Mark's EPOXY Granite 18"x20"x20" (700mmx780mmx780mm) vertical mill build

My last project was converting a PM45m (RF45 clone). Exceeding the performance envelope in every way of my Rf45 clone is the base line goal while being on a budget.

Old mill I am going to replace with this build:
Attachment 269762[

I have been accumulating components over the past year. I have come to hate dovetails and am going with linear guides:

THK SHS30 for X axis 18”/700mm travel
THK SHS 20 for Y&Z axis 20”/780mm travel
I am going to be using either THK or NSK -32mm ground ballscrews
Epoxy granite for all major components and use of precision ground steel plates cast in place for the linear rails, servos, ballscrews etc.

Here is the basic design. I am using simple straight forward forms for the molds and incorporating plumbing in the casting to provide for lubrication and cabling:

Attachment 269742

Attachment 269744

THK Linear Guides- I have a longer section of rail coming Monday for the short one:
Attachment 269752

Here is a pic of the mold for the base with the 1/2" thick precision machined steel pads that are being cast in place for the linear guides:
Attachment 269754

The steel is ground to .001" on all sides and 1/2"x3"x36". Each bar weighs a lot. I am aligning the bars and bolting them to each other with the cross members embedded in the epoxy for strength. That way I can pre drill the holes and fix them precisely by bolting them through the mold using coupling nuts which will serve as anchors for them.

Attachment 269756

I have 5 gallons of US Composite 635 and slow hardener coming- I will try to finish the mold for the base this weekend and drill and align the steel pads for the linear guides.

More to follow but I will probably try to go with a 10hp spindle and 10k minimum rpm.

I am still finalizing the epoxy granite mix but have built molds with melanine coated board with composites previously.

This is going to be a heavy machine...

Thanks for looking- suggestions are very much appreciated.

Mark

that going to be heavy machine good size too look forward to seeing more

Quote:

---

Originally Posted by daniellyall

that going to be heavy machine good size too look forward to seeing more

---

Thanks for your interest. The base alone is 19"x37"x8" (482mm x 940mm x 203mm) deep epoxy granite with re-enforced steel and the ground plates embedded.

Looks interesting. How will you align the two segments perpendicular?

Are the linear guides purposely ordered do did you happen to pick them up for a good price?

I'm finding these epoxy builds fascinating, I'll also be following your progress with this, it sounds like a great project.

cheers, Ian

Quote:

---

Originally Posted by Voyager77

Looks interesting. How will you align the two segments perpendicular?

Are the linear guides purposely ordered do did you happen to pick them up for a good price?

---

The point of using the 2 big ground steel bars in the design is to provide a way to line everything up- the y axis rails, and the the column for the z axis both sit on the same precision ground bars to .001".

First I am aligning the two 3x36x1/2 precision ground bars parallel using a 12" granite surface plate. Then I will align them on the same plane by cross bolting 1 x 1 precision ground bars to fix them in place aligned.. Once cast into the epoxy base this assembly forms the basis for both the y axis and the vertical column for the z axis which will also use 2 ground steel bars aligned the same manner as the base.

Finally, I am going to cast in place a 15x15 ground plate on the bottom of the z axis. After reading about using grouting for alignment, I will use this method and have jack screws to do the final adjustments and then grout.

All the ground bars are being drilled and will be through bolted in the mold using coupling nuts to hold them in place.when I cast the epoxy granite. This should provide a precision ground and aligned surface to do final indicating of the rails.

Today I managed to drill the steel bars for the 30mm linear rails. I was worried about the prospect of drilling 18 holes through tool steel all aligned but it worked. I had to do 3 different setups on my rf45 clone for each rod. I bolted one of the linear rails direct to the table and indicated it straight and then used that as a straight edge for steel bars. Really made me wish I had a bigger mill! At the end, every single hole witnesses the linear rail holes.

Attachment 269906

After all of this is done, the column will mount on the bars cast into the design. The base of the column will be a precision ground 18x18x1 plate cast into the that will also sit on the rails - I have to drill those in a different pattern.

As far as the size of the linear rails, they were selected on the basis of the largest I could afford and fit in the design. I got all of them on Ebay and they were new stock. I averaged 200 per axis for rails and guide blocks.

The Y axis has the most forces on it because it has the x on top so I went with the extended version THK SHS30. They are going to be 15" for maximum rigidity. I can get a decent amount of deflection cutting steel on the RF45 and this is one area I wanted to really improve.

The X and Z have less force and going with the 20mm THK SHS to provide more range of motion.

Attachment 269904

Thanks for looking.

Mark

Are you going to machine a reference edge into your rail mounting bars? That is something THK recommends in their mounting guide.

bob

Quote:

---

Originally Posted by rowbare

Are you going to machine a reference edge into your rail mounting bars? That is something THK recommends in their mounting guide.

bob

---

I am going to use a ground steel bar bolted on top of the drilled bars pictured above for a reference edge. I only just came across this point and changed the size of the long bars to a wider size to allow room. It seemed easier than trying to machine a 3 foot step in a pre-ground bar.

FYI, there is a great pic of how this works in this installation guide link (Item 2): http://www.rockfordlinearmotion.com/...llation-guide/

gt40 are you sure that base and column are big enough? 482mm x 940mm x 203mm at 2,500kgm^-3 is only 230kg (507lb). I think you changed the Y travel to 15"??? This is better than the original 20" of Y travel, but a industrial machine with ~15" of Y travel will weigh 2,000kg+ (4,400lbs+) and still be considered light weight...

You mention your RF45 deflecting too much while cutting steel but how will this machine differ? Your build is obviously early stages but I can't see it being any more rigid than the RF45, and a 10hp spindle will only make matters worse... Getting 20" of travel from 780mm rails leaves only ~250mm carriage spacing (end to end) to support the table / head with.

Am I right in thinking you're expecting the steel plates to remain perfectly flat and parallel even when cast? Or are you going to scrape the plates flat or something once cast?

Sorry if I sound really negative, it seems to me that you haven't planned this out sufficiently. By the time you've finished you'll probably have spent enough to have bought a working used machine, which will be rigid and precise enough to do the work you want without all of the hassle of building the damn thing.

I hope this helps,
Fenza

Quote:

---

Originally Posted by Fenza CNC

gt40 are you sure that base and column are big enough? 482mm x 940mm x 203mm at 2,500kgm^-3 is only 230kg (507lb). I think you changed the Y travel to 15"??? This is better than the original 20" of Y travel, but a industrial machine with ~15" of Y travel will weigh 2,000kg+ (4,400lbs+) and still be considered light weight...

You mention your RF45 deflecting too much while cutting steel but how will this machine differ? Your build is obviously early stages but I can't see it being any more rigid than the RF45, and a 10hp spindle will only make matters worse... Getting 20" of travel from 780mm rails leaves only ~250mm carriage spacing (end to end) to support the table / head with.

Am I right in thinking you're expecting the steel plates to remain perfectly flat and parallel even when cast? Or are you going to scrape the plates flat or something once cast?

Sorry if I sound really negative, it seems to me that you haven't planned this out sufficiently. By the time you've finished you'll probably have spent enough to have bought a working used machine, which will be rigid and precise enough to do the work you want without all of the hassle of building the damn thing.

I hope this helps,
Fenza

---

I am only 4 years into cnc and it is at the serious hobby stage. I want to make some 3d stuff in limited quantities in steel and I want it to be more accurate than my RF45 clone- these things are the basis for doing this build. In a perfect world, I would have Brother Speedio but as the wife won't sign off on that just yet, this is the current upgrade.

As far as the design, I am trying to keep each major components at 700 lbs / 300 kg. It will also be bolted on a much larger subframe that will add considerably to the mass I expect the full mill with the subframe( still finalizing the design) will be around 2500 lbs/1130 kg. I got all my rails, thk and nsk 32mm ballscrews and basic materials for 2k so far. Add my time and spindle + electronics pneumatic draw bar and I hope to be well under 10k. I think if I can achieve this price point with the components I am using it should be a major step up from my current mill and hard to beat in terms of cutting ability, accuracy compared other mills in this price.

I am still tweaking the vertical column. 11x12x36 is the current dimensions and re enforced with large steel tube cast inside. Should be about 480lb with the steel inside. This is alot better than the 100Lb or so weight of the RF45 column but it could be bigger.

Most important, when I finish, I will know the machine inside and out and be able to maintain and fix it if need be. Modern VMC's can be very complex and expensive to fix and there is the learning curve. I also would rather start with new stuff than learning fix someone elses mess.

To your comparison with an RF45, I am not sure how familiar you are with these machines. The column is 7x9 tapered hollow cast Iron, as is the base. I can pick up and move the column, the base or the table by myself. The whole machine is only around 800 lbs/350 kg without the motor and stand. Going from a RF45 to this build will allow me to replace dovetails with THK linear rails, 3 times the mass and the room to fit proper sized ballscrews and servos.

I am still tweaking the vertical column. 11x12x36 is the current dimensions and re-enforced with large steel tube cast inside. Should be about 480lb with the steel inside. This is alot better than the 100Lb or so weight of the RF45 column but it could be bigger.

You brought up an interesting point on the steel rails. I tested a 1/2"thick piece a foot long cast at the bottom of a box mold placed on a granite surface table. It was dimensionally unchanged. The ground steel will be through bolted to the mold and cast in a horizontal position with the epoxy granite on top in the same manner as my test. We will see for the bigger piece. I guess I will scrape or grout as needed.

For my part, the biggest thing I am trying to work out is mixing 3 - 4 cubic feet of epoxy granite per pour for each section. I have planned for a total of 25 gallons.

Thanks for the challenging questions!

In my opinion, 10 hp is too much to expect in a light machine like the one you are building. 5 HP maximum, and even that is pushing the limits of your current design. My recommendation is 3hp especially since you are thinking of doing 3D work, wwhich 90% of the time is done with small cutters or ball mills and at high rpms.
I think you will be pleasantly surprised what you can do with 3hp.
I would concentrate on accuracy and speed of motions and spindle rpm.

Just my thoughts.

Oh, and one last thing, add a fillet or chamfer on you column where the base attachment points are. At 90degrees you have a weak connection there.

Quote:

---

Originally Posted by G59

In my opinion, 10 hp is too much to expect in a light machine like the one you are building. 5 HP maximum, and even that is pushing the limits of your current design. My recommendation is 3hp especially since you are thinking of doing 3D work, wwhich 90% of the time is done with small cutters or ball mills and at high rpms.
I think you will be pleasantly surprised what you can do with 3hp.
I would concentrate on accuracy and speed of motions and spindle rpm.

Just my thoughts.

---

Thanks for the suggestions. As far as the spindle size, I am happy to spend less money on a lower HP spindle but I have a 3 HP 8k capability on my RF 45 already and I wish I more power and rigidity. I also converted it to 30 taper with a Pneumatic bar. I will definitely get more than my current 3.

Quote:

---

Originally Posted by G59

Oh, and one last thing, add a fillet or chamfer on you column where the base attachment points are. At 90degrees you have a weak connection there.

---

Yeah, I did the cad sketch to layout things in relation to each other and it doesn't really have all the fine details. I understand about putting chamfers on corners like you said. Actual mold will have that on all internal corners too because it is easier to release from the mold as well as stronger.

Really appreciate the suggestions- keep them coming.

We'll good luck anyways. It's an interesting build and I look forward to the results.
for comparison, my 5hp mill weighs in at 2100kgs and similar work envelope, cast iron.
I'm not familiar with granite so I'll be watching and learning as you go along.

Out of curiosity, was that 8000rpm spindle driven 1:1 by that motor?
A 3 hp spindle has no problem doing a 0.25" DOC at full WOC, with a 2.5" face mill in 4140. Even less trouble in mild steel.

Quote:

---

Originally Posted by G59

Out of curiosity, was that 8000rpm spindle driven 1:1 by that motor?
A 3 hp spindle has no problem doing a 0.25" DOC at full WOC, with a 2.5" face mill in 4140. Even less trouble in mild steel.

---

Here is a pic of the motor:

Attachment 270098

Pulley setup:

Attachment 270100

Full build thread is on the zone here:

http://www.cnczone.com/forums/vertic...49951-cnc.html

Quote:

---

Originally Posted by gt40

Here is a pic of the motor:

Attachment 270098

Pulley setup:

Attachment 270100

Full build thread is on the zone here:

http://www.cnczone.com/forums/vertic...49951-cnc.html

---

We'll done on the PM45.
Anxious to see this build progress.

Update:

Out with the old, in with the new... The ballscrews I was looking at got sold but I found some bigger ones.

I got a 40mm THK ground screw and another ground ballscew for Harding that is a double nut. Both are larger than my original design so everything got supersized.

The new base is 42" by 22 3/4" x 9" deep I was able to get the mold almost ready to cast. All I have to do is drill holes for the ballscrew mounts and a few more for the column mounting.

Here are some pics:

Old base mold next to the new one:
Attachment 270778

Length:
Attachment 270766

Width:
Attachment 270768

THK 40mm ballscrew- this thing is really heavy, new old surplus-. It must weigh 50 lbs/22 kilos:
Attachment 270770

36mm ballscrew for harding machine. It should be arriving next week:

Attachment 270774

Here is a pic of the layout for the mold base. It is almost ready to go- I will be radiusing all corners in the mold with grout. The pipes are going to be cast in place lift points- stick a pipe or lift strap through them. Keep in mind all of these pics are from the perspective of the bottom of the base looking up or upside down:

Attachment 270776

Final shot showing the x axis linear rails laid over the mold to gauge foot print- The wider base will support the x axis better.
Attachment 270764

Thanks for looking

Looks interesting.

But I am also bit sceptical about the precise ground bars of steel you are going to use for rail support. My personal opinion is that it is waste. It will be quite impossible to set them to such accuracy as to not require machining (grounding) after the epoxy is hardened. Personally I would just use milled bars and then, when the casting is done, let it be machined on one clamping with column plate as to have ortogonality.

Well I will be interested in how you take the build on.

Double