Epoxy Granite Inquiry

Say I was building an industrial sized cnc mill out of EG. It is not practical to prepare an entire batch for one casting.

From that point there seem to be two reasonable options. Cast the frame in parts and bolt them together. Alternatively, still cast the whole frame, but in layers. So some percentage of the frame each 'layer'.

I am interested in the latter description. There are not many completed examples of EG machines, and the ones that do exist are not well documented.

Now after typing your ear off, I am wondering if anyone has experience with casting EG frame in layers? I imagine that it would be substantially weaker that casting the whole frame at one or bolting pieces of the frame, but that is just a guess, I honestly have no idea.

Thanks for reading through the essay, additional thanks to any responses, relevant or not.

One word: "rebar"

WolfmansBrother said:

Say I was building an industrial sized cnc mill out of EG. It is not practical to prepare an entire batch for one casting.

From that point there seem to be two reasonable options. Cast the frame in parts and bolt them together. Alternatively, still cast the whole frame, but in layers. So some percentage of the frame each 'layer'.

I am interested in the latter description. There are not many completed examples of EG machines, and the ones that do exist are not well documented.

Now after typing your ear off, I am wondering if anyone has experience with casting EG frame in layers? I imagine that it would be substantially weaker that casting the whole frame at one or bolting pieces of the frame, but that is just a guess, I honestly have no idea.

Thanks for reading through the essay, additional thanks to any responses, relevant or not.

Click to expand...

For this vary reason concrete is cast as a continuous casting, casting in layers will give you a discrete line/interface. There are work arounds like surface roughening rebar etc... Cast iron is awesome why would you screw with epoxy granite. Also why not fab it, should still be way better than epoxy granite.

very interesting.....please keep us posted with pics if possible of your build

IMHO its not as hard as it sounds, you just use a real slow setting hardener, most resin manufacturer’s offer a range of grades - hardening rates, have mixer or 2 near mold, mix, pour, rinse and repeat. compact as you go. Its really not hard to mix - continous pour concrete, most nuke base pads are done in one pour and there thousands of cubic meters, your talking a object what a dozen or two dozen mixer loads, should not be all that hard to do? Other options use a prepeg grade resin that needs a post mix bake to cure, a lot only need 60-80C so should not be too hard to do? The stone on a cold day + refrigerated resin will give you a month or 2 before they start to set up, hell even a week or 2 at room temps a common uncured time span!

Othere options is ditch mixing the resin and granite all together, simply vacuum impregnate it, its the common - every day composite process of choice to minimize excess resin, can't see why it would not work for this task?

Is worth adding, granite epoxy bases only work well with very high stone to resin ratios, nothing like a self levelling runny mix, think more like the granite barely has just a coating of resin and is very much relying on a lot of stone to stone contact to be stable, not stone - resin - stone contact. though its the later that very much holds it together, its the stone on stone near direct contact and high mass that gives it its thermal and mechanical stability.

adama said:

Its really not hard to mix - continous pour concrete, most nuke base pads are done in one pour and there thousands of cubic meters, your talking a object what a dozen or two dozen mixer loads, should not be all that hard to do? Other options use a prepeg grade resin that needs a post mix bake to cure, a lot only need 60-80C so should not be too hard to do? The stone on a cold day + refrigerated resin will give you a month or 2 before they start to set up, hell even a week or 2 at room temps a common uncured time span!

Click to expand...

Thanks for keeping it real.

Lots of EG threads on the zone.

WolfmansBrolther, where did that name come from?

One issue with a project like this is that when the epoxy cures (crosslinks), it gives off heat. Thick cross sections will cure at a different rate from exterior surfaces with the attendant problem of locked in stresses.

You would need to work closely with the compounder to get a slow curing mix to allow the temperatures to be as uniform as practical.

As is the case for any casting/molding, uniform sections are your friend.

Tom

machtool said:

Thanks for keeping it real.

Click to expand...

Whats the issue? Its bog std in the nuke world to do massive pours in one hit, its real common on a lot or smaller highly loaded commercial stuff to do one hit pours of over 100 truck loads. Can't picture him wanting more than a couple of cubic meters hence weres the problem so long as you have a resin with at least a 24hr pot life?

Prepeg resin is avalible to buy, its not like hes looking for a half mug of resin is it for a machine base? hence buying a few muti gallon tubs you can get a lot of options that aint going to be available on the shelf from locally suppliers. Getting a 1 ton mass to 60-80C is not too hard to do, just a simple metal enclsure and a propane heater will get you there in a hour or 2 for something bridgeport sized? Being able to take a few days to mix and compact it surely means he has no issue doing the pour in one hit?

lowCountryCamo said:

WolfmansBrolther, where did that name come from?

Click to expand...

Well, it was many years ago now....

One option is to cast the parts and to cement those together with epoxy. I am using this technique in constructing assemblies out of granite blocks and plates. If the two mating surfaces are flat (very flat, like a surface plate) the epoxy layer is in order of few microns and the joint stronger than the parent material.

billzweig said:

One option is to cast the parts and to cement those together with epoxy. I am using this technique in constructing assemblies out of granite blocks and plates. If the two mating surfaces are flat (very flat, like a surface plate) the epoxy layer is in order of few microns and the joint stronger than the parent material.

Click to expand...

Most engineering-grade epoxies have an optimal bond thickness, and I usually see 25 to 75µm values. I'd be interested who makes a formulation that's in the 2-4µm range. And flatness isn't as critical as surface prep (cleanliness, roughness, mix ratio, etc.) and chemical compatibility.

Milland said:

Most engineering-grade epoxies have an optimal bond thickness, and I usually see 25 to 75µm values.

Click to expand...

That's what I've seen as well, and more towards the higher end of the range. Applications that require uniform film thickness utilize spacer beads.

.004-.006" here. I think they were using plastic shims. May have been application specific in this case.

I've seen much thicker used to compensate differences in CTE between glass and metal housings.

If I were gluing granite surface plates and parallels together, I'd grind the bond area for texture and film thickness, leaving pads of the lapped surface for alignment. But I don't know any better.

Milland said:

Most engineering-grade epoxies have an optimal bond thickness, and I usually see 25 to 75µm values. I'd be interested who makes a formulation that's in the 2-4µm range. And flatness isn't as critical as surface prep (cleanliness, roughness, mix ratio, etc.) and chemical compatibility.

Click to expand...

In high precision assemblies (for example stages with sub-micron motion accuracy) and when cementing granite parallels etc in the assembly, the aim is to have a very thin layer of epoxy as epoxy has about 8 times larger coefficient of thermal expansion. Using low viscosity epoxy, the film thickness is controllable by the pressure. In test I found that the thickness is indeed a few microns and the joint strength (with proper preparation) higher than the granite: the joint breaks pulling the material and not at glue line.
I have no experience doing it with epoxy-granite, but being over 90% "granite" I think that epoxying separate assemblies should work - the glue thickness could be larger in this application.

Depending on the epoxy goop, they are indeed amazingly strong, and in any casy stronger than granite.

Granite is not "strong", whatever that means, but absorbs vibrations very well and has great crush resistance.

If you do plan to do something like this, I would advise first looking for or locating the rest of the bits you are going to use.
Even the cheap stuff, by the way, is quite expensive (for anything good).

Unless you are an experienced MT builder, of course, we have no idea.
Several gotchas, and I speak from lots of experience.

My first versions used concrete in some assys, about 11-12 years ago, and today I use tool steel.

There are or may be several good reasons to build tools - I have many reasons.
"Cheap" is not one.

In any case, if you are serious, use some cast iron as well and epoxy them together.
Then bolt onto the CI bits.
Hint .. mill the CI pieces flat first.
Std bolt holes tapped for linear guides are a good idea.
Alignment ledges.
Fixture holes.

There are a ton of threads about this on CNC zone but last I looked, all everybody does is mix some up in a baking pan. They have ideas about building bridge mills and other machines. They never get farther than that because they need precision equipment to make their molds.
I use EG a lot and a proper mix cannot be done by hand unless you are talking cup size batches or want to end up sweating with blistered hands. I use a 5hp electric mortar mixer and I had to add a second pulley reduction because it would almost stall with a 400 pound batch. And I do cast in layers but do it wet on wet, 400 pound batch to partially fill 4 molds, another 400 pound batch to top off.
A proper mix is very strong and stable. I poured some leftover mix into a plastic bin and popped out a 4 x 8 x 12 block. I had a 5 pound weldment that I wanted to break so I started pounding on it and was chunking out my concrete floor. I put it on that block expecting it to break but my pounding with a 3 pound hammer only nicked the surface slightly.
Your project will only be as good as your mold and you must think this through and include every threaded insert, threaded linear guide mounting rail, etc. Machining after cure requires diamond tooling and lots of water.
Also note that a concrete vibrator is useless on EG. The viscosity of a proper mix will not transmit the vibrations. A slow back and forth movement will.
Once I have my molds filled I roll them to an area where they are allowed to cure for 48 hours. I level the molds with shims under the wheels. They must be level because if not some of the mix will very slowly start pouring over the sides.
Once the pour is done it is time for cleaning the mixer. Steam or hot water pressure washer is best but I don't have that. I use a cold water pressure washer. First thing I do is to put about a half gallon of water into the mixer while it is running. One by one I throw in the empty aggregate paper bags. Then more with the pressure washer. I get off as much as I can and have a special dumping hopper I made and dump the whole mess into that. What is left will cure. I fill the molds about 10 times before the mixer really needs a hard cleaning. Ear and eye protection and an air hammer. Hand tools like my trowel or garden hoe get heated with a torch and the epoxy softens, burns and stinks, flakes off. I do that outdoors.

Scruffy, can you post pictures of what you are making?

OP said:

There are not many completed examples of EG machines, and the ones that do exist are not well documented.

Click to expand...

What are you talking about?
Hardinge, for one, made quite a few machines that way since the late 80's.

They apparently don't make as many with the process as they used to; my guess is that in the international sourcing market, it is no longer an inexpensive way to get to the complete machine. So some are now CI with Harcrete damping fills, and some are completely Harcrete where the properties are essential to the performance (mass, very low vibration).

SUPER-PRECISION® by Hardinge

billzweig said:

In high precision assemblies (for example stages with sub-micron motion accuracy) and when cementing granite parallels etc in the assembly, the aim is to have a very thin layer of epoxy as epoxy has about 8 times larger coefficient of thermal expansion. Using low viscosity epoxy, the film thickness is controllable by the pressure.

Click to expand...

Nope, you control it my adding a small amount of spacers of sutable dia to the epoxy on mixing, were talking like just a small pinch or so of sutable micro ballons or other compatible additives, you can not control glue joint thickness with just pressure in any kinda reliable way, add a pinch of spacers and hay presto you get that guaranteed minimal glue joint thats idiot proof and low agro to do. Pressure alone and its real easy to end up with a staved joint and your strength falls off, the above is a aircraft certified approach, though i believe it probably came from the performance boat world, its certainly been around for a very long time and is very much field doable in more or less any circumstances with any clamping means and totally reliable!