass fibre for instance and is readily available. Sika have both as well... Peter

reo in structural slabs is used for different reasons to in machine parts. Its the tensile strength and crack bridging component. Machine parts do not need high strength and grout should not crack. But using a glass fibre is a good solution to improve strength and stiffness.

Since its a table top and mass maybe an issue consider adding pearlite to lighten it. Plus I expect this will improve its dampening properties... A thicker but lighter slab will be stiffer. Geometry wins.

Overview of Perlite Concrete - Perlite Institute

or look up aircrete and foam it a little.... or make the entire table from a chunk of aircrete!

[maxki]

Yikes, not using steel is a bit scary, I think I need to do a whole lot of reading to know which fibers to select, how much to put, etc, It's certainly less "idiot proof" than steel !

Weight is not so much a concern, as long as it's not above 100kg, which seems easy to achieve, even with steel.

For vibrations, it seems that, given that the slab will have to contact the steel cabinet, vibrations will make their way, even with a slab without steel, Unless I put rubber between the cabinet and slab, in which case, the steel in the slab will loose it's detrimental effect.

A steel frame inside the slab also has the advantage that I can bolt 4 screws, near each corner of the slab, that will become the legs of the slab, who's height that can each have a bolt and a metal plate under the bolt, to interface the cabinet structure.

I believe the name for such a height adjustment system is "jacks crews".

Thanks for you suggestions, I appreciate the "thinking outside the box", and the expertise !

Hi Max - Be careful using traditional steel reinforcement. I would not use steel in it at all... If it is connected to the frame in such a way that it is in the loadpath then you defeat the purpose of the "damp" slab. Thats because loads take the path of least resistance which in structures is the stiffest path. So the load will preferentially travel down the steel. If your going to use grout for machine builds use a steel fibre reinforcement or a glass fibre reinforcement that creates a more homogenous mass with no direct load paths (except thru the mass). Counter tops use glass fibre for instance and is readily available. Sika have both as well... Peter

reo in structural slabs is used for different reasons to in machine parts. Its the tensile strength and crack bridging component. Machine parts do not need high strength and grout should not crack. But using a glass fibre is a good solution to improve strength and stiffness.

Since its a table top and mass maybe an issue consider adding pearlite to lighten it. Plus I expect this will improve its dampening properties... A thicker but lighter slab will be stiffer. Geometry wins.

Overview of Perlite Concrete - Perlite Institute

or look up aircrete and foam it a little.... or make the entire table from a chunk of aircrete!

[peteeng]

Hi Max - If you have a steel frame inside the slab you have neutered the whole idea of using the slab as a damper. The load will stay in the steel as steel is 200GPa and your grout is 30GPa. So you have just added mass nothing else. may as well make the frame form thicker steel to achieve the mass. Since yuou like reading heres some more info that covers some of this stuff. Peter

Notes - p20 states that for a very damp machine "material factors .... are too low"

By the way - there are many companies now that use steel fibres in their structural concrete with no steel mesh. This has allowed for very rapid construction as laying mesh is time consuming and is something that needs to be checked etc etc. Forming up and pouring in "reinforced" concrete is very easy... Peter


https://fibermesh.com/

[joeavaerage]

Hi,
it seems to me that you are making a large project for yourself for little or no benefit.

It has already been established that the table on which you sit the machine will have little to no influence on the machine vibrations. You can literally make the table top out of
wood. A bunch of 100x50's glued together will form a strong (enough) beautifully damped table top to which you can screw the machine, and way cheaper than any UHPC or EG.
Save your money for new carbide tools, or that Kurt vice you always wanted, or a coolant tank and pump....or whatever. There are many things to spend money and time on
but to waste it on a table top???

Craig

[peteeng]

Hi - I agree a timber plinth is all that's needed. Audiophiles place their turntables on timber plinths and they swear by it. Peter

[maxki]

Hi Max - If you have a steel frame inside the slab you have neutered the whole idea of using the slab as a damper. The load will stay in the steel as steel is 200GPa and your grout is 30GPa. So you have just added mass nothing else. may as well make the frame form thicker steel to achieve the mass. Since yuou like reading heres some more info that covers some of this stuff. Peter

Notes - p20 states that for a very damp machine "material factors .... are too low"

By the way - there are many companies now that use steel fibres in their structural concrete with no steel mesh. This has allowed for very rapid construction as laying mesh is time consuming and is something that needs to be checked etc etc. Forming up and pouring in "reinforced" concrete is very easy... Peter


https://fibermesh.com/

You are probably right that a steel skeleton in the concrete (or rebar), will cancel any dampening benefits gained from using concrete.

The question I'm now asking myself is how much effort is it worth spending to get a "high performance" slab ?

Joeaverage suggests (and you seem to be in agreement), that even a maximaly dampening slab would not add that much benefit.

Which makes me favor low cost, low effort solutions.

Two bags of sika-212 cost $45 CAD total, obtained with 10 minutes of driving, and the steel is free (from my scrap pile), the "low cost" is satisfied.

Now, if throwing a bag of fibers in the mix can do the same as a steel skeleton, it would be added benefit at low cost, so I'm all for it. Same for throwing a bag of perlite.

My reluctance comes from my inability to predict anything about the slab, given the fibers (and or perlite) I can obtain.

Also, "professional" fibers (the ones that come with technical sheets) are kind of hard to obtain: the local "specialty store" that sells the sika exotic products, doesn't hold them in stock, and could not offer any guidance into which of the dozen sika fibers would suite my project, let alone how much I need to use.

I see a lot of low cost fibers on Amazon, but they tend to not have technical sheets.

Is there a process/algorithm to concoct a fiber+concrete recipe, for a slab design, which can use cheap/generic commodity fibers ?

If yes, then a fiber concrete slab remains "low cost/effort", for an elegant solution.

Otherwise it becomes a "project" in itself, and the suboptimal "slab with steel skeleton" design becomes tempting.


BTW, thanks for the "Topic14-DesignOfMachineStructures.pdf?", really interesting stuff !

[joeavaerage]

Hi,
just as a thought experiment imagine the machine sitting on a concrete floor, not bolted to it, but just sitting on it. Would the machine work? Would it vibrate when cutting?

Now imagine that you bolted it down to the floor. Would anything change?

If the base of the machine were flexing, say twisting and bending, then bolting it to something solid would restrain some of that flexure. If however the base of the mill is not flexing,
or at least only microscopically then bolting to the floor or in fact any other solid object will have no effect.

The four major sources of compliance (flexure) in any mill is the column, the headstock, ie the Z axis saddle and extension to the spindle, the Y axis, saddle and X axis, and lastly the base.
By far and away the worst (for flexure) is the column, closely followed by the headstock, with the stiffest and least flexure occurring in the base.

If the base is bolted to some solid object then some of that flexure will be transmitted to that object and the properties (stiffness, damping) will contribute to those same properties of the base.
If however the base does not flex then no flexure is transmitted to the object to which it is bolted.

The column on the other hand is not constrained in any way by bolting the machine down....so the column will continue to flex and vibrate irrespective of whether the machine is bolted down or not.
The headstock is similar. Does bolting the machine down in any way restrain the headstock????? No, it will therefore flex and vibrate as normal.

If you follow that argument then you can place the machine anywhere you like, say on a table. If the machine is accelerating back and forth the machine might be inclined to topple over
so it is usual to bolt a desktop machine to the desk. Having the desk or table be reasonably weighty will reduce the amount the combination sways around under axis acceleration.
It is not entirely a waste of time to make the desk/table heavy, but neither is it critical.

Craig

[peteeng]

Hi Max - You seem to be fixated on a slab for some reason. Buy damped machine feet and that's money well spent. This slab/plinth thing is going in the wrong direction? If its a test case for machine parts maybe a good idea but in terms of machining benefit you won't get any. Re: inability to predict. With all of these materials nearly everyone is unable to predict. I do dynamic analysis for a couple of companies and I'm always telling them this is fuzzy stuff unless they commit to a serious materials testing program (elastic and sonic). Which they don't do due to $$$... so even us pros are guessing at some stuff... If you have simulation that does modal analysis it will calculate the critical (natural) vibration modes of your structure & parts. The next step is to add damping and unless you have real data on that for the material you/everyone is guessing...

If you ran the monolithic slab you will see that its modes are well away from any of the machine modes. So the plinth is not doing anything to "absorb" vibes in regard to helping the machine quality of cut... The dampest solution is to make a box and fill it with sand. Any "solid" has quite a low damping ratio. This is explained in the Topic 14 article. But the load does not travel through your plinth. The machine loadpath is between the tool and the part (ie the machine) that's the path you need to investigate.

So 1) define exactly what you are trying to achieve here? 2) if your doing something right the forum is very quiet, if your doing something wrong the noise is very load! Peter

In regard to machine mounting there are two camps:
1) The machine is mounted free on its own feet. This camp relies on the machine mass for dynamic stability. The machine does not accelerate fast enough (generate high enough inertia loads) to move the machine mass. The bigger the machine the more it moves into this territory. There are some mill makers for instance that expressly forbid machines to be connected to the floor, probably for thermal reasons. But there are other makers that recommend the machine is bolted to the floor for better stiffness and stability. My 300kg router if I ramp up the accles will dance easily...
2) Bolt the machine to the earth. This effectively makes your machine infinite mass. Some makers recommend this and provide machine feet that allow this. This solves the dancing machine issue if the machine is light and has high accel potential. This is particular to fast small/medium machines and most robots are screwed to the planet as they move very fast and are very light... Keep at it. Peter

[maxki]

What is it that I'm trying to achieve is actualy simple: I just want to build a cabinet to put my new bench mill ! ;-)

The concrete slab is first and foremost a solution to the problem of needing a top plate on the cabinet to rest the mill on !

I thought that the slab could have a secondary benefit of damping.

Now thanks to you and Craig, I know the slab won't provide much damping, and that the need for it isn't much there to begin with.

But I still need a plate for my cabinet, I excluded wood, because it becomes a mess with the oils, coolant, etc, and there can be occasional sparkles, etc.

Given that a concrete slab is cheaper than a metal plate, I think I will go with the steel reinforced slab desing, because it's a no brainer to design and to get done.

It will be more or less like this :


Thanks again for the great advices !

Hi Max - You seem to be fixated on a slab for some reason. Buy damped machine feet and that's money well spent. This slab/plinth thing is going in the wrong direction? If its a test case for machine parts maybe a good idea but in terms of machining benefit you won't get any. Re: inability to predict. With all of these materials nearly everyone is unable to predict. I do dynamic analysis for a couple of companies and I'm always telling them this is fuzzy stuff unless they commit to a serious materials testing program (elastic and sonic). Which they don't do due to $$$... so even us pros are guessing at some stuff... If you have simulation that does modal analysis it will calculate the critical (natural) vibration modes of your structure & parts. The next step is to add damping and unless you have real data on that for the material you/everyone is guessing...

If you ran the monolithic slab you will see that its modes are well away from any of the machine modes. So the plinth is not doing anything to "absorb" vibes in regard to helping the machine quality of cut... The dampest solution is to make a box and fill it with sand. Any "solid" has quite a low damping ratio. This is explained in the Topic 14 article. But the load does not travel through your plinth. The machine loadpath is between the tool and the part (ie the machine) that's the path you need to investigate.

So 1) define exactly what you are trying to achieve here? 2) if your doing something right the forum is very quiet, if your doing something wrong the noise is very load! Peter

In regard to machine mounting there are two camps:
1) The machine is mounted free on its own feet. This camp relies on the machine mass for dynamic stability. The machine does not accelerate fast enough (generate high enough inertia loads) to move the machine mass. The bigger the machine the more it moves into this territory. There are some mill makers for instance that expressly forbid machines to be connected to the floor, probably for thermal reasons. But there are other makers that recommend the machine is bolted to the floor for better stiffness and stability. My 300kg router if I ramp up the accles will dance easily...
2) Bolt the machine to the earth. This effectively makes your machine infinite mass. Some makers recommend this and provide machine feet that allow this. This solves the dancing machine issue if the machine is light and has high accel potential. This is particular to fast small/medium machines and most robots are screwed to the planet as they move very fast and are very light... Keep at it. Peter

[peteeng]

Hi Max - Good video - consider it practice for when you cast your next machine! Use lots of wax on the mould... Peter

[joeavaerage]

Hi,
if you are using coolant you need a much bigger table top. You need sides to surround the whole machine otherwise coolant will go EVERYWHERE. I had flood coolant on my first
mini-mill but it was not particularly well thought out.........and I seemed to spend a lot of time cleaning up coolant leaks.

My new mill is vastly better, but still not as good as it should be. Hopefully sometime over the next twelve months I'll get to re-do the coolant system. I want a 100l tank, a pull-out tray filter and be able to remove the tank easily
to clean it out completely. I promise you'll have to do it every once in a while....and making it easy to do is a priority. I did my coolant tank over the weekend (annual event), took several hours, where with good design I should
be able to do it in under and hour.

Craig