Roland PNC-3000 CNC

Does anyone know what it would take to get this up and running? I have a project I'd like to use it for.

Though on the cnc tool inventory, I have not seen this item. Anyone know where it hides?

This machine is legendary for the frustration it has generated. It’s been retired so long that, perhaps it is not the best machine for Hive13 for various reasons. It may be a solid tool, though, in the right hands.

So, I’d say there’s absolutely no harm in trying. Go for it. I see a potential for diminishing return, but maybe the legends are wrong…


Dave, it's on the lounge roof, above the big CNC.

Lorin, any particular reason it's been a frustration? Looking online shows they are annoying to communicate with as built, since they don't use G-code, but there are work arounds and some people have replaced the controllers with ones that work with Mach3 just like the big CNC.

I specifically would like to use it to machine the backing plates for some enamel pins, which I'm sure at least a few other hive users would enjoy doing. If it's too much work to get it running though, I'll pass.

Last person that got it running was Jim Shealy, if I remember correctly. And yes above the lounge, on the right side when looking at it.

Is what you want to do something you could do on the shapeoko? Might be better suited to putting your efforts into.

The pins are typically milled from bronze or copper sheet, and pretty good repeatability is required to get the small details.

I'm not sure what the shapoko's capabilities are, but it may be possible.

Shapeoko is actually pretty stout, even though it may not look it. Ryan, Dave, and Myself got it working quite well, but nobody’s had the time to do the finishing details:

  • Strain relieve and tidy up the wiring.

  • Make a case / enclosure for it.

  • Dedicate a monitor, keyboard, and have a place to put them along with the Pi Ryan set up.

  • Other minor details.


Kevin, I don’t think the Roland would be “too hard”, but it won’t be too easy. Obsolete software, proprietary code, etc make things a hassle. If it’s worth it to do workarounds, go for it – I’m sure it will work.

Is that the one people have talked about selling several times because sells for more money than it is worth? Aka, they are kind of special pieces of equipment that really hold value only for the people that already use them?*

*All my information is based on heresay from random people at the hive while pointing upwards at the hive.

Yes it is the one on top of the lounge.

I was looking into reviving it a year or so ago but apparently there are some “quirks” that make it quite difficult to use.
The main one being that the collet is some weird version and difficult to find.
Also it uses some weird proprietary language, but people were working on getting it to talk Gcode.
I found one on eBay that went for over $2000. So I’m thinking the collector market may be more logical.
However if someone wants to fire it up and dig in a bit more, I’m happy to help.


I’d help too. It was a serious PITA when shealy got it going with dosbox and roland software. However, I did a search and it looks like the motors, etc nicely match existing open source CNC electronics. Since there’s a some PID loopback spindle and other intricacies to handle, smoothieware would do it quite well, and here’s someone who has done it:

I’d be willing to help out once I get some of my higher priorities out of the way. I have plenty of familiarity with smoothie…


I'd love to be the one to lead a smoothie conversion project, but I'd need some serious help from y'all that have done this type work before. Let me know when you feel up to diving into it with me.

Did a quick search and found some interesting links:

I dug around ebay and couldn’t find anything. I’m thinking it has fallen off the interested list.
We should probably resurrect it. :thinking:


Does anyone remember the collet issues Jim had? If I remember correctly he was using paper somehow to make things work. This could end up being a bigger project than just electronics. Also with the limited workspace I’m not sure that the Shapeoko couldn’t accomplish the same task.


In terms of POI on time invested.
the SHAPEOKO will finish faster than starting in on the Roland. It also has limits in collets and tools, but has worked, Will work. Just needs some Finish work and a home corner in HIVE.
There are also a couple of other projects simmering and taking time.
I would encourage finishing them, before we dig out the Roland.
Regarding a smoothie project. Lorin and I have interest as this CPU is picked to drive our Delta printer project. Collaboration is a good thing.
So not saying no, just evaluating project priorities.

I do think engraving as a base for an enamel pin or other jewelry is a worthy capability. I’d like to see small fine metalworking added to our capacity.
This would also add jewelers saws, more riffler files, mallets, a jewelers anvil, ring sizing, polishing, silver soldering. An oven to drive off wax for Investment casting.
An oven for enamel. This stuff can all fit on a well designed bench

That’s ROI

Dave, 100% agree that other projects need to clear up first.

I think the spindle is ground for a proprietary collet, and that there is only one size to choose from. (I have a fairly large pile of small mills from a bulk purchase of tooling, some might work.) Both Centroid and Masso have really interesting controllers, pretty cheap all things considered. Far more user friendly than older software guaranteed. I do know a fellow who is doing a tool and die apprenticeship not but a few blocks from the hive, I could ask if he needs a practice project :slight_smile: But, time/money/energy… I think the machine to look at for engraving and small modeling might well be from Datron. Desktop extremely high speed milling.