[pippin88]

I have personal experience of epoxy creeping when used under linear rails / ball screw mounts. I don't have a measurement, but there was a visual impression / outline in the epoxy.

I would not use epoxy under rails for a metal mill.

[jaguar36]

Hi All - All epoxies bond to clean steel very well. I'm unclear why people would place a 6mm piece of plastic under their rails if a better way was available.

Because a better way would be much more expensive. Yes, ideally we would all be mounting our rails on nice machined, stress relieved and ground cast iron frames. But each one of those operations on a significantly sized (say 4') frame is going to be extremely expensive. If instead you can use an epoxy a couple bucks and sacrifice a tenth or two of accuracy that's a great trade off. The creep rates of epoxies designed for mounting linear rails at room temperatures is very low.

PS. Glad to see you're using a Nastran based solver now rather than Simsolid.

[peteeng]

Hi Jag - Is creep data available for epoxies designed for linear rail leveling? Sorry to be pedantic but usually to prevent creep, fillers have to be added usually long fibres. Since this can't be added to self levelling epoxy and aggressive hardeners can't be used to increase the curing hardness then I'm unclear how this aspect of curing can be improved in epoxies. I use epoxies for all sorts of things and I have been involved in creep testing of laminates so this is an interest for me... Peter

[joeavaerage]

Hi,

If instead you can use an epoxy a couple bucks and sacrifice a tenth or two of accuracy that's a great trade off.

It is a compromise. If you determine that the compromise is favourable to you then go right ahead. At the design stage, which is where this thread is, then you would try to design such that no such compromise is required.
For example make the machine components in such a way that they can be machined if not cheaply but at least economically, as opposed to some tortuous and extremely expensive set up with outsized parts.

I have seen people do the same thing when designing a cast part. They seem to imagine that any part they can conceive can be cast, only to find that the moulding costs have gone through the roof because
the part requires multipart patterns with multipart sand cores. Had they spent some thought on the moulding process they may have designed a part which requires a single part pattern in a simple drag and cope
with greensand alone and no pesky sand cores and save themselves a fortune.

Designing hobby CNC is as much about designing parts and assemblies which can be economically made with the least possible compromises.

Craig

[jaguar36]

Hi,



It is a compromise. If you determine that the compromise is favourable to you then go right ahead. At the design stage, which is where this thread is, then you would try to design such that no such compromise is required.
For example make the machine components in such a way that they can be machined if not cheaply but at least economically, as opposed to some tortuous and extremely expensive set up with outsized parts.

Sure, everything about engineering is about compromises. Anysort of machining/grinding that hits the same tolerances as an epoxy bed is going to be an order of magnitude more expensive. (Unless you can do it yourself or have a friend of course). Thats a good chunk of money that can be spent on other components or more material for a stiffer frame where you will get a much higher bang for your buck. Now if you’re happy with the stiffenes and tolerances of an extruded shape to mount your rails too thats great too.

[joeavaerage]

Hi,

Anysort of machining/grinding that hits the same tolerances as an epoxy bed is going to be an order of magnitude more expensive.

That is correct. I would not tolerate epoxy mounting so I had my cast iron axis beds, cast obviously, and then precision machined, Cost $10000NZD, of which $6000NZD was machining.
Had my design been more complex where I could not machine the beds individually the machining cost would nave been even higher. I had to design such that the machining could be done
on a moderate size machine with as few setups as I could or face a $12000NZD machining bill.

Craig

[peteeng]

Hi All - I have a piece of short rail here and some suitable epoxy so its about time I did this in person vs make some models. So I'll organise a pour asap. Attached is the epoxy I shall use and it cures to 85 shore D in 14 days. I don't have a hardness tester but its crisp and snaps like toffee after 24 hours. If someone has a data sheet for an epoxy levelling system to compare it with please publish it here... Peter

[Chris-M]

Hi Peter,

Would Diamant DWH be a good candidate? https://www.diamant-polymer.de/en/products/dwh/
It looks like it comes in a few different variants.

Chris

[peteeng]

Hi Chris et al - Diamant and Moglice come up quite a bit in the forum for the european builders. I looked at their site and they don't have a leveller. . I've used epoxy over decades for "setting" joints and doing liquid shimming. Mainly correcting fits and filling gaps in machinery and yacht structures. Its perfect for that application as its strong and has very low shrinkage. Peter

I ordered a hardness tester

Diamant have a self levelling system. Its viscosity is 500mPas the ATL resin is 240mPas so it flows even better...

[peteeng]

Hi All - My shore hardness meter arrived so I set up some epoxy for a test. The R118 has a max hardness of 84 HD at 2 weeks. In 30 hrs we are at 80.5 HD so I'm happy with that. I'll track it over the next couple of weeks. 80HD is the same hardness for the white melamine surfaces I have around the shop. The next thing to do is to make a steel bed and set up a levelling experiment. Peter