Epoxy-Granite machine bases (was Polymer concrete frame?)

_
A total disappointment. I rushed through the process and produced a sloppy dud!! Not only did I go over my weight estimate, but it's exactly 150lbs- can't ship Fedex!

http://i233.photobucket.com/albums/e...IMG_3017-1.jpg

http://i233.photobucket.com/albums/e...IMG_3026-1.jpg

Steel ways have shifted during casting, totally out of whack now

http://i233.photobucket.com/albums/e...IMG_3032-1.jpg

http://i233.photobucket.com/albums/e...IMG_3015-1.jpg

Here's an insult to injury: 5"x5"x10" foam core is showing through the back!!

http://i233.photobucket.com/albums/e...IMG_3036-1.jpg

Arrghhh!!!! I really need some helium now..

Graham's law covers effusion not diffusion. Effusion is where gases of different molecular weights flow through a small holes at different rates. Diffusion is the process by which gases with different concentrations on opposite sides of a semi-permeable membrane travel opposite the concentration gradient to reach equilibrium.

Diffusion is the the process by which a vacuum chamber will come to equilibrium with its contents due to the fact that there is vacuum in the chamber and trapped air in the part. The rate of diffusion is proportional to the concentration difference on the two sides.

Effusion is what would happen if you put a bottle of helium at atmospheric pressure in a chamber and placed a pinhole in the bottle and watched the helium atoms slowly leave through the hole and enter the chamber.

Helium is of interest here because it is capable of escaping through tiny holes that other substances are not. This is why it is used to test high vacuum systems for leaks. Hydrogen is even better at escaping through tiny holes but also flammable and can effect the chemistry of some things like metals.


Walter,

My condolences on the X2 column assembly. It looks like you had a lot going for you but that it just went wrong! I suppose this says that jigs to hold everything in place are going to be critical. The other thing is that judging by the place where the foam showed through, it looks like the intended E/G section was not thick enough.

It was definitely a brave step to make that part and I am sure that there are lessons we can all learn from it. I think the first lesson I'd take is to try to make castings as simple as possible. You are the Edison of Epoxy Granite right now and you just found the first of "a thousand things that did not work".

Despite the failure, it's a great service to the community that you did this and I want to congratulate you for trying such that we may all learn from this experiment.

--Cameron

Sorry to confuse matters. I used the word diffusion as it's commonly understood, whereas effusion is a bit more rareified (chair) (chair)

John

Hi Walter - Thank you for posting the pics. I think your actual need may be nitrous oxide.

I came up with a simple test for the He - 2 containers of honey, bubble air in one, He in the other till it has a froth. Turn off the bubbles and watch the time for foam reduction. I used to keep a cylinder of He in the garage, so that would have been a very fast test. Need to look around.

It is not so hard to calculate / estimate that He will be better from a simple diffusion perspective. If we were really hard core, I suspect that there are more complex bubble mechanisms esp. with air to deal with. I think this is the point where my old Chem Eng books would say "The proof is left as an exercise for the reader". :)

What I really need is a valium. I'm ready to throw this thing out the window!!

EDIT:
(I just googled nitrous oxide. That could definitely work)

Hi everybody!

I'm very impressed by your the work here, espetially about Walter's!

even if the X2 base if a failure, the portique (dunno the english word - I'm French) is realy impressive :rainfro:

also you could use this failure to test -if not already done- the resistance of the steel bar inserts to traction!

Personnally this is the major question I ask myself abour E/G.
I don't doubt it'll handle Klbs of load but I'm worrying about comportement of the inserts under side/facial traction...

any clues?

walter

I see that you use some kind of a thin plastic film for release agent i suppose. That just gave me an idea.
If you had a very elastic balloon you could fill it with E/G and get rid of trapped air in the corners. Kinda like injection molding or hydroforming.

Ok someone look at this;
http://www.fibreglast.com/showproduc...y-20.html#2430

Then checkout the mechanical properties of Epoxy 2000 / Hardener 2020 on page 2;
http://www.fibreglast.com/documents/458.pdf

This looks much better than EPOTUF 37-127.

It's $90 a gallon for resin and $38 for a quart of hardener.

Jack

i looked at this stuff but was unable to ascertain how thin the thin was? i am trying not to end up with a huge collection of small quantities of epoxy. have you see the thin stuff?

Quote:

---

Originally Posted by walter

What I really need is a valium. I'm ready to throw this thing out the window!!

EDIT:
(I just googled nitrous oxide. That could definitely work)

---

Hi Walter - Can't help you with the valium, but if you are near San Jose CA sometime, I have some Wild Turkey Burboun here.

I wonder if it is possible to recover any expensive parts from the EG. A common technique to break up something like this is to cryogenically freeze it (liquid N2) and then rap on it with a hammer - often will break up like glass.

It might be possible to do it with dry ice as well - maybe worth trying.

Jack,

The stuff from fiberglast.com is interesting but remember that the chances of making anything resembling epoxy granite are a bit low with a mixture that hardens in 20 minutes. That eliminates fiberglast 2000/2020 as a possibility. With 2000/2060 or 2000/2120, there would be enough time to make the composite and these look like nice epoxies although their viscosities are about a third higher than the epotuf 37-127/37-606 combination.

In general, I think these epoxies are cost prohibitive for most readers of this discussion as through uscomposites.com, 5 gallon quantities of the 37-127/37-606 combo approach $30 per gallon. It's easy to find epoxies that perform better than 37-127 but engineering is not all about performance; it's also about cost. Since we want to build heavy structures that damp vibrations, there isn't a particular need to use the strongest epoxy. The epotuf 37-127 is strong enough that we already fail in the granite, not the epoxy.

My calculations show that as long as flexural strength of the composite exceeds 2ksi and the modulus exceeds 2000ksi then it will be fine for basic machine parts which AustinT has already shown is quite doable. There are probably cases where the higher performance epoxies are beneficial and they should lead to stiffer parts and the ability to make reasonably sized beams longer than about 3 feet. I'd certainly want to do a calculation before using the expensive epoxies on a part as they might be more expensive than necessary. For parts like a minimill base, there is probably not much advantage as the loads are mainly compression with a bit of flexural.

My personal favorite supplier for higher tech epoxies is www.appliedpoleramics.com These guys make epoxies for composite ballistic armor.

At any rate, good find Jack. I've found a lot of epoxy suppliers but missed that one.

John,

Sorry for being flippant in my comment about effusion vs. diffusion. Since we were talking about deairing, that's why I prefaced my comment with the bursting bubbles nonsense. No disrespect was intended.

Regards all,

Cameron

Quote:

---

Originally Posted by ckelloug

John,
Sorry for being flippant in my comment about effusion vs. diffusion. Since we were talking about deairing, that's why I prefaced my comment with the bursting bubbles nonsense. No disrespect was intended.
Cameron

---

Cameron - no apology necessary, and no disrespect felt. :cheers:

I must admit, in retrospect I was thinking more about the air coming out from between the lumps of resin/aggregate, ie small "holes", whereas the more difficult problem is the air in the resin, where you'll have a sort of resin bubble foam.

I'm glad I'm still opting for spinning my molds, but a bit faster than - :wee:


Regards
John

John, You said that you were spinning your mold; what are you making that requires you to spin it?

Jack

Walter (et al),

Your tenacity inspires me. Please do not stop.

An X2 base product is certainly an appropriate target for your technology; the genesis if this thread begins with your acknowledgment of a similar item by another.

The unavoidable misalignment of metal rail mounts etc. should be planned on, no? I will enjoy learning how you machinists actually address this issue. The embedded mounting strips could be proud of the surface (and extra wide) so a secondary op can square them up and drill the holes? This would add a significant step, but relax the need to be really, really precise with alignment of embedded rails. Hmmm, would also require a larger mill be available. Just thinking out loud here.

These words are not costing me too much: I am one of thousands reading this thread who probably haven't materially contributed to any of your individual efforts. I might participate by buying something someday. Please keep it up.

Thanks
-Mark

Walter,
sorry it did not turn out the way you hoped, but your casting is certainly another step in the right direction. If nothing else it is a proof of concept or a precision planter. You could mount a few flower pots on the z-axis and it would look great next to your front door. :) If it makes you feel any better, it looks like you're one small step away from a perfect product.

I have been planning on a very similar casting, but am concerned about the ability to create a perpendicular structure with any accuracy. Why not cast the base and column as 2 pieces an bolt them together? A one-piece structure could certainly be cast if the mold were big/ridgid/expensive enough, but then it would be hard to vibrate and a real chore to build. Do you think that your existing mold is the limiting factor?

Also I was considering using a thin-walled aluminum extrusion as an embedded core (that would get filled with E/G) to which the steel ways would be fastened to keep them aligned during casting. Some have commented that this would create problems with resonance. Would this still be true if the mass of aluminum was relatively small?

Anyone have any thoughts?

Hi Jack.
Lots of pages ago, I decided to take the approach of building up a frame from smaller elements, rather than attempt a single casting.
To that end, I've been looking at designs for a simple cross section that will give me a beam that could be easily modified after casting for different gantry ideas.
My flavor of the moment is towards an air bearing surface formed from two parallel square beams, each rotated 45 degrees.
Regards
John


Edit I should have added that my end point is a 24" gantry router

Walter,

How did you attach the steel ways.

Where they attached to the mold or just braced in place?

Thanks everyone! I appreciate the comments and ideas.

Yugami,

See the post here http://www.cnczone.com/forums/showpo...postcount=2131

I couldn't afford large precision angle plate so I used what I had in stock. It was suppose to be a vibration damping test platform, that's all.

I made it in my spare time at the cost of $150 and was going to give it away in return for test data and report from user. I still want to do that, just need to find somebody with a mill-drill in his garage.

Cheers!

Quote:

---

Originally Posted by walter

Thanks everyone! I appreciate the comments and ideas.

Yugami,

See the post here http://www.cnczone.com/forums/showpo...postcount=2131

I couldn't afford large precision angle plate so I used what I had in stock. It was suppose to be a vibration damping test platform, that's all.

---

Walter,

Thanks, somehow I managed to miss that one.

Enco regularly has angle plates on sale. I think there last flier listed a 12" ish one for about $45. I'd keep an eye out for something like that. should reinforce that seem.

If the issue is flow to the bottom of the mold then why not start the pour there. What if you place multiple small tubes vertically into the form before adding the aggregate? Then inject the fluid into the bottom first and begin retracting the tubes as you fill. If this was done on a vibratory bench you might not even need to retract the tubes. When the fluid reaches the top of the mold then you know it is full. Even though you would need quite a bit of pressure to force the fluid into the mold, the mold its self would see very little.