Address

Author Topic: MendelMax - The Big One  (Read 4737 times)

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
MendelMax - The Big One
« on: May 30, 2012, 03:04:39 PM »
I decided I'm going to build a MendelMax with a 1-meter square build envelope.

It sounds simple, but it will not be easy.

My plan is to simply scale the MendelMax frame as described on their site, http://mendelmax.com/tikiwiki/tiki-index.php?page=Scaling. It seems somewhat simple, however, there will be some large engineering challenges.

  • Y axis - This is a large issue. Need to keep the bed quite stiff and prevent bowing of the rods due to the weight of the carriage, frame, and bed itself. 8mm rods would not be enough. Since it is built on T-Slot, i will look into slide bearings designed for this task, from t-slot manufacturers. The drive will also need to be largely upgraded to sling about a heavy carriage. This will either be a large NEMA34 stepper, or possibly a servo drive.
  • X axis - this will be another challenge, at the least requiring beefing up of the rods to keep the Z true during moves. The weight of the carriage will not increase, therefore it should require relatively small modifications. The motor for X will need to be upgraded slightly if faster print speeds are desired, but since the weight of the carriage is the same, only small increases in belt drag due to length will need to be accounted for.
  • Z axis - This should not be too hard, as not much changes here. It's wider, but not much heavier. I will most likely use larger steppers to drive the heavier load of using thicker smooth rods..
  • Drive system - This will need to be upgraded. Possibly using larger current-handling drivers such as GeckoDrives. Otherwise, the Y axis may need to be driven by a servo system. This may be complex if the controller does not handle direction and step commands as steppers do. I will need to locate and price out systems..
  • Bed - This will need to be made so it is flat and smooth. Can most likely be accomplished with a combination of slightly thicker aluminum and glass. Possibly countersinking the screws into the aluminum bed, then setting glass atop it. The glass will need to be thin as to not add too much weight to the Y axis. Speeds will most likely be lower. The bed will be heated by either one, or multiple 110v mat heaters to bring it up to temperature.
  • Extruder - This will be a standard Wade's extruder with hobbed bolt. Heater will be a Budaschnozzle powered by a cartridge heater. Will use a fan to help increase build speeds.
  • Electroncs - Possibly a custom solution here, depending on the size increase of the steppers involved. May switch to a separate Arduino shield that will breakout the pins to the drivers being used.

The BOM I have begun modifying is here: https://docs.google.com/spreadsheet/ccc?key=0Akfb5p719f2-dDA0cWxUM1o4TkNUNlZBRnVyRFNoc2c

So far, the total for the T-slot alone is about $180. I will begin pricing out drives and metal shortly, most likely from onlinemetals.com.
« Last Edit: May 30, 2012, 10:51:01 PM by Chorca »

ThantiK

  • FamiLAB Member
  • Full Member
  • *
  • Posts: 237
  • Resident 3D printer guru.
    • View Profile
Re: MendelMax - The Big One
« Reply #1 on: May 30, 2012, 03:48:38 PM »
This is an effort to use those massive stepper motors in your boxes, isn't it...


Also, as far as engineering challenges: You'll probably need to upgrade to larger diameter rods, with appropriate linear bearings.  Will 10mm rod work for this build envelope?  As mentioned in another thread somewhere, you can brace rods from underneath and just use 270 degree linear bearings to clear the adjustment brace.


I'm getting much better with OpenSCAD so maybe any editing that needs to be done to existing mendelmax parts I can be of assistance with.
« Last Edit: May 30, 2012, 03:55:59 PM by ThantiK »

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
Re: MendelMax - The Big One
« Reply #2 on: May 30, 2012, 04:05:36 PM »
Actually, since it's all going to be tslot, i was thinking of using the V-bearings that you can get specifically for T-slot. High-end stuff.

f00bard

  • FamiLAB Member
  • Newbie
  • *
  • Posts: 0
  • Da Real MVP
    • View Profile
Re: MendelMax - The Big One
« Reply #3 on: May 31, 2012, 09:24:28 AM »
Would Makerslide be appropriate here?  I see they mention that on the Mendelmax site.

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
Re: MendelMax - The Big One
« Reply #4 on: June 02, 2012, 03:32:45 PM »
OK.

So, after chatting with ThantiK, f00bard, and doing a bunch more research, I've decided that the Y axis will best be run on a ball screw type linear actuator, as a belt just won't provide the rigidity and torque that I can get in a small servomotor.

Calculated that the bed, with carriage, Al, and glass tops, will weigh roughly 37lbs, including a couple extra for whatever linear slide mechanism I decide to use. That is a lot of weight to sling around.

A ball screw will magnify the torque of a servomotor enough that it can be easily used on the machine. This will keep the size of the motor down, and make things more manageable. It will also look way cool. I believe a single ball screw with some sort of slide system will be enough to make the axis work. Not sure if MakerSlide should be used here, or just some large linear slides (LM12UUs), but it will need to be able to handle a lot of weight. I'm leaning towards a fully-supported system like makerslide (it's already aluminum extrusion) or a fully-supported linear bearing setup that I can mount on the frame.

After a chat with ThantiK about worries of structural rigidity, I've decided to go with 4040 instead of the usual 2020 extrusion, to make things beefier. This will need more design modifications, as I'd like to make the printed angles at the top beefier, and I will also need to modify most of the other parts to make things fit. I want to print the components out of black ABS, as I will be getting black anodized extrusion.

I've updated my BOM, and added in a few things that I need to purchase along with the extrusion (blind bolts, t-slot nuts), and the current cost for the frame and a couple other little things is approx $416.

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
Re: MendelMax - The Big One
« Reply #5 on: June 02, 2012, 07:28:19 PM »
Just ordered a GeckoDrive G320X servo drive. Now to get a motor an experiment with how this whole thing works...

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
Re: MendelMax - The Big One
« Reply #6 on: June 05, 2012, 04:38:02 PM »
Just for giggles, someone in #familab asked how much a solid block of ABS would weigh, 1 meter square (the build envelope of this machine)

We, being nerds, decided to calculate.

Mean mass of ABS = 1.07 g/cm3 (http://www.wolframalpha.com/input/?i=Acrylonitrile-butadiene-styrene)
Convert to m3 = 2359lb/m3 (http://www.wolframalpha.com/input/?i=convert+1.07+grams+per+cubic+centimeter+to+pounds+per+cubic+meter)

I don't think my servomotors will be able to move 1 ton.

Thus, anyone operating this machine: Do not build a 1m3 cube with solid infill.


P.S. Cost of 2359lb of ABS from MakerBot: $51,408 - 1,071 1kg spools

Chorca

  • FamiLAB Member
  • Sr. Member
  • *
  • Posts: 272
  • My primary function is failure.
    • View Profile
    • Flickr
Re: MendelMax - The Big One
« Reply #7 on: April 25, 2015, 01:02:46 AM »
Oh, this. This is done. Long-done.
http://imgur.com/a/t6Yto

It's a piece of crap cause nothing works well. I'm making a bigger one.