New build in Toronto needs advice

I came across the MPCNC in March and thought wouldn’t it be nice to have a do-(mostly-)anything machine from hardware store stuff. So I got started building. First, thank you for this piece of art! Been lurking ever since and now I need the community’s wisdom.

(I also thought about the 2beigh3 but cost of threaded iron pipes totally put me off. )

I had no 3D printer to begin with so I bought a Wanhao i3 - $500 - which just about ate the price advantage - although it’s a fine, very hackable machine itself.

I sized it for 24"x18" after this guy:

Thinking I could just buy two 10’ EMTs and have my rails. Alas I botched one of the cuts and need a third. Plus it ruined my dingy tubing cutter from my plumbers kit, meant for copper pipes.

Anyway, I printed the parts and have the conduits, still waiting for the bearings and electronics from China, and actually got tired of waiting so I bought a RAMPS kit from a local shop. Still can’t really put it together until the bearings arrive.

Now it’s time to think about table, electronics box, spindle. And I need advice.

My goals out of this build:
Large format 3D printer (more on that later)
Light aluminum milling
Computer controlled mitre saw (no compound mitre though, lol)

Stretch goals: Laser, drag knife.

I’d love to have as much parts retail-off-the-shelf (that I can walk into a store, buy and take home) and reusable as possible. By that I mean it will share as much parts as possible with Wanhao and when it comes time to dismantle either, I can reuse the parts. That means I’ll move the Wanhao’s MK to MPCNC for 3D printing needs - or should I go the other way around, get an upgraded 3mm printing kit for MPCNC and move it to Wanhao when needed?

For hookups I can do either DB9 with plain old 24-gauge wires or RJ45 keystone jacks with Cat5e patch cords. This gives me provisions for 8 wires which should be enough for one axis.

Power is going to be from an ATX supply because I have so many left over.

RAMPS box: I am considering these two: with this modified panel

First one I like the size and look and feel; second I like the ergonomics and 80mm fan, which I also have too many leftover.

I’d like to use one box for both MPCNC and Wanhao but I don’t think Marlin can switch hardware profiles on the fly… at least the hardware is or can be made common.

Firmware: Ryan’s. I haven’t bought a screen yet, but I was able to control the motors on the bench via RepetierHost. Does ours have all the latest upstream updates?

Have a 20yo Dremel variable speed rotary and printed a Hicwic-compatible mount for it. Needed to modify it to fit my older Dremel.
Got a Dewalt 660 for $100 on sale but also trying a $150 Makita RT0701C because of variable speed and possibility to use it on my router table (needs modding though). Dewalt 611 costs too much.
Both have 1/4" collet. Dewalt comes with a 1/8" collet. I would need to get one for Makita, if one even exists.
Tell me which one to keep. I’ll return the other.

This is the big one. I could just buy a piece of MDF and mount the build on it, leaving some space to the right for electronics.

But I like to have a motorized drop table with at least 180mm travel (Wanhao’s usable Z). My plan for this is to home and level Z with table fully up, power down, reconnect Z to 2-4 steppers with allthreads wired in series mounted to the table, so the table moves down as it prints. In CNC mode I would latch the table at the top. Optionally I can build a short, unpowered Z axis that just sits on the center, bolted to it by a short length of allthread, for use when the Z axis is rewired to this drop table.

There are two table designs that caught me:

I’m going to put wheels underneath. Tell me why I should pick one over the other or just go MDF board. For background: 3/4" EMT are $9/10ft, 2x4s are $3.5, 4x4s I haven’t got a price but they look to be all pressure-treated for outdoors.

How does my dream sound?

Dui’s table (and entire build, really) is great. If you understand what he’s doing, and can replicate the quality, then you’ll be very happy.

I have an original duplicator i3, and it works fine, but the print speed is mostly limited by the underpowered extruder, IMO. So if you have to move the print head from one to the other, I’d prefer to not use the wanhao one. After I had the MPCNC, I made an MP3DP, and it works great. You can buy the extruder from vicious1, it’s at a very good price/value.

What do you mean by “Computer controlled mitre saw (no compound mitre though, lol)”? The thing I call my mitre saw (or miter saw hehe) is used to crosscut boards at different angles. The MPCNC isn’t going to be the most effective at doing that (but it can be done, but you’ll spend more time setting each cut up than actually cutting).

Also, in case you didn’t know, all these parts are in the vicious1 shop, which pays for Ryan’s time as the designer, support, advocate, referee, etc, and it pays for the forums. I think it’s fair to mention that if even if you don’t buy anything from the shop, it would be fair to throw him some bones via the donate button.

Welcome! I’m excited to see what you build. Seems like you’ve got quite a few interesting projects planned.

If you mean the printed parts, it will be a while before my build is operational, and I can print something else in the meantime with the Wanhao.

It won’t really be a “computer controlled”. I’ll just key a move command into RepetierHost to have it do a cross cut for me, based on the angle I need. Think using the 660 and Dremel as a spiral saw, but MPCNC moves the tool instead of my hand. You have a point though. At $160 maybe I should just go get a 7-1/4" sliding compound mitre saw anyway.

Okay, I was going to quote that and and give my two cents but I think I can sum it up. Before you go super crazy with it, build it small first. you have some lofty goals having not used a MPCNC yet. Large format printer, and Aluminum milling do not mix well. Not that it can’t be done but better off making 2 machines.

I don’t want to dampen the enthusiasm, I love it, but a small basic build will really really teach you a ton of stuff. Then go crazy with it. It does not have to cost you any extra to do it that way, cut your tubes to the length you want, make the legs really short and shrink the corners in to make a 10x10 workspace and use it a few times before you change everything.

There are tons of ways to cut corners but be careful cutting too many at one time. Dremel, isn’t going to be much fun for long. Not all ATX PS will work I had two that would brown out, I had some suggestions on how to fix it but a $8 PS works perfectly.

All the parts I had on my generic I3’s all died and like Heffe said the extruder was the worst part.

So you do literally mean mitre saw. I thought it was some miscommunication.

It can easily cut a straight line, but there are some gotchas. First, the machine’s origin is slightly different each time you release the motors (or power cycle the machine). Also, you will want the board parallel to one axis, so you’ll want to have some kind of rail. Not a huge deal, the rail can be smoothed by the CNC to make it parallel, but you’ve got to register it somehow, like with dowels or screws. Then, to cut something like a 90 degree cut, 18mm deep, 6" long is a different gcode file than 45 degrees or 1/2" or 3.5" long. Plus, the actual cut will be maybe 5-15 minute long. So you can see why a circular saw and a board, or a miter saw could be better.

That being said, there’s an argument that having one fewer machine. If you want to do anything more complicated, like a mortise, or a half lap, or half miter but end with a straight cut, then it’s becomes a more fair fight. Even cutting out entire rectangles might be more accurate with the MPCNC, but not as fast as a table saw. I did see a video of a cabinet maker that preferred CNC, just because he didn’t have to cut through the board all the way, so he would save material in the long run.

So, back to one of my more basic questions and still keeping with my “one less machine” motto, am I better off keeping the 660 or Makita? I’ll just print the mount for one of them.

I don’t rout that often - That old router was given to me.
On the other hand 660 is noisy!

Hello buurin,

Well, that’s a pretty clear plan you have I think, and most of it seems to be doable in my opinion.

I’ll talk more specifically about the table, since other people can provide more expertise than me in the other fields:

The table frame is fairly easy to build, costs close to nothing and has the advantage of possible motorization. But: you need to be very precise when you cut it, the final quality of your machine will entirely depend on it.

The most critical thing is that you need to drill perfectly perpendicular holes to pass the vertical tubes through at each corner. The tubes should fit snugly in them, otherwise the whole structure will end up being pointless, so they absolutely need to be vertical and at equal distance from one to another. I didn’t manage to make this perfect, I have a few millimeters of difference, so my machine is not totally square. This isn’t a big deal for me since I don’t need a crazy precision, but you really want to make sure you have the tools to do this part as precise as possible, so:

-A jointer, to make sure that the wood posts are perfectly flat and square
-A very good drill press, to be sure that your drilling is done at 90 degree
-An appropriate 25mm drill bit
-A good and accurate miter saw, to cut all the parts as precisely as you can.
-Patience to make perfect measurements

I didn’t have any decent tools to do that, It doubled the workload and made things unperfect.

My first plan was to build this frame using aluminum extrusions (9090 profiles). But since it was a bit expensive for me, and since I wasn’t sure that the MPCNC would actually perform well, I didn’t do it (yet, because I intend to, whenever I’ll have some time and cash…). You can consider it maybe if you have the time and money. There wouldn’t be much things to change from the original design, except a few 3D printed brackets to hold the MPCNC corners.

Also, the motorized platform is something I really, really want to do for a while, but it is a bit complicated to do mechanically speaking, especially if you want something steady as a rock… I currently have three ideas of moving table designs, not decided yet which one I should try…

Anyway, let me know if you need advices on how to build the table, I’d be happy to help :slight_smile:

Well If you can find that 1/8 collet, go for Makita without hesitation.
Even if I like Dewalt stuff, Makita is waaaayyyy better quality. Their tools are meant for professional use in the toughest conditions.

hello i have problem i need help i give the g code to my mpcnc and dying rong plant whats is happen now sam body i have mega 2560 ramps 1.4 (29.1 KB)

Hi Theodoros,

1rst thing: please create your own thread. This one has nothing to do with the problem your have :slight_smile:

second: Please give a more exhaustive explanation of your problem, how can we know what is going on with the information you provided?

My guess, based on the very very few information you gave us, is that you wrongly connected at least 2 of the axis. Try moving manually the X, Y and Z axis in repetier and see if they move accordingly.
Also, post pictures of your machine and any other relevant informations, like how did you create the Gcode??

1 Like

The 660 is pretty worthless outside of a CNC machine, IMO. 9/10 times I use a router it’s to round over some edges. The 660 can’t do that, because it doesn’t have a base. Is there a mount for the Makita? That would be my only concern.

Yea, the 660 is basically just a rotozip.

Phew, the bearings finally arrived, only took them two full months…

at least I can now go ahead and assemble the moving parts.

Turned out I have four of the same rollers. Do I have to reprint two of them?

Probably should have started this in the advice section.


Now I have to think about the table it goes on.

Looks great!

I’ll have to think about the table later. Meanwhile, the build must go on. I found a 3/4" plank with a crack down the middle and sliced that up for a frame. I’ll go with a shorter rail that was an earlier miscut, shortening the Y travel. I’ll cut a new rail for the Y gantry when I build a new motorized table, or when I need to do a bigger cut.

Being an lapsed AFOL, the Lego blocks are my attempt to build jigs to align the rails. This is gonna be long.

Wow, I really like yellow and grey. Good choice on colors.

Getting closer, hope the build has gone well so far.

Grey is a recommendation by the local shop for the “industrial” look, yellow is because my home is a Dewalt shop, for better or for worse. Since my dad bought a Dewalt 20v hammer drill that have pretty much displaced all other drills in the home including a corded Makita. I think the system is overpriced but I don’t want to buy into another system either.

It’s been a long road / Going from there to here

[attachment file=37338]
The Dremel is only here for a pose…

Had I known it’s going to take this long, I’d consider doing it up in red and white instead. O Canada!


It still need more work done before it can do the pen test and get dirty. I’ll ask in a new thread. But with all steppers connected, I know its mechanical limits. This G-code moves from home to far end as fast as it could without a tool: :stuck_out_tongue:

G0 X620 Y330 F7500

You’re on the home stretch! That gcode is really fast 125mm/s!