Build size and Tube/EMT materials

Okay im waffling on which way to go. First CNC foray. so CNC, Hardwoods (Inlays and smallish 3D parts), Acrylic/MDF/Ply for templates and jigs and such. Carbon fiber is probably not smart to cut. I doubt I would need metal so ill consider that a bonus. Maybe some drilling guides in metal would be nice. At this point, I think I am leaning towards a recommended build size (17.75x13) to try to see the limits in tolerance etc. Or perhaps something slighltly bigger say (24-20)x18 might be able to more things. With the recommended size I would probably do most ukulele build tasks. I am sure I would have size envy at some point. I think I am more interested in the precision for advanced inlays then the max size. I can always build more or upgrade later. So I think my most important need desire at this time is inlays with high precision.

Some found comments:

“Your recommendation for the expensive tubing is based off a large build with a very tall Z axis. That Z of yours is compounding any flex several times. A normal-sized 2x2 build with the minimum Z screams through wood. If you look at my speed tests vs yours, since it is the same tubing, you will see how much faster I can cut than you just with a normal sized build. You needed the DOM mainly because of your choice to do an extreme Z.”

“Using the default values, or very close to them, in the calculator are highly recommended.”
“2’x2′ footprint MPCNC is a really fun and easy to use machine, anything larger and your CAM will have to be very accurate to work well.” - Introduction to The MPCNC - V1 Engineering Documentation

“My MPCNC is 24"x 24”
I found that I have approx. a +2mm rise from the edges to the center of the carve area."

Questions:

  1. if EMT works great up to 24x24 why is the recommended build size so much smaller (17.75x13 odd size BTW why not 17x17)? Guessing the deflection. Hmm, should I even go smaller for tight precise inlays?

  2. If I decide to upgrade later I would likely need to switch to SS/DOM and the larger parts. Is that ALL that is needed to change from the original kit or would I need a whole new kit? (Longer bolts etc).

  3. If the EMT works so well at the recommended size, is there a good reason to spend the extra on SS/DOM now for a standard recommended build size? (Besides the upgrade path).

Any other words of wisdom appreciated.
Rick

  1. Yes, small as possible but still do the job is ideal. Larger builds flex more, period… laws of physics.

  2. You’d also need longer z screw maybe, longer belts, perhaps longer wiring harness… not sure if I missed anything here, but the last one can be a biggie.

  3. DOM is a lot more precisely manufactured, besides flexing less due to thickness. EMT is not intended to be very uniform, has a welded seam, galvy coating, etc… it’s not a precision product like DOM.

3 Likes

The Z length affects rigidity a lot. If the Z reaches twice as far, that particular flex gets quadrupled. You can easily carve into 4" stock, but build a table where the work is lower, or prop up the stock so the MPCNC can reach it without sticking out too far.

The gantry does sag a bit. The big impact for you is that V bits will be wider or skinnier with errors in Z. There are several ways to reduce that impact:

  • You can surface your spoil board with the CNC. This makes the surface of the spoil board match any sag (we are talking about a few mms over a large area, not visible to the naked eye). The weight of the router is a lot. The loads from a small vbit engravignng are near zero, so the sag should be consistent.
  • You can use a touch plate, or otherwise set the Z=0 to the top of your workpiece. The error near where you zeroed it out will be very small.
  • You can build your XY dimensions smaller, or use larger tubing. That will male the sag from the gantry tubing smaller (half size is close to 4x more rigid).

I measured by Z error by setting the Z=0 to the top of my spoil board and I “drilled” a pattern of 4mm deep holes in a square pattern. Then I measured the depth of each hole using my calipers. I found the deepest to the shallowest was about 4mm, but the average was closer to 2mm. That was on a 36"x48" LRv1. 4mm seems like a lot, but I could easily have cut a 10" long pattern without 0.5mm of error. These numbers are from my memory because it was a few years ago with the older design.

I’ve also seen you mention about saving time. CNC is not automatic. Every time you design a job, it takes real time. Especially as you are learning. The CNC doesn’t move much faster than a human. The way it will save you time in production is 1) it frees you to do something else (in the same room) while it is working. 2) If you use the same design to make many parts, the second part is way faster (because you don’t have to do more CAD/CAM).

IMO, you should build a smaller size first (especially Z, if you are doing engraving). You will learn a ton just doing a few projects. You will quickly be able to intuit what way it will work and which ways it won’t.

I don’t know your budget. The larger tubing (25.4mm OD vs. EMT at 23.5mm) is valuable at any size. But it is definitely not a requirement. You don’t have to buy stainless. Regular DOM steel is fine. I built my first mpcnc with EMT (which was 24"x36" and worked great). Then a LR1 (36"x48") with DOM. My LR3 (30"x48") is back to EMT.

You can always change the machine later. Building a larger Z just means replacing the tubing on the Z and the legs. Changing from 23.5 to 25.4 tubing means all new printed parts (but all the electronics are identical). Making a larger XY is 6 more tubes. Making XY smaller is just a couple of cuts.

Learning CAM is challenging. A smaller machine means you can get away with more mistakes while you are learning. You can do it though.

3 Likes

I’ll preface these remarks by explaining that I’m using 3/4" conduit, and am happy with its performance. My current working area is about 16"x21". I may expand it to 21"x31" some day (big enough for a full sheet of dollar tree foam core), but that will require freeing up space in my shop. With that option in mind, I made all my motor and end stop connections long enough for the larger working area, and my “long” rails now can become my “short” rails if/when I do complete the expansion.

I would use larger tubing from the start if you reasonably expect to expand the machine beyond the size that smaller tubing is expected to do well with. Basically, make sure the plastic parts you start with support the tubing size you want to end up with. Then, when you increase the size of your machine you’re just replacing a couple of rails.

Thanks. Can you somehow quantify the compromises for this size EMT? IDK perhaps deflection and accuracy? You are happy with it but are you having to “work around” the limitations?

My initial goal for the machine was to cut our free plans for “foamie” aircraft designed to fit on 20"x30" sheets but shortened one set of rails as it would then fit in a convenient spot on a counter in my shop. It lives in a basement shop, and to get to it I need to carry things down a staircase with a 90* bend and through a finished family room living area, so I don’t work on full sheets in the shop, I break them down upstairs before bringing them to the shop.

My primary limitation is where the machine can live in my shop. The sides are open, so I could slide longer work pieces through, but I haven’t actually done any large pieces. I don’t have significant rail sag, and work in wood and a little in foam.

I think you are getting hung up on the wrong things. For the right reasons.

You are very worried about accuracy, but you can get almost any accuracy you want out of pretty much any machine. The trade-off is speed. When we discuss these things, we are lumping them together. So a large machine with EMT makes the same exact cuts as a large machine with DOM, just slightly slower (lower material removal rate).

Inlays are very very shallow cuts in very nice wood. Just about as light duty and ideal as it gets. Finishing passes bring our tolerances very tight in dense wood.

Why we keep stressing smaller builds is two fold, smaller is more rigid, faster, more forgiving to program to get the accuracy most expect at good speeds, and the next thing is do you plan on doing 30"x30" inlays? So many people have build 4’x8’ Low riders but I honestly think there might have been 2-3 projects that used anywhere near that in all these years.

You decide the largest thing you will cut and we will help you make the right decision. If I thought DOM was worth it I would let you know I promise. I want to look as good in your eyes as possible when you build my machine. The MPCNC lets you slide in larger material than you can cut so there is no reason to fit all your material in the usable area. Make the usable area as small as possible. If ~$80 is not a huge deal to you feel free to buy DOM and don’t give it a second thought either way (if you have a metal shop nearby sometimes you can get it for half that).

I think you have noticed I step in anytime someone says something is necessary, If it was that is what the instructions would say. You know the real reason I provide 25.4mm capable parts…the is the OD of EMT in parts of Australia and a few other regions of the world.

2 Likes

Thanks, that helps a lot actually. I have decided on the recommended size build. Should suit my needs and if not I can rebuild. I have just been agonizing over if EMT and low cost (~25) and easy to source is good enough without an upgrade path or if I want 1" (some kind of steel) and upgrade path. It’s a bit more work to source and I’m trying to figure that out now.

I can get some at Orange in 3ft (mild? steel) but its $$ instead of $. I’ve called all the local places (too small job) and finally found 1hr away found a place willing to give a quote on SS (either they don’t think they knew what DOM is or I dont know what to ask) but I had a rough time with the accent over the phone, I doubt that quote will ever come.

Found some online DOM ~.08 instead of the .12 which is nice but they are 8ft and that raises shipping to $200’ish (almost double price). I can get 6 @ 4ft mild steel for like $100 delivered. Which might be good enough to move forward. So, I a trying to find work out my options.

Something you may consider adding, how may pieces of pipe needed depending on size. For the recommended size I came up with: 3@8ft (really 2@8ft + 1@1ft would work), 6@4ft, 7@3ft or 5@3 ft + 2 @ 2ft - IF I did that properly.

3 Likes

And flex it will… at least with mine. I can’t tell you what to build with or what size to make your machine, but I can tell you the limitations experienced with mine.

I love my mpcnc. You can see a bunch of things that it has made in the gallery.

With that said, I built my machine with a 2’x3’ working area. 3/4" emt was used for rails. Don’t use 3/4" emt if you go that big. I’m sure if 1" DOM was used there wouldn’t be anywhere near the flex mid span. I’ll be shrinking it down to 24x24 soon. Realistically, the bulk of what I do will fit in the smaller footprint.

You can still produce decent quality, but I find multiple passes with small doc will do it, but the time factor increases. Time is a luxury I don’t have much of lately.

Just my 2 cents, but you can keep the change! :joy:

1 Like

Thanks Matt, I decided to go with the recommended build size (~ 18x13x3.25? inch) 1" OD for the added rigidity and upgrade path options. I can switch to DOM/stainless or make it bigger this way to avoid having to buy new kit parts gong from EMT to 1". I went low cost and got mild steel alloy (HREW) for now. I almost went milled stainless (304) which seems way cheaper then the other stainless. Not quote sure what all these mean and getting tired of reading material properties. I’ll maybe put some penetrating oil or paste wax on the HREW for rust avoidance. Hopefully this stuff is smooth enough? I’ve seen some evidence of people using it but not much on performance.

I got the pieces pre-cut at MetalsDepot online which claim only 1/16" tolerance (better then others) for not much extra. We shall see if I have to fine tune them. BTW, long tubes seem to greatly increased shipping so watch that folks.

Now I am on to boards and leaning a bit towards the Rambo and figuring out what else I need to include. I just fond out a friend is going to print the parts so I can check that off the list. Getting closer.

1 Like

Really wish I had thought that out before the build, but even with the limitations of long 3/4" emt, I am very happy with my machine.

1 Like