Use of epoxy for levelling

[JAZZCNC]

I guess not needing epoxy casting is first prize and in hind sight, I'd be very surprised if the end result is more more accurate than using the filing and shimming process as described by Jazz.

It's not more accurate it's actually much less accurate if scraping/shimming is done correctly but it is much easier and less time consuming for for those not equiped or experienced.

Regards not using a bridge can I ask how you know there is no difference in height between one side and the other.? Ie both rails on same plane.?

[gavztheouch]

It's not more accurate it's actually much less accurate if scraping/shimming is done correctly but it is much easier and less time consuming for for those not equiped or experienced.

Regards not using a bridge can I ask how you know there is no difference in height between one side and the other.? Ie both rails on same plane.?

epoxy method is nice because it makes sure your rails are on the same plane. How do you achieve this with a straight edge and shimming?

[Boyan Silyavski]

I liked the epoxy the first time on the 1000x400 machine. i hated it on the 3000x1300 , untill i figured my mistakes and made it right. The way as i see it if i have done correctly this one, i will stop using epoxy in the future when i am building smaller machines that fit inside that one.



I still don't see how sb will shim properly 3000x1300 machine. I mean to make try to make the machine not relatively straight but absolutely straight, which are 2 very different things. Though for 95% of the people relatively straight is more than straight enough.

By relatively straight i mean long rails not perfectly on 1 plane, one rail as a guide and the other straightened via the carriage, and so on. After the table is surfaced it will not show unless you have 500mm Z axis and are doing 500mm high details. If you get what i mean.

I spend a lot of time aligning and mounting my rails and even with all in one plane, even with a helper. Man, trying to make all absolutely straight is a big waste of time. It goes so slow. And you have to triple check everything. If sth not right do all again. Holding carefully 15kg straight edge does not speed the process up, i just can not imagine what will be if the rails were not in one plane.

I know sb. could argue here but let me tell you something. Remember, i am speaking of trying to make an absolutely straight machine here. On a machine with 1350mm wide working area the rails are 1800mm separated. Using 2000mm square edge across is extremely difficult to square the second rail to the first. Why? Cause you move one side 2mm and the precision square still shows its square. You have to scratch the straight edge with precision square and judge by the friction if its square or not. There is no other way. So yeah, tell me now if at the same moment you have to check if the rail is not wavy, twisted and so on, and what happens if the rail itself is not straight from transport so you have to push and pull here and there while screwing. You need to grow some more hands. Now another thing. Even 1 tiny dust below the rail and it shows. So it slowed even further all.

So my suggestion is to speak of relative or absolute straightness when speaking of how it is done, cause its different. But as i said, my guess is that people most of the time are meaning in their heads absolute straightness and at most achieving in real life a relative one

Dean again i agree with you, in a way . Absolute straightness /or a try at it/ in a DIY CNC is needed only if you make aluminum molds, surface aluminum machine beds and rails and generally intend serious production of aluminum pieces. Man i am not preaching OTT. But if sb spend considerate amount on BOB, Steel, motors, rails and so, he should at least give a try for an absolute straightness. Hence the bloody epoxy.

[mitchejc]

not relatively straight but absolutely straight

Silyavski, concerning rails and things there is no such a thing as absolutely straight. If it appears to be absolutely straight you are just not measuring it fine enough. Just getting back at you for saying my fans suck

Apologies to the OP for the digression.

[Neale]

Thanks for the picture, Mitchejc - that's exactly what I was meaning. X1 and X2 may be "level" in the sense of horizontal to the limits you can achieve with self-levelling epoxy, but they do not have to be in the same plane, just parallel. It's the same error and corrected the same way as not having the feet of the gantry exactly co-planar. It's not ideal, but once set up, your friends will never know the difference...

I have to agree that epoxy is a second-best. If we could use self-hardening water, I might believe that the surface would be ideal, but even low-viscosity epoxy doesn't run that well, and I suspect that you also get surface imperfections because of internal stresses caused by shrinkage. If I had a surface plate 1.8x1m, I would consider scraping (proper engineers' scrapers and everything, I've used them before) the top of the rails to it. In my case, I was looking to take out a dip in the 100x50 box sections of something around 1.5mm. I don't know why it was that bad, whether the box section was warped when I bought it or if it is welding distortion or what, but it would need quite a lot of filler and/or filing to get that flat to acceptable limits, and it just ain't that easy to do. As pointed out, even if you could use a straight edge to get the top level along its length, measuring twist is really difficult even though the profile rail manufacturer's specs are pretty tight and you need to get it right (or at least consistent along the length of rail).

[JAZZCNC]

I still don't see how sb will shim properly 3000x1300 machine. I mean to make try to make the machine not relatively straight but absolutely straight, which are 2 very different things. Though for 95% of the people relatively straight is more than straight enough.

Your fooling your self if you think you can build and more importantly measure to Absolute levels. Every DIY build is a compromise between accurecy and engineering abilty.

Jazz, the way I see it the X rails don't have to be on the same horizontal plane as long as the two planes are parallel ie. if rail X1 is 2mm higher than X2 it must be 2mm higher everywhere along the length. There's probably other ways to do it but I epoxy casted a T shape reference surface on my base and used that to set up my Y and Z. The machine in the extreme example below will cut perfectly well as long as the y rails are parallel to the ref surface who's horizontal plane is parallel to the two x-planes.

Agree completely but again your only has accurate as your reference surface and abilty to measure. Not every one as large granite surface table for reference surface to work or cast from and without one then how sure are you your reference surface is accurate.?

If I had a surface plate 1.8x1m, I would consider scraping (proper engineers' scrapers and everything, I've used them before) the top of the rails to it. In my case, I was looking to take out a dip in the 100x50 box sections of something around 1.5mm. I don't know why it was that bad, whether the box section was warped when I bought it or if it is welding distortion or what, but it would need quite a lot of filler and/or filing to get that flat to acceptable limits, and it just ain't that easy to do. As pointed out, even if you could use a straight edge to get the top level along its length, measuring twist is really difficult even though the profile rail manufacturer's specs are pretty tight and you need to get it right (or at least consistent along the length of rail).

It's all about having the right tools whether it's hard or not. With Precision straight edges and Engineers levels it aint that difficult. But here lies the problem becasue at DIY level 99% don't have or need these tools and the epoxy method with bridge makes it possible for less tooled up DIYers to achieve acceptable levels of accurecy for gantry router style machines.

The key word here being ROUTER. . Remember these are mostly Gantry based router machines being spoke about and the nature of the beast dictates lower accurecy is acceptable and high precision accurecy unobtainable.!! . . . .Again Horses for courses and use which ever method gets the job done for you in the easiest way possible.!

[Rich]

Have you considered whether the fault lies with poorly mixed resin? The cure is a polymerisation. If the relevant components aren't available in sufficient quantities, the density of the film may change, leading to variation in final thickness?

Epoxy resin tends to be quite sensitive and there's little mention of preparation in this thread.

Using a good set of scales will ensure the right amount of resin and hardener is used. Really important if you want to resin to perform per manufacturers specifications e.g. hardness

Mixing. Spend at least five minutes mixing before use. Hand mixing is fine, an electric mixer is easier, but a slow speed. Avoid mixing in air. Once you have mixed for a minute, use a pop stick to ensure the resin sitting on the side and in the crease of the cup is integrated properly. Mix for the remaining four minutes. Once the batch is mixed, poor into a separate cup - leaving a little behind, so as to avoid any un-mixed resin transferring across.

The surface to be coated needs to be clean and dry.

The video was helpful in some respects. Personally, I wouldn't put a heat gun near epoxy without wearing a mask - air bubbles or no. But he's right epoxy loves heat and it does affect the viscosity of the poured resin. Personally, I'd lift the temperature of the resin prior to mixing using a water bath, especially here in the UK. I'd also be more interested in ensuring the ambient temperature of the room remained constant - especially in a hot climate.

I understand the target thickness for the film is 3-5mm. There's no reason why this need to be done is one pour: http://www.westsystem.com/ss/epoxy-chemistry/ A second layer can be applied once the first layer has started to gel, sufficiently. This might well be the best approach, assuming the underlying cause is human error

I'm not entirely certain what steps are taken after the film has been applied, but if the variation remains say 0.3mm i.e. human error is not a factor, could a high viscosity resin/compound then not be used to seat the rail? I believe that what is taking place here? http://www.moglice.com/ . You'd need to wait around a week before seating the rail. Ambient cure epoxies take a while to reach a state of final cure, which could be a problem, as you don't want an ultra thin layer of epoxy being exposed to pressure, when not chemically bonded to the layer below .... all my epoxies are heat cure, so I'm not in a position to run any tests. So lets hope its human error and applying two layers to create the desired thickness produces a lower error rate.

Oh yes. Vinegar is a good solvent for any split epoxy.

Personally, for this job, I'd been thinking about urethanes - but I'm not any way near needing to do this yet.

[mitchejc]

Regards not using a bridge can I ask how you know there is no difference in height between one side and the other.? Ie both rails on same plane.?

Jazz, the way I see it the X rails don't have to be on the same horizontal plane as long as the two planes are parallel ie. if rail X1 is 2mm higher than X2 it must be 2mm higher everywhere along the length. There's probably other ways to do it but I epoxy casted a T shape reference surface on my base and used that to set up my Y and Z. The machine in the extreme example below will cut perfectly well as long as the y rails are parallel to the ref surface who's horizontal plane is parallel to the two x-planes. Mine, does not look like this as the x height difference was minimal:-) I used little skrew adjusters and then epoxy putty to shim my gantry bottom parallel to the ref surface and gantry front perpendicular to the ref surface and also used the ref surface to set up my Y rails and Z. Casting the ref surface does add a bit of extra work but once its there its very easy to set up Y and Z as opposed to some other methods I've seen.