This guy tested a few. The video is long and rambling and definitely not PG rated (don’t watch with kids around), but images speak for themselves (ACF is one of the two shiny sectors).
My main concern would be if it will be sticky enough to collect sawdust.
That’s why I said paste wax. It’s try to the touch in a few minutes.
Saline feels more realistic…
I’ve encountered ACF-50 (and a similar product called CorrosionX) in the aviation industry. Both remain tacky for a long time. I think they’d be a saw dust magnet.
I agree with @barry99705, that paste wax is a good option. There’s also a variety of products sold for use on cast iron table saws and such that would work OK. I think they’d all need occasional re-application.
Conduit is still cheaper than DOM tubing. 5ft of 3/4 conduit is about $5 so a 2x2’ machine would use about $15 worth of conduit (3x5ft), the same machine made out of 1" tubing would come to ~$50 for tubing. Not sure if that $35 alone is worth keeping the conduit version around, but given conduit is galvanized, it could be slightly less maintenance.
Given for longer spans one really wants a larger tube, I would look into a possibility to unify metric and imperial with a 1.5" version. 1.5" is exactly 38.1mm which should be close enough to use the same parts. I did some quick googling and found a bunch of tables from Sizes of steel tubes and pipes - Železiarne Podbrezová . 38 or 38.1mm is listed in all US, EU, JP and RU standards tables. I also cross-checked what tubing is available in Estonia where I’m originally from. We have a very low population (1.5M) so anything remotely rare would be hard to get. I found both stainless and carbon steel to be available in 38mm. Of-course more research is needed to fully gauge general availability of that size.
Another size that has significant overlap between the standards is 31.8mm or 1.25". I did not see this readily available in Estonia but 32mm is available which can be forced into the parts but I don’t know what the ramifications would be. @vicious1 what is the tolerance of the parts that hug the pipe? I know the documentation says “interference fit” (0.05mm @ 25mm) but what happens if the pipe is larger/smaller than that?
EDIT: 1.25" EMT is 38.4mm OD so in theory if it’s possible to design for 38.2 ±0.3mm it should work for conduit, imperial and metric tubing.
EDIT2:
I ran some numbers to see how much weight advantage would a 3/4 conduit version have over a “generic” 1.5" version which would be one reason to keep this version around. 3/4 EMT has wall thickness of 0.049". Everything in the 1.5" size seems to start from 0.065". So the gantry of a 2x2’ build would “gain” about 990g (850g → 1840g). However, the middle itself w/o rails on my MPCNC (with all the bolts and bearings and Z axis) weighs 2308g so the weight gain would be about 30% w/o a tool. For CNC use, DW660 adds another 1500g so the total gain is only 18%. 1.25" is available in 0.049" wall thickness thus the weight gain there is ~310g or only 9% of the total gantry weight of a 2x2’ machine. I personally wouldn’t worry much about the weight gain but then again I don’t 3D print with my machine (yet).
Yeah, I figured, 
. Really dislike this 3 version hole I dug myself into…This version might currently already fit 25.4-25, maybe 1-2 different parts…23.5 all new. Maybe, I guess I should really look into this and see how far off it will be. Hmmmm.
Pastewax it is. So in the end, rust is not an issue, with some light occasional use the bearings will ride on a clean path, this is just for some aesthetics I think. Anything will probably work just fine, If you have an issue using some wax on the rails once a year, spring for the stainless. I am just trying to offset some of the cost, meet in the middle.
That is what I am working on. Currently larger or smaller really really effects the angle of the gantry tubes, compounding a few times, more than just tolerance stacking. The new one, might have much much less of an effect. It does still rely on the Z axis to help set the angle but in a very different way. Not compounding means the effects for a different size are multiples less error. With this in mind I will put a little more thought into it.
At some point it just can’t be that variable. For example think of the corners, We could pretty much use anything smaller than the intended design and it would work pretty good. At some point the offset block would be pretty far off but still functional, then eventually you would run out of the clamping gap. The feet could be made in two pieces and that could accommodate a larger range. BUT everything is a bit stiffer if they are a more exact fit. So 25-25.4…really no big deal, feet-corners-rollers-gantry…then you hit the Z axis. That needs to be a precision fit. That in turn defines a lot of the center geometry. The more size flexibility you build in, in the end, you end up loosing space to mount a tool.
I started this new one with the intention of working with all three sizes, the 120 bearing offset will just barely fit all three, but then I think I would sacrifice rigidity just for me not wanting to maintaining three sets of parts, so we are back to 3 sets.
Hmmmm, I have some ready for testing, back burner.
So now really before I would even consider going bigger I would have to see really proof the rails impart any significant amount of error vs plastic parts in a reasonable sized build. I really tested the crap out of my 24"x24" footprint build and with a huge load and the frame deflection was not measurable with my dial indicator. Remember it is that first very tiny load we fight, the initial endmill engagement with the material, not the large constant load. One it gets a bite and loads up everything is fine as the system gets to the end of the elastic zone.
LR…yes that could use larger rails for sure (or a different cross rails system), but the MPCNC…I really don’t thinks so other than allowing for larger builds as this would also mean larger plastic printed parts which in turn means more springy elastic zone. So we would have to use other materials in the build, no longer an MPcnc (pro version?).
P.S. This thread is really helping. Some of the details I did not care to define or “solve” are really getting worked out. Thank you everyone!
I agree that for 2x2’ machine 25mm-ish rails are a good fit and there is really no need to go any larger. However, even when there is LR which can be built to be of a full-sheet size, the sweet spot for me would of been a 35x35" (2x2’ work area) machine. Why? Because I could then fit a 2x2’ project board in there 
I enjoy building a lowrider, but I’m not sure if I would of started it if I had the guts to build my MPCNC larger in the first place (my current work area is 16x14x3").
So yeah, maybe 1.25" would be a good fit to “extend” MPCNC half way towards the lowrider but there is no cheap EMT option there. Or perhaps keep 23.5 as the smallest/cheapest option and unify the metric and imperial to 31.8-32mm range for anyone who wants larger span or smoother DOM rails.
Perhaps a survey could also give some insight to what people in this forum think.
A survey to figure out what is really commonly available (as EMT) would be good! I would only do larger if it was EMT as shipping long pieces of metal is expensive. We started with 23.5, then added 25, then Australia popped up with 25.4 as the common EMT size…nuts! Survey where you have to provide a link to your countries standards…This should be available by google search but only in the native language.
Maybe just a post with a few specific questions. An actual survey would lose a lot of the detail, and we would have a hard time making jokes in the middle of it.
Is 38mm tubing common? I would have expected 35mm or 40mm instead. But I live farther than anyone from another country or ocean.
I think standards are fairly easy to come by. The link I posted above has tables of tubing standards for EU/JP/RU and US. US Conduit standards are easy to find as well. I don’t know a source for UK/AU and maybe IN standards, but given these are English speaking countries, I’m sure I can find these after a bit of research. I hope @forcerouge can help find standards for CN.
Once we have these all, we should look where the sweet spots are. E.g. sizes that cluster close together e.g. within 0.5mm of each other. However availability is another story and this is where I think a survey can help. E.g. once we’ve identified a few sizes that are close, asking people which size they can source the easiest.
I don’t think 38mm is easy to find here in the Netherlands. 25mm, 30, 35, 40 and so on is easy to find, affordable, can be delivered cut to size for just a few extra bucks.
I’m sure it would be possible to find 38mm tube but it would probably only be for large customers and 6m long with ridiculous delivery costs (for a single tube).
Looking at big box store pages in northern EU countries (Estonia, Finland), I find 25mm and 32mm being the most commonly available sizes.
Since most things are produced in China, you can pretty much find any size here. I made a quick search for 38mm and 35 mm tubes and got litterally thousands of different results for both, so absolutely no problem.
What ever happened with the 1" DOM testing? I think I found a source that’s local(no shipping or waiting to start printing). Otherwise i’ll need to go with one of the EMT’s. Will either still work with the MPCNC 2020 version?
The deeper I get into these forums the more I find. This rabbit hole just keeps getting deeper and you guys are trying to cut my guide rope.
Thanks!
That is what I am building the new one with.
I’ll give them a call at lunch. Hopefully I can pick it up on the way home. Thanks Ryan
I picked up the tubing last night, but the guy was out so sold me stainless instead. I think it was still cheaper than the normal EMT tube.
Awesome!!!
Did you double check the OD, Some stainless is 25.6mm+ not much of a difference and should work but still not ideal.
As luck(mine anyway) would have it the battery in my calipers was dead, but it looked like it was between 25.4 & 25.5.(using the scale on the caliper) I plan on printing a foot tonight for a test fit.