IPGregory - MPNC Build Thread

[IPGregory]

Intro and criteria for this build.

My 2nd Acorn Build Thread. 1st one is in my sig below (QueenAnt Pro V2). I plan to take a different approach with this one. The last one was really more of a blog and got pretty long with lots of inline pictures and details spread across many posts. This time I want to keep it much shorter (This intro aside) and keep the pictures in an online album to show the progress. I hope this makes it easier for people to see the progression and if needed to help with any issues I may run into and need help with. But more importantly it should keep Keith happy...

Little background. My 1st build (and still current at time of writing) machine is a 1m x 1m sized kit from China of an upgraded OpenBuilds type machine. It was the replacement for a basic 3018 that I used to dip my toe into the CNC world. It is a 110v, aluminum extrusion frame machine using a 2.2Kw water cooled, 24K RPM spindle. Linear rails and Ballscrew motion driven by Nema 23 Closed Loop Steppers from an Acorn (4). 725mm (x) x 785mm (y) x 110mm (z) work area.

I hadn't long had it up and running before I realized it likely wasn't going to take me where I wanted to go with my projects. My plan for it was primarily Acrylic with some Wood and the occasional non-ferrous metal like Aluminum or Brass.

My first issue was I had undersized it. It was OK, for the Acrylic type work, but for wood, I was spending a lot of time tiling or what I wanted to do was unrealistic to do on the machine.

The second issue I ran into the first time I tried to cut some aluminum brackets as part of the machine build. The attempt failed miserably (There was smoke, there were many broken bits, fire was avoided, but it was close). 95% of this was on me and my lack of knowledge and experience of working with Alu, but I also realized the machine wasn't going to grow with me very far before it would struggle and become the limiting factor.

At this point I sat back and considered what I wanted to do. I had learned by now that an Alu Extrusion Frame Machine is never going to be very rigid. No matter what upgrades or other parts I bolted on, this would be an underlying problem that would be difficult if not impossible to overcome. Without upping the rigidity, anything above plastic or wood type work was going to be limited and any detailed plastic work might suffer finish quality. Increasing the size was possible, but it came with 2 problems. 1, it would cost almost as much as replacing the machine and 2, it would just increase the rigidity problem.

Just to clarify, I am not looking to work in steel or higher with the router. Plan is to get a Mill and Lathe down the line for that type of work. Non-ferrous is the limit, but I want a machine able to handle it relatively comfortably with reasonable finishes and Operation run times.

At this time, I started to research what else was out there and what options I had to solve the problems, either by upgrading or maybe replacing. I looked at different designs, different price points of kits and ready built machines and so on. What I found was that other than the still relatively small (and yet to be released at that time) Langmuir MR-1, pretty much everything out there in the Hobbyist price range was pretty much the same as what I had. The drive systems might vary, the sizes might be different, different controls setups, different spindles, etc., but they were almost all still Alu extrusion based machines or worse. Alu frame aside the machine I had was actually pretty well spec'd with ballscrews, HGR rails, closed loop steppers, etc., compared to much of what else was out there even at 2 or 3 times the price or more.

So I kept looking and researching. I wasn't in any rush, as I had a running machine I was using and continuing to learn and gain experience with. Also, the more I used it, the more I understood potential limitations, design choices, etc.

Eventually I ran across something different.

[IPGregory]

Inklings of an idea for build 2

I ran across something called the Print NC (PNC). On reddit of all places. When I first ran across it, I saw it as an oddity. It was a Mild Steel RHS framed machine using 3D printed plastic parts for motor mounts, screw plates, etc. Since one of my main issues was rigidity, plastic parts were nowhere on my wish list for the next machine. But it intrigued me for some reason, so I carried on looking into it while still actively researching the options for the problems with my machine.

There is a web site and it has a full parametric Fusion 360 CAD model of the machine which I downloaded to play around with. The reddit group is not very active but it points you to a discord that is incredibly active and full of ideas and options. The more I read into it and played with the CAD, the more interesting it appeared.

The initial design was created and matured by a guy called Mark Hogan (Hoges on the discord/reddit) in Australia. He wanted a machine that pretty much anybody could build with basic hand tools and a 3D printer. Kudos to him for the work he put in. He made the CAD model, he arranged pricing and supply of a parts kit with Chinese vendors, he set up a web site that included a frame size calculator and BOM creation tool that allows you to order a kit or lists all the parts so you can source yourself, he made example videos of the assembly tasks like drilling and tapping the steel, etc. He put a LOT of work into it. Then he turned round and made it community license and handed it over to the community. (We need a thumbs up Smily).

But it's still got plastic bits, and I (at the time) didn't own a 3D printer, why are we still talking about it?

Well yes, it has plastic bits... And no, it doesn't... ??

The objective of the stock PNC as I said above is to allow somebody to easily make a fully working CNC Router with minimal tools. The expectation however is that once up and cutting they could if they wanted to, use the machine to bootstrap itself by making parts to replace the plastic (and wood sometimes) parts with milled aluminum replacements. Apparently, many build the stock machine, with the plastic parts and use it quite happily, as it's a very capable machine based on user feedback in its stock form. There are however 'Pro' and 'Mod' versions of parts that the community has developed and shared that allow you to replace those plastic bits with metal ones.

I am meanwhile reading the discord, chatting with people who have built one, messing around with the CAD file, substituting plastic bits with Alu bits and generally playing around. I am not however finding anything that's looking like a solution to my machine problems anywhere else.

But an idea is forming.

[IPGregory]

The design for build 2

The more I look at it, the more it becomes apparent, the PNC and my current machine are almost the same. Maybe that's why it intrigued me to begin with. Well, that and the fully 'build it yourself' aspect. My machine doesn't have plastic parts and the PNC doesn't have an Alu extrusion frame. If I upgrade the plastic parts to milled Alu parts that I can (hopefully) make on my current machine, I end up with a steel framed machine that can be made in whatever size I want.

Hmmm.... My problems currently are Alu frame and too small, doesn't that answer the issue? Fires up Fusion....

Beginning of the year, I ditched the Stock official CAD file and started new. I worked out I could just fit a 48x36" work area on my existing bench and into the space I have in the garage. This would allow me to tile up to a full 8x4' sheet if I needed to but was big enough I could get away without tiling most of the time. So I drew up the basic PNC base frame, Y and X setup in this size. Then added milled Alu Y Rollers, X Rollers and all motor mounts from the PNC Pro options the community had developed. Added the plan to re-use my existing bench, steppers, drivers, spindle, all electronics and my Acorn.

Milled PNC (MPNC) V1 was born.


With the dimensions worked out, I contacted ThreeDesign, the company Hoges set up and ordered custom length SFU2010 ballscrews and HGR20 rails with HGW20CC carriages, cable chains, couplers and so on. Just the stuff I needed to change due to the frame size change. I looked around and the prices he had negotiated were better than I could source the parts for, plus I wanted to 'give back' in some way for the work he had done.

While waiting for those I continued to refine the design. I looked at the Pro parts and decided some of the design choices might not be what I would have picked or worked best for me. I looked at the inline drive setup with the steppers overhanging the frame. I decided I wanted a bench that had some form of built in chip gathering and an enclosure that my current bench didn't have. I don't have a 220v outlet in my garage. The only one is in the kitchen on the other side of the house, but I decided after talking to various people who have run both that 220v was a better option for the machine and I would add an outlet in the garage, so the spindle and VFD changed, and so on.

Maybe I was going to build twice, but I wanted to try and avoid crying twice. Long story short, it got to MPNC V10...

This was what I set out to build. My Nema23 3.0nm steppers had turned into ClearPath Nema23 SDSK servos, the servos were inside the Y rails and the gantry and running belt drives to allow for a bigger work area in the same footprint. It was up to 50x36" cutting area in the same frame footprint.

I ordered the changed/added parts (Except the servos for now). Hopped off to Home Depot and picked up a new load of lumber and ply for the new bench and started the build.

Oh, and I ordered a 3D Printer...

[IPGregory]

Steel was a PITA to get. Apparently very few companies want to deal with small orders and just can't be bothered to respond or come back with ridiculous prices. But after a few weeks of searching, calling, emailing I finally found a small local company who was happy to take my money and picked up the frame steel.

I didn't order the ClearPath Servos because I was still on the fence about ballooning the budget that much at the time. The design worked with either steppers or servos still then, so I held off (luckily).

Why the 3D Printer if I wasn't printing the drive parts? Well as part of going over the design, I had realized that there are many parts that are quite acceptable to be made of plastic without impacting the machines operation or strength. Chain supports, sensor supports, cable guides, covers, etc. I had wanted one for a while but hadn't pulled the trigger. This project was the perfect reason. Time to learn that then..

Steel cleaning and initial hole marking commenced. Bench design was done so that build started as well. Needed somewhere to put the machine during build after all. So that was the primary focus.

Winter turned to spring, then summer arrived. I live in the Desert SW of the US and my garage is uninsulated and not air conditioned. All physical build work on the main machine came to a halt as humans melt in those temperatures.

With nothing else to do over the summer but build the control and electrical cabinets (drawers), Fusion called and why don't I make a few 'tweaks' to the design crept in? Another long story short and as fall approached and temperatures cooled, MPNC V13 was now 'the Design'.

What changed? Well, a lot. The servos were now a fixed part of the design but had morphed into Nema34s on the X and Y, now mounted outside the frame tubes on the Y, but still in the frame on the X. Y Drive screws had migrated to the sides of the Y tubes from the tops. Luckily, I hadn't ordered the Nema23s, but I now ordered the Servos. I had ordered the Power stuff over the summer as it was part of the controls stuff, I was working on in the house.

Why Nema34s on a little bench top router? Blame the Gantry. One part of the PNC base design I wasn't sold on was the Gantry and Z Assembly. The stock setup has a rail top and bottom of the gantry that both limits Z Clearence and I believe can be subject to variable nod over the length of travel as the rails are on different faces. I completely reworked the Y Rollers, the X Roller Drive and the Gantry. Instead of an 4x3x0.188" RHS steel tube, it was now a 6x4x0.25" Extruded Aluminum Tube. Wait, what? Aluminum? Didn't we ditch that?

Well kinda. We ditched Aluminum Extrusion. Extruded Aluminum Tube in thicker wall is of comparable strength to the thinner RHS steel tube in the PNC stock design but has significantly better dimensional accuracy. Basically, the steel can be fairly off and is difficult to correct due to internal stresses, so machining it in thinner gauges is problematic. Alu tube is better off the shelf. I moved both X Rails to the front face of the Alu Tube to make it easier to align them and then completely redesigned the Z Assembly for the new setup. Part of this increased the Z Clearence from 4" to 6.5" so I also beefed up the Z assembly to cope with the extra lever forces. Finally, the Gantry tube will be filled with steel black pipes for cable routing and air flow for the servo and UHPC for damping and mass. Y rails and Base rails will be filled with grout and PVC Pipe to provide damping and mass while still allowing cables to be routed through them.

So the Y servos were upped to Nema34 to cope with the increased weight of the gantry, Z assembly and spindle. The X Servo was upgraded because the Spindle is being upgraded to a heavier ATC unit as well as the beefed-up Z Assembly. The Z Servo remains as a Nema23 with a brake because the Spindle assembly will have gas struts to counterbalance the weight. X and Z servos are belt drive. X is HTD-5M x 15mm with 36 tooth pulleys. Z is HTD-3M x 9mm with 30 tooth pulleys.

Work area is up to 50x40" or a tad under for the same frame footprint and drive parts due to the side mounted screws.

It's come a long way..

[IPGregory]

That pretty much catches it up to now as of posting this.

Production of the Alu milled parts is underway on the current machine. It has its challenges as the machine is really not up to it, but by slowing everything right down, detailed CAM simulation runs, careful selection of the Op types and tooling, it is making usable parts. I have a problem with 2 sided operations at the moment, but I have a plan to try and get past that (hopefully).

Some of the parts I might choose to remake when the MPNC is up and cutting to get better finishes, etc. It will also help me learn the new machines capabilities and limits.

The MPNC will get a brand-new Acorn (4), an Ether1616, a WMPG, 3D Probe and Toolsetter that are all here ready and waiting. I temporarily put the probe on the current machine to help with the parts runs, but it's destined to be permanently installed on the MPNC. The ATC spindle will be used as a 'Quick Change' rather than 'Auto Change' for now, but there are plans for a carousel and a new metal bench in the future to make full use of it and also to better support the machine. I just need to buy a welder (to go with my new 220v outlet when its installed) and remember how to weld since it's been 20+ years since I burned holes in anything.

That's it, I'll shut this down for now by saying I added a shared album of pictures of the current machine's (QueenAnt Pro V2) build in my sig below if you want to see a comparison.

The MPNC picture album is also up in the sig and here - https://photos.app.goo.gl/ees2SER6C2Aw8AJ89. I will continue to post to it as progress continues.

This was a bit longer intro than I planned but nearly a year of work in 5 posts is an improvement over last time where it was about 11 or 12 pages worth...