MPCNC Build - Texas

Hello,

Downloaded some of Ryan’s robo3d mods 5-6 years ago and saved the MPCNC project for a later date. Today is the day!

Dusted off the Robo3d and got my dual end-stop kit a few days ago and it went together pretty smooth. It’s not pretty yet but it’s mostly functional. I’m having one issue with the z-axis and before I started tightening bolts willy nilly I figured I’d ask for a second opinion.

The z-axis falls after raising up a few millimeters. I’m not sure if it’s because my bolts aren’t tight enough to support the weight of the spindle or some other issue.

Here’s a video, thanks for any help in advance.

https://www.youtube.com/watch?v=fczsA_UAoRI

 

I would make sure your coupler is on tight. I don’t know of any other screws that you could tighten.

I’m assuming this is a rambo controller flashed by Ryan?

1 Like

The coupler was too far down on the motor shaft which didn’t leave enough room for the lead screw. Took apart the Z axis and adjusted everything, seems to be working good now. Thanks!

 

I forgot to print a pen mount… Not bad for a ball point pen zip tied to the dewalt!

 

 

1 Like

Grats on the great leap forward! Time to get it dirty!

Another CROWNing achievement!

I’ll let myself out.

2 Likes

Where are you in TX? I am in south TX

Lubbock… lol

Printing some cable chains and ordered some PET sleeving on Amazon. I was doing all my testing with Octoprint since that is what I was running on my printer but have since switched over to CNCjs, which is really awesome and much better suited to this machine. I’m going to build an enclosure this weekend since my machine is in my bedroom.

Something like an xbee would be better than bluetooth. Wireless for the jogging and setup seems fine to me but running a job that way seems like it will cause problems.

I originally made my table 36x36 however that didn’t leave hardly any room with an enclosure to run cables.

I made the table 42x42 and used some tongue and groove bead board I found at Lowes with quarter inch plywood to make a simple enclosure. It’s not anywhere near airtight but should keep the majority of the dust inside. I may redo this in the future or just move the machine to my shop.

 

I also got my CNCjs setup working without running a long cable to my server by using Docker swarm on a raspberry pi. There were so many ways to do it but I settled on that option since I already had a spare pi.

I could have simply installed CNCjs on the pi itself but then it wouldn’t be integrated with all of my other infrastructure. I don’t plan on leaving the CNC alone like I do my printer, so it doesn’t really matter if I’m able to access it outside of my network anyways but my OCD likes it the way I have it now.

 

Hopefully I can finish setting it up and dialing it in later today and finally make my first cuts!

2 Likes

Docker is a very good way to do it. I keep my home iot scripts running in containers. It just makes it a lot easier to replace them and they are all one off configurations, so it doesn’t make sense to keep whole images for them.

I have started running my cables above the corners. I show it on my red and black build, opposite of what is shown in the instructions. Smaller footprint and less dust collecting corners.

First cut was a success! (after a few failures)

 

I wasn’t setting the temporary zero offset in CNCjs. I’m not 100% sure what the difference is between work coordinates and machine coordinates are but from what I gather I wouldn’t have to do this if I wasn’t using endstops.

 

Thanks again to Ryan for the amazing design and the people who helped me in this thread. I’m excited to make some stuff with this thing, problem is that I don’t know what to make.

 

 

https://www.youtube.com/watch?v=KHOMnI34jLg

1 Like

Was some carving with a v-bit on some (thankfully) 1 1/4" dense foam and the coupler let go of the Z axis. Thankfully no damage.

 

That coupler is the only thing that has given me issues with this machine so far. Is there some trick to keeping it secured to the leadscrew? I’ve tightened the grub screws as much as I can.

There should not be very much force on the lead screw. Have you lubed it and made sure it moves extremely easily by hand, no binding?

Another trick is to run the grub screws all the way in without the lead screw there to open up the holes then add some threadlocker and re seat the lead screw.

My Z axis parts barely fit the conduit, even after generous filing. Even now my Z axis stays up by itself with the motors off. Could this be part of the problem?

I printed the parts myself, everything else is from you.

The lead screw moves fairly easily by itself, no binding or anything. I’ll try some thread locker.

 

Maybe, hard to say honestly. Loosen up the tension bolts. The fact that you had to file anything is extremely bad, it it did not fit as is something is wrong.

The only two parts that required filing were the upper and lower Z brackets. I printed those right after I had to take apart my printer to clear the clogged nozzle. everything else fit perfectly minus those two parts. When I had the Z axis off today to apply thread locker to the grub screws I noticed that both of those parts are sliding much easier now. Probably to do with the heat inside the enclosure and actually being on the conduit for a few weeks. I was going to reprint them but I honestly think they are fine.

While messing about today I learned not to carve out large chunks of MDF without some sort of dust collection setup. Even though the machine is enclosed there is a layer of dust over just about everything, and the inside of the machine required a thorough cleaning to get dust out of all the nooks and crannies where the bearings roll.

I got some of the scrap 2x4s from my first attempt at building a table and started to push the machine to it’s limits. It cuts way faster than I thought it was going to do. I had been running at a measly 8mm/s feed rate with a 1mm depth of cut. So far I’m up to a 15mm/s feed rate with a 5mm depth of cut and it can easily do more.

It’s also cutting perfectly accurate. Anybody have some good ways to test and calibrate repeatability without a dial indicator?

Going to get another week or so of test cuts under my belt while I settle on a first project. Can’t wait!

 

 

1 Like