Friday, July 21, 2017

Ultra MegaMax Dominator (UMMD) 3D Printer

UMMD has lots of  white and UV lighting built in.  The UV really makes fluorescent filament prints look great!

Ultra MegaMax Dominator (aka UMMD) is my third 3D printer design/build.  I used much of what I learned from its predecessors, and tried a few experiments, resulting in a very high quality machine that produces very high quality prints.  Time will tell if it meets my reliability goals.

Basic specs:

This project started about a year ago with the design of the XY mechanism.  That sat for about 6 months until I decided to finish it up, adding the Z axis design and finally the frame and enclosure.  I posted progress reports at the reprap forums as the design work and construction progressed.

A lot of the work was done in the machine shop at the Milwaukee Makerspace.  I could not have progressed as far or as fast without the people, equipment, materials, and tools available at the makerspace.  I strongly recommend that you find and join your local makerspace.

Upcoming posts will be using a lot of CAD drawings, photos, and hand drawn diagrams to explain the details of the design, hopefully without too much text.  This printer was designed using DesignSpark Mechanical, a very powerful, yet easy to learn and use, free-enough CAD package.  I will be posting the master file for download, but you'll need a copy of DSM to open and edit it.  A few people have asked, and the answer is no, DesignSpark can't export the design to a different CAD format.  You'll just have to load DSM and open the file with it.

Here is a link to the final design file:

Someone was able to convert the rsdoc file to a STEP file so it could be brought into Fusion360:

In the future I'll be switching over to Autodesk Fusion360 because of its built in ability to produce files for CAM.  Fusion360 CAM files can drive CNC mills and routers at the makerspace.

UMMD printing a fluorescent green vase with PETG filament.   The blue dots in the background are reflections of the UV LEDs in the dual layer polycarbonate enclosure panels.

Parts sources

I bought the used, 40x40 mm t-slot from a local scrap yard for $1 per lb.  Cast aluminum tooling plate came from Howard Precision Metals in Milwaukee for $2 per lb.  Motors, belts, linear guides, power supplies, and miscellaneous hardware were purchased through, ebay, and Aliexpress.  A few other parts came from the Milwaukee Makerspace and from people at the makerspace.  All of the machining of parts was done at the makerspace.

I won't be providing a BOM for this printer, but you'll be able to gather a lot of it from the CAD file. Your chances of finding the exact same parts I used at reasonable prices are slim to none, so you're going to have to adapt the design to the parts you have or can get.  If you can't figure out the necessary mods to the design, you might be better off buying a kit.

Details, details...

Check some of the other posts for details about:

Print Bed and Support Structure
CoreXY Mechanism
Z Axis Design Rev 1
Z Axis Design Rev 2
Z Axis Design Rev 3 (final)
Frame Construction and Enclosure


  1. Hi Mark,

    i've read a lot of your post and find them very interesting. I was recently offered a job at my local maker space as a 3d printing instructor. Is there a way to get in touch with you via email or something? I've designed a few printers myself and i'm pretty sure i could share some info with you. I also have a few questions.

  2. patnap71@gmail.comApril 7, 2018 at 1:40 AM

    Do you foresee any issues if I did this in 700x300x300? All the design features should still work just lengthen here and shorten there...

    1. I think it will work, but you'll probably have to get a custom bed heater made.

  3. Hi Mark, Love your articles on making 3d Printers and the Technical stuff. I have learnt a lot from your site and I would like to thank you for that :)
    I purchased a TronXY 5SA as my 1st printer and almost instantly discovered so many engineering problems with the design. I started searching online for the answers only to find that nearly all the owners had the same problems and they all had different fixes which generally were plastic parts. My background is in the Motorsport/Engineering field and it was only after finding your articles in the Open Build forum that the answers became clear. Since, then I have been making nearly a whole new printer, although I have also made a 1 metre new Frame now too. I plan to print ABS for a SuperKart Bodywork and wonder if you could point me to some tips for large ABS printing like some of the stuff you do. One of the questions I have is, when printing large items, say 1/2 metre to 1 metre, is there a problem with the layer cooling too much before the next layer is deposited? Thanks again - would love to see some of your large ABS prints :)

    1. Thanks for the kind words.

      The largest ABS prints I have made were 500 mmm tall vases with walls about 1.8 mm thick (3 perimeters with a 0.6 mm nozzle). Heating the enclosure to 50C kept them from splitting.

      From what I have seen of Superkart cars, I would think that vacuum forming would be a better way to make the body, and certainly much faster than 3D printing.

  4. Hey Mark,

    I was looking for some info on belt-driven Z axes, and it looks like I have (accidentally) stumbled across your blog MULTIPLE times researching different 3D printing topics, and didn't even realise it was the same blog!

    The UMMD design solves many of the issues I had also noticed with 3D printers (I own a Creality Ender 3, which I've already replaced a bulk of parts with machined CF and linear rails, and changed to 3 point leveling), but I've also wanted to do a larger format printer for a long time. I help run my local Makerspace, and have access to basically the same tools you do, as well as plenty of people there who can help me if I get stuck.

    If you could go back and do it again, would you change anything on the design? I've got your Fusion model, and have started the process of changing everything over from freedom units to metric.


    1. Sure, there are a few things I would (and did) change... See:

      I changed from SmoothieBoard to Duet controller (made the machine very quiet) and settled on a BondTech BMG extruder (great!), and moved the extruder and motor over the X axis bearing block with a teflon tube running down to the hot-end. I am now looking at flipping the brackets that hold the bed assembly on the Z axis over to bring the bed up another 40 mm, and then I can move the hot end 40 mm closer to the extruder.

      I switched to 40 tooth pulleys on the Z axis and put in glass core belts instead of the steel core belts. The belt clamp design was changed after there was a problem with the original design.

      I'll be running some tests with an accelerometer on the hot-end to check mechanical resonances soon.

  5. Hello Mark,
    this is an awesome design - after dipping my does with cheap printers I'm starting to fiddle about with something bigger and sturdier, hence my look in your designs.

    One (probably silly) question:
    you have the bed suspended by belts, that is supported by a single motor and (to my understanding) relies on the holding torque of the motor to keep it "up" from the time you turn the machine on. Would it make any sense to introduce a counterweight mechanism, so that the bed weight (and any friction introduced by the counterweight pulley) is offset by it? That way (in theory), while the Z motor is free, when you move the bed around it should "stay" where you leave it, and hence the motor should only account for moving the bed with precision, with less strain on both the belt and motor.

    I will probably test the theory out when I have the build done (it's gonna take a while), but would like your feedback.

    Thank you in advance!

    1. Thanks!

      The Z axis uses a 30:1 worm gear reducer on the motor that completely stops bed movement when the power is cut. Take a look at the post on the Z axis design version 3, and a more recent post on Z axis updates.

      The problem with counterweights is that they unload the belt/motor which can lead to backlash in the Z axis if you set the printer to lift on retraction. With the worm gear reduction there is no backlash because the gears are loaded by the weight of the bed assembly at all times.


Leave comments or a questions here and I'll try to post a response as soon as I can, if I can figure out how to get it working...