Second-hand MPCNC - rebuild

Greetings! Within the last few months, I have inherited a sizable (42" x 30") CNC machine that was presented as being based on MPCNC, but in a state of disrepair (many parts are broken/cracked) but otherwise complete except the tool. At the time I hadn’t heard of this amazing project, but have spent many hours reading and researching the background and history. It wasn’t until after I had started re-printing the 23.5mm Primo set of parts (90% done) that I realized they didn’t match! Going back through the documentation, I now understand there are several iterations including three different sizes (this one is 23.5mm). The questions and issues I’m facing now are as follows, and I’m hoping someone can shed some light and/or give some tips or direction:

  1. Which specific model is this? It appears to be a Burly but from reading the forums, there are different versions, and even some earlier that I may not be aware of… I’ve attached an overview photo. If any close-ups would help, I’m happy to take more.

  2. Based on the answer to #1, since the upgrade kits from Burly to Primo are no longer available in the V1E shop, where would be the best place to source parts? So far, I’ve been able to find bulk bolts and screws at a reasonable price from McMaster-Carr, and the Pulleys/Idlers/Belts either from the V1E shop or Amazon.

  3. Can the controller be identified from the photo? As you can see it’s running a super old version of Marlin (1.1.0-1). Is this directly upgradable to a more recent version?

  4. Maybe the most important question is which direction makes the most sense… Should I start over, re-printing Burly parts to use as it was originally built? Should I attempt to directly migrate to a Primo and fill in missing parts as I go through the assembly instructions? Should I disassemble everything and clean/inventory first then proceed? So many options here…

Thanks in advance for info and insight on what you would do!


It appears to be the older Burly model. I’d open the controller box and look at the board, could be anything, as a lot of DIY’ers used various boards and displays and modified the firmware for the board they had.
I have 2 Primo’s with the SKR 1.2 pro boards and the TFT35 V3 E5 touch screen.
Get a good pic of the board and post it

Welcome to this wonderful hobby, the rabbit hole is deep. :yum:

If you are really interested in this as a hobby I’d recommend just upgrading to the Primo.

Either way, make sure you are shortening the legs to minimum size, you are not going to have a lot of fun like this. :sweat_smile:

1 Like

Myself, I think you will find a Ramps controller in there, if so you will find an arduino with a ramps controller on top.

I had one, for a while. I moved my machine manually and and it knocked it out.

Just be careful if you move it back and forth by hand. If you see the screen light up, you are going to fast and generating electricity!

I would highly recommend that you upgrade to the Primo. It is a much better design. I upgraded from a Burly to a Primo and am very happy I did. You will be able to reuse the hardware on the Primo.
Keep us updated on your progress as you move forward.

1 Like

Here’s a few photos of what’s inside the controller housing. Couldn’t wiggle out the bottom boards, they seem to be glued in place. Maybe someone can identify from these? (Thank you!)

1 Like

Yes, I’ve had a great time so far just reading and learning. Figured it was about time to get some real live discussions and answers. lol

Do you mind elaborating on the “why” of shortening the legs? I feel like the answer could be obvious, like the rigidity of smaller dimensions, but I don’t want to make any assumptions. I appreciate the help!

The gantries aren’t easily moved due to the broken parts, but I will certainly take care not to move them too fast. :grinning:

Ramps 1.4/Arduino Mega. ReprapDiscount 12864 display.

If you use dual endstops, I believe the correct firmware is below (Someone wil correct me if I’m wrong)

Thanks for the recommendation. I read a lot of the threads when the Primo was announced when folks already had Burlys up and running… always fun watching the birth of new products. I’ve got maybe 2 more days of printing (I’m in the middle of the Core with 70-30-70 infill), then I should be able to get started in earnest. I’d better get those other parts here pronto… I will try to document as best as I can, maybe it will help the next person in the same boat. Thank you, I’m sure there will be plenty more questions coming soon.

1 Like

That was fast, thank you very much!

Many thanks, I will download and read the documentation on this release. So, this begs the question: Should the MPCNC use endstops at all? I believe I’ve read that many (most?) do not recommend using them, but I’m unsure why. I know on my 3D printer I have them on X/Y for safety, and I had to disable the Z endstop for the auto leveling probe (and PEI bed). What would be the reasons against using endstops for MPCNC? This Burly build does have at least 3 that I can find, but all of the clamps are broken so they’re just hanging around unmounted. Thank you!

1 Like

The endstops are used to set 0,0 (x & y) but more importantly they square up the x and y axis. You can run without them for sure - I started out with no endstops or probing on my burly but when I built my laser MPCNC (burly as well) I swapped the miniRambo to that machine and upgraded my CNC burly to a Rambo and added the dual endstops and probe.

I’d say you’re safe either way, with or without the endstops but I also cannot say how much of an impact the auto squaring had on my machine as I was always too impatient to “do all the things” to test and tune the machines. I guess I’m more of a bang it together and get it making/doing stuff guy. Perhaps others can fill us in on the real world impact/benefit to the auto squaring. I guess I should also add that repeatability is another reason the endstops are nice but you could set up a machine without endstops to give you pretty good repeatability too.

Ryan somewhere explains it in the manual. The longer the legs, the further the router needs to travel away from the core before cutting, so there is more chance for play. Not a good explanation, but maybe it is sufficient.

Some observations:

  • I only see three heat sinks → implying only three drivers → implying serial X/Y motor connections → implying no (or useless) endstops. (I suppose I should elaborate: you need five [5] drivers to use endstops for squaring. With just three, there’s no way to move only one side of an X/Y gantry, they always move in tandem. SOP is to set up hard stops that are known square, move [slowly!] the gantry against the hard stops, then turn on the machine, allowing it to start at a known square home.)
  • Longer legs mean the bit is reaching further from the plane of the gantries (the median of which is the plane of greatest rigidity), and therefore the more torque can be applied to the moment arm of the Z-axis just by the simple action of cutting stock. You can alleviate some of this by putting your stock up on blocks, and keeping your cuts as close to the gantry plane as possible, but then you have a large amount of Z-axis sticking up, generating inertia, and potentially causing issues if anything pulls on it. General best practices say go for short legs with a removable spoil board for thicker stock/vertical milling, and be very, very mindful of anything you attach to the Z-Axis.
  • The Z axis/Tool holder for the Burley can fit in a Primo, but it’s not an exact match (IIRC), so you may need to either rebuild that portion, or Frankenheimer it.
  • The Burly specs 6mm belts, the Primo specs 10mm belts. So, technically new belts, idlers, and pulleys. Ask how I know…
1 Like

This is good info… it means I can keep the hardware but not use them until later if necessary. As a matter of fact, the Dual Endstops page of the documentation (Dual Endstops - V1 Engineering Documentation) specifically mentions new users should not use endstops at first.

Makes total sense, thank you! I was actually very surprised at the viability of an MPCNC, but when I started to print at higher infill densities as instructed by the documentation, it made such a huge difference. My first attempt at a 70-30-70 Core failed at 116mm this morning… guess I need to suck it up and print the entire thing at 70, or use a different slicer (Cura doesn’t do variable infill directly).

Many thanks for your observations, this is the type of feedback I love to read and learn from. I went back out to double check everything and did find a total of 5 endstops. The strange thing is that there’s two in one corner, one in the next corner, and two on the Z-axis(?!). At least that’s where all of the failed clamps are located.

I appreciate the further explanation on leg length… I will trim it down to keep Z-axis play issues to a minimum.

Copy that on the new Z-axis and other parts. My plan at this point is to finish printing the new Primo 23.5mm pieces, completely disassemble the existing Burly to inventory/clean parts, then start from the beginning with the Primo build instructions. New bolts/screws from McMaster-Carr and new belts/pulleys/idlers from Amazon should be here in the next day or two… just in time to build a table for it. Thanks for the tips and feedback!

That’s a damned good observation!

To fully use the auto-squaring capabilities of the firmware, each stepper needs to be individually controlled as well as having an associated endstop switch.

So, you’d need to populate the two empty stepper driver locations on that RAMPS to fully use dual endstop firmware. I’ve marked those spots up in yellow below.

I prefer not to use series stepper wiring for a variety of reasons, and the drivers are really inexpensive- so I’d probably add them while doing the rebuild (if you stick with the RAMPS, which is fine for now)

1 Like

I appreciate the advice here, thank you! It took me awhile to digest this info while diving more into the hardware (as I was waiting on tool and material shipments). As you guys noted, my current Burly is wired for series stepper motors since single drivers are installed. I would need to add the two new drivers (and jumpers) to the E0/E1 on the RAMPS 1.4, so that these correspond to the X2 and Y2, correct? Then these would need to be defined in the Marlin firmware as well (which I think there’s a version for duals already). I’ve decided to go ahead and add everything from the start, so I don’t have to mess with it later. This includes the independent motor control and all 6 endstops (X, Y and Z, both min and max) unless there’s a compelling reason not to… I figure it’s easier to do it all at once and wire management would be simpler.

Today I’m finishing up the table, then will start tearing down the old in preparation for the new! Thank you everyone for your input and guidance!

1 Like

Z does not need endstops for the Burly. There is nothing to be gained.

I appreciate the info - this is an inherited Burly that is in the process of being updated to a Primo. Trying to decide whether or not to do it all up front or mess with it later on after gaining more practical experience.

Same goes for the Primo. :stuck_out_tongue: The Home-Z was removed in the Fluid Confing anyway. :slight_smile: