Newbie. Going to Build large VMC 800mmx500mm iron casting Frame filled with epoxy gra

[Chmfhm]

First of all sorry for my bad english as i am using google translator

Lets start

I have learned a lot from this community as well as from YouTube and online tutorials on cad in solidworks and nx and cam in mastercam and now I am quite comfortable with these software. I spent 16 hours daily in studying learning and working on the softwares and operating the machine

I bought a 3040z machine from china with linear rails 1.5 kw water cooled spindle and iron casting bed year ago. I bought as a hobbyist and to learn. I can cutting aluminum and mild steel with 3040z machine on 18000rpm with hsm style

And now i want to build a my own vmc with the help of this community members. I will upload my design pictures within couple of days with all details. I will also open a channel on youtube to upload videos during this build

Now I need some spindle information. I want to know if 12000rpm belt driven spindle can do high speed milling on steel? Mostly on p20 or sometimes 4140 steel and aluminum high tensile strength. And I also want to know if any belt driven spindle comes with rpm more than 12000rpm?

And last i let you know that i want to jump into tool making and mould die bussines and i'll start from here.

Similar Threads:

• New Machine Build- Large Epoxy granite VMC building
• Epoxy Granite VMC build (preview)
• Bare cast iron frame for machine build - any documented builds?
• New Machine Build- CNC lathe build, steel frame, HIWIN rails and epoxy granite headstock
• Quartz filled epoxy...

[awerby]

A 12k RPM spindle can cut steel with a very small tool, under 1 mm. But it's too fast for larger ones - optimum RPMs decrease with tool diameter. Some belt driven spindles can run at low RPMs without losing torque; the answer to your question depends on the specific spindle you're planning to use. But belt-driven spindles in general are better for this use than the self-contained spindles used on routers, which generally can't slow down much without losing torque and overheating.

[Chmfhm]

A 12k RPM spindle can cut steel with a very small tool, under 1 mm. But it's too fast for larger ones - optimum RPMs decrease with tool diameter. Some belt driven spindles can run at low RPMs without losing torque; the answer to your question depends on the specific spindle you're planning to use. But belt-driven spindles in general are better for this use than the self-contained spindles used on routers, which generally can't slow down much without losing torque and overheating.

Thx awerby for your prompt reply.

While working on cad designing base of vmc. I found alot of types linear rails. My plan is to put 45mm linear rails. But i cant finalize which type of liner rails. There are hgh hgw egh egw mgn and mgw types with ca ha or c series
I am expanding my vmc size 1100 into 600 to mill big parts. So would you or anyone on this forum would describe which type of liner rails will be suitable

[pippin88]

Roller (rather than ball) bearings are more rigid.

Every linear rail manufacturer has detailed information about loading ratings etc. You will need to become familiar with those.

[Chmfhm]

I have made this CAD design so far. The machine size will be 800×500×350. Cad Designing of the machine with measurements.

I plan to make it with epoxy granite with steel reinforcement inside with welded steel bar rods.

All yellowish color will be epoxy granite filled attached with steel plates in grey color

Specification will be

3205 c3 ball screws XYZ
Linear rails HGR35 on all axis
Servo motors 2.3kw 2500rpm on all axis
Direct Drive spindle bt40 12000rpm with 5.5kw servo driven
Y-axis plate made of cast iron or steel (need suggestion on this)

All item's Source is Alibaba

I need good suggestions according to my cad dimensions on both side pillar sizes and also need suggestions on the x-axis beam are they ok according to the design to mill steel mostly? Also need suggestion on the controller which will go best and easy to work with Mastercam Postprocessor for 5 axis as I have a plan to make turntable later

[peteeng]

Hi CHM - Since you have machining capabilities I suggest you make an all aluminium machine. EG is not stiff (relative to other materials) and if you use a welded steel internal you run the risk of poor load paths which will not take advantage of the full geometry. Steel is 200GPa and EG is 40Gpa stiffness at best. Strain travels down the stiffest path so making a design as you have explained can result in poor machine stiffness, compared to say the same part in aluminium. If you read my Milli thread it will cover a lot of materials. I have just had some carbon fibre epoxy laminates tested and they achieved 38GPa. Carbon fibre is significantly stiffer then granite and sand (typically 70GOPa and CF is 210GPa) so I can't see EG getting to aluminium stiffness or cost.

My current conclusion is that laminated metal is the stiffest and dampest approach. This also means no moulds are required which is a bonus for a one-of machine. What you have drawn is fine just do it in laminated aluminium and it will be at least twice as stiff as what you have described and have a similar weight and be easier to do.

Epoxy is more expensive by volume then aluminium as well... and the laminated metal will be more stable then EG and you can machine it anytime you like (eg for upgrades or modifications vs EG which will need specialist machining... you mention a 5 axis upgrade for instance. Peter

regarding linear rails - you will need high preload cars to achieve the best from your mechanics. Check that the supplier can provide these. Normal order cars will be zero preload or light and these will have slight to lots of hysteresis as they change direction. This leads to inaccuracy and to premature wear if highly loaded often. This is the usual case with a mill that you are describing...... Also if you look at commercial machines of the size you are designing the structural elements are much larger then yours and they are steel or cast iron so are significantly stiffer. In my experience you should start by designing the Z axis first as I think you will run out of space once you design a Z axis stiff enough for your purpose. By designing the frame first you now have limited the geometry available for the Z axis. But you have to start somewhere. So sort your spindle design and Z axis before you get too concerned about the frame.... eg in this size machine high rail designs are becoming the norm, the walls keep the swarf in and are significantly stiffer then single column designs....keep at it. Peter

[Chmfhm]

Hi CHM - Since you have machining capabilities I suggest you make an all aluminium machine. EG is not stiff (relative to other materials) and if you use a welded steel internal you run the risk of poor load paths which will not take advantage of the full geometry. Steel is 200GPa and EG is 40Gpa stiffness at best. Strain travels down the stiffest path so making a design as you have explained can result in poor machine stiffness, compared to say the same part in aluminium. If you read my Milli thread it will cover a lot of materials. I have just had some carbon fibre epoxy laminates tested and they achieved 38GPa. Carbon fibre is significantly stiffer then granite and sand (typically 70GOPa and CF is 210GPa) so I can't see EG getting to aluminium stiffness or cost.

My current conclusion is that laminated metal is the stiffest and dampest approach. This also means no moulds are required which is a bonus for a one-of machine. What you have drawn is fine just do it in laminated aluminium and it will be at least twice as stiff as what you have described and have a similar weight and be easier to do.

Epoxy is more expensive by volume then aluminium as well... and the laminated metal will be more stable then EG and you can machine it anytime you like (eg for upgrades or modifications vs EG which will need specialist machining... you mention a 5 axis upgrade for instance. Peter

regarding linear rails - you will need high preload cars to achieve the best from your mechanics. Check that the supplier can provide these. Normal order cars will be zero preload or light and these will have slight to lots of hysteresis as they change direction. This leads to inaccuracy and to premature wear if highly loaded often. This is the usual case with a mill that you are describing...... Also if you look at commercial machines of the size you are designing the structural elements are much larger then yours and they are steel or cast iron so are significantly stiffer. In my experience you should start by designing the Z axis first as I think you will run out of space once you design a Z axis stiff enough for your purpose. By designing the frame first you now have limited the geometry available for the Z axis. But you have to start somewhere. So sort your spindle design and Z axis before you get too concerned about the frame.... eg in this size machine high rail designs are becoming the norm, the walls keep the swarf in and are significantly stiffer then single column designs....keep at it. Peter

Thanks peter for looking in.

Well i am not familiar with metal laminated can you share link on this.

Also in our country i have access to aluminum oxide 320mesh.Which cost me about usd1.2 per kg

What's your thought on epoxy+ aluminum oxide instead of epoxy granite

[joeavaerage]

Hi,
epoxy-'anything' including aluminum oxide is about as stiff as a limp noodle 20GPa to 40GPa, and 40GPa can still only be achieved if you are lucky/good.

You can design and build epoxy-'anything' machines to be stiff but it requires much larger sections than if it were made of steel (205GPa), cast iron (110GPa) or aluminum (70GPa).

It turns out that epoxy-'anything' is actually quite an expensive build relative to steel and/or iron by virtue of the volume of materials. Steel can be cut, bent, cast, welded, drilled and tapped,
none of which you can do with epoxy-'anything'. Steel structures, certainly welded steel structures, require thermal stress relief and finishing machining, neither are cheap but very doable.
While you cannot/need not stress relive epoxy-'anything' you will have to finish machine the inserts for rails, ballscrew mounts etc. Thus the finish machining costs between both
technologies is similar.

Craig

[peteeng]

Hi CHM - I have made samples of parts with epoxy ALOX and they are effectively grinding wheels. Even though ALOX is 200GPa stiffness it does not translate into much part rigidity as the particles are spherical. eg the carbon fibre samples I made have a higher solids volume ratio and a higher modulus then ALOX samples I made yet only achieved 38GPa modulus. This has put me off the epoxy cast direction using minerals. Granite and other materials are 70Gpa so they won't achieve over 40Gpa. Plus all the commercial mineral cast data is via compression tests and this is biased as high solids volume constrained will give you a high modulus., If you constrain water in a container & compress it you get nearly infinite stiffness yet we can't build stuff from liquids.

If you have flown on an airbus plane lately the plane is probably made from GLARE a laminate of aluminium and glassfibre. GLARE - Wikipedia

Fiber-metal laminates in the spotlight | CompositesWorld

Constrained-layer damping - Wikipedia

These are thin sheet applications, for machine parts think about 12mm or 16mm thick plies to make whatever you need. Or thicker 25mm?

So 1) build from steel or aluminium. Ideally monolithic sections not thin tubing unless it is highly trussed. Do not worry about damping as this is a secondary consideration. Get the initial machine rigidity correct and vibration will not be a problem 2) If you want to minimise material vibration use laminated metal. This is called constrained layer damping. It provides superior damping to cast iron for instance. You can make parts in 16mm or 12mm plate, laser cut or machined on your mill then epoxy it together like a loaf of bread. Then finish machine. In this way you will need no stress relief or post heat treat. These parts will be stable and strong if done correctly. Stacked parts can be bolted together or dowelled. Think about it more like an IC engine design/build philosophy. Bolted connections are strong and damp.

I like aluminium as it does not rust, its less dense than steel and is easier to machine on my router. But if you can machine steel then make all the small parts, bond them together then paint. Then finish machine.

You can even have steel outer layers and aluminium inners to save weight as the steel outer will be providing the bending stiffness. Ideally have the steel zinc plated so it does not rust. Keep at it. Peter

[Chmfhm]

Hi Peter and Joe,

Thanks for your inputs.

@peteeng.

i go through your milli thread and hope you get what you want to achieve as i saw you spent 2 years on milli with alot of analysis.

Well Before EG casting my idea was to laser cut 15mm stell sheet then weld whole shell of machine with a joint which is in picture and filled with EG. whats your thought.

i can go into iron casting with whole machine. but problem is i have no source atleast 1200km far away to machine whole casting frame.

Whats your opinion on rigidity of machine according to my old thought

[peteeng]

Hi CHM - I think the idea of making a steel shell then filling with a very high cost low performance material is totally flawed (I'll get some flack over that). I think making a steel shell is a good idea but you must then stress relieve it and finish machine it. Iron casting has issues as well as CI needs to be post heat treated to stabilise it or you leave it in a paddock for a couple of years to settle down. All materials have +'s and -'s and you have to balance them out to a solution that suits you. I just reviewed some material costs and for me the cheapest is CSA grout, but it is low modulus and needs inserts and finishing. Since all build philosophies need final finishing to make it true and correct to the required accuracy, I feel that monolithic or laminated metal is the way to go.

Cold casting has become popular because of many reasons. These may not be the correct reasons for you. Laser cut steel is economic and accurate for the "rough" parts. Stay away from welding, huge can of worms unless you commit to thermal stress relief. Bonding and bolting "small" parts together to make big parts is a valid and common approach in my experience and side steps heat treatment and material stability issues as you start with stable materials. Machines have to be rigid and to be rigid you have to start with the stiffest materials to give yourself a chance of getting to a rigid machine... Hence alum or steel is the starting point IMHO...

You say you want to become a commercial toolmaker. The cheapest machine is one you buy ready to go. If you have a business model and it works then you lease a machine and get it earning... Building a toolmaker quality VMC is not for the faint hearted or for someone with shallow pockets. Tools and moulds require machines that can remove metal at very high MRR's otherwise it takes weeks or months to get to final sizes and then you don't make money. The size of machine you are describing means you are making moderate size moulds. Since moulds require 50% or more metal removed from the billet this means huge amounts of swarf so you have to think thru swarf management, enclosure design all things that others have already been through... Your journey has just started Peter

[joeavaerage]

Hi,
I agree with Pete, filling a steel shell is just a waste of filling material. The outer shell, being steel is as stiff as an old boot and will flex just a few um for a given force. The
filling material will have to comply, namely that same few um, but given it is a stiff as a limp noodle the energy transferred to it is minimal, and it potential vibration damping
is likewise minimal. If you were going to spend extra dollars do yourself a favour and just make the shell thicker. Rigidity beats vibration damping any day of the week.

I used cast iron, or rather had some grey iron cast for my new build (1 year) mill axis beds (3). It is expensive....approx $10NZD/kg....but it is good. I then used 32mm med tensile steel for the frame.
This was about the best compromise between stiffness, budget and with satisfying vibration damping properties.

I did consider epoxy-granite, but once you do the calculations you realise that cast iron and/or steel beat it for stiffness (at a given price) hands down.

Pete is also right, you need to spend a good deal of time and money on the enclosure, coolant pumps, filtration and chip handling. Even a small mill like mine seems to have me chasing
coolant leaks and chip blockages all day long!

Craig

[Chmfhm]

Hi Craig and Peter,

Thanks for your Valuable input.

In my country we mostly have old bridgeports of 80s or 90s mostly on which people do Tooling and mold works. They are very slow and quality of finished work is not good also had low tolerances of final product.

I want to be in this bussines but import of new machine cost is very high as 1usd is 250pkr

Cast iron price is usd1.4/ kg here.

So thats why i want to make it by myself to control price range.

As a starter i want to make one machine to check flaws and errors. As i want 5 machines in total.

My budget is USD 10k to 12k for 800x500x400mm table size machine. I come with some new ideas like makeing H beam style with flange size 20mm thicknes h beam size 400x300 and truss design on both sides in grey cast iron

[peteeng]

Hi CHM - Well design it out and cost it out before you do anything. CAD time is cheap and flexible and will resolve various issues. There's lots of hurdles to jump and rabbit holes to get lost in... Peter

[Chmfhm]

A question comes to my mind that if cost wise iron cast and Epoxy granite machine are made same at the end, then why Chinese companies and European companies are preferring Epoxy Granite nowadays.

Can someone give solid reason on this

[Chmfhm]

A question comes to my mind that if cost wise iron cast and Epoxy granite machine are made same at the end, then why Chinese companies and European companies are preferring Epoxy Granite nowadays.

Can someone give solid reason on this

[joeavaerage]

Hi,

Cast iron price is usd1.4/ kg here.

That may be the price of the metal but it's a small fraction of what it costs to cast something.

The price I quoted was for the three axis beds completed, a total of 345kg. It includes the price of getting the pattern made, getting the molding done, the actual pour
and the fettling afterwards.

A question comes to my mind that if cost wise iron cast and Epoxy granite machine are made same at the end, then why Chinese companies and European companies are preferring Epoxy Granite nowadays.

But are they? The vast majority of CNC machines are still cast iron and steel. Those few epoxy granite machines that are out there are actually more expensive than the same machine in cast iron.
In those machines where epoxy granite is used it typically some special property of epoxy granite that is sought, usually thermal properties.

Go into any tool and die making shop and see what they have....99% or more will be cast iron/steel.

Craig

[joeavaerage]

Hi,

My budget is USD 10k to 12k for 800x500x400mm table size machine.

That budget sounds very low. I spent that much and for a much, much smaller machine.

I was lucky and got three 32mm BNFN THK C5 ground ballscrews second hand for my project, if I'd had to buy new I would have paid something like $2500USD each. So just the ballscrews
will come close to your budget.

Craig

[peteeng]

Hi CHM - Production machine building has many aspects. Casting hot has issues such as you can't achieve various shapes or sizes. Post heat treat is needed or years of time to allow the structure to stabilise. Heat is energy and energy is now expensive. Plus scale comes into it. You can cold cast very large objects vs metal casting is limited in part size. Plus these days the cold cast production machines include wiring and cooling passageways that you can't achieve via hot casting are "cast" in. Usually, a foundry is something that is contracted and casting cold allows you to bring that operation back in house to control quality and cost. Foundries are considered "dirty" operations per the environment and the environmental costs of business are increasing. Freight from China to europe and US used to be cheap so they contracted Chinese foundries (who did/do not have environmental rules to comply with) but now freight has blown out to a huge cost fraction so companies are bringing work back in-house.

Cold casting solves many of these issues. 3D printing of large metal structures will overtake some of these processes.... They will print the tool blank these days vs billeting, then finish machine to reduce the swarf output and up front material cost. If I was a production machine builder, I would choose cold casting over hot casting hands down (which means CSA grout or mineral epoxy) ... But then cold has its own issues. Your issues of building a one of a kind machine within your resources are totally different to production machine builders issues. I bet you can't finish machine the rail foundations to the manufacturers spec as you need access to a much larger machine then the machine you are building to accomplish this operation. So you will be looking at epoxy levelling and things like that which is a very poor solution for your machine. To get to where you need to be you also have to start at the end and work backwards. Do you have a local heat treater? large scale machining access, cranes etc. The size of machine you are dealing with is easy to do in CAD but in the real world there are many hurdles for you.

You are in what is called the "fuzzy front end" of the project. The correct path has not yet been identified. You will filter it down eventually. Keep at it... Peter

By the way large H & I sections are optimised for long columns or long beams. They have excellent global bending stiffness but poor geometry and local stiffness for machine parts. Shear transfer and shear deflection dominate machine parts thats why they are very solid around rails and hard points vs thin and hollow. If you stay with plate parts you can optimise your geometry better then by using standard sections which are optimised for a different application. again - keep at it and keep chewing the elephant. Peter

[joeavaerage]

Hi,
your budget sounds more appropriate for a mini-mill rather than the machine size you are proposing.

It seems that every time you double the size of a machine the cost goes up by ten-fold or more.

With the budget you have I would suggest a very much smaller machine, one which would be satisfying and successful whereas a much larger machine would be so
compromised that its success would be unlikely. For example you could buy three 400W servos for the cost of one 2.3kW servo, instead of using 25mm or 32mm C3 ballscrews
at $2000-$3000 each you could use 16mm C5's or if you still need to economise C7's and get all three for less than $1000.

Craig