Jason Moyer published a new build:

Read more about this build...

 

My build plate will be 14" diameter. the ball screws are 16mm OD so for this application they are huge and overkill. the columns are 1.5" PVC and the only reason for the guides is to take out any deflection in the screws in the center. This whole printer is around 38" tall or so.

 

Yea i know i went a bit overkill for the size of the machine. i have it about 2/3 built now i just haven't posted the pics. i am waiting for the ball screws to come back from the machine shop before i can put the upper frame together. i built it heavy because my biggest complaint with the Makerbot Replicator 2x that I have access too is just built too flimsy. I've been a machine designer for the past 10 yrs or so and i can and have run machine tools myself. if you are trying to calibrate a machine to run a layer pass of .1 or 2mm (appx .010") how can you expect that kind of accuracy from a machine that will have the carriage flex or the bed flex over .030" just by looking at it wrong?! the concept of that just seems absurd to me. The ball screws are 16mm and will be plenty rigid for what im doing but i have the column pucks on a single bearing to help with screw flex as its trying to push the end effector around the build area.

If its too slow i will change the drive configuration and do a 1:2 belt drive on the screws. thats the beauty of ball screws. no backlash and will take the higher speeds. I've been told over on the RepRap forum that my electronics will be borderline for controlling this machine. So i'm pretty much planning that as the first upgrade after I get it up and running. Then I will re-purpose the Arduino/Ramps setup for something else. maybe another printer...maybe a scanner...not sure yet.

Domes are no joke. I've built dome style fire-pits using my cnc plasma table and they are fun to figure out that's for sure.

 

Not as of yet. i have not completed the build. I want to make sure that it works properly before i release any files. I am hoping to have my ball screws sometime before the weekend so i can start getting some real motion out of it.

 

no... im using 200oz/in NEMA 23's for the screws and a Nema 17 for the extruder

 

thanks. eventually i want to go to a higher amp stepper but for now i am going to try and keep the speeds down to try and minimize the amp spike. im not sure if i can bypass the on-board steppers with the duet board im using now.

 

not necessarily. ball screws will work in any orientation. the benefit is the zero backlash which is what you need for this type of printer. the weight of the head and arms will keep tension in one direction but switching it will induce backlash. using acme or std thread could cause the print to not be correct. they are also made for heavy loads...which i wont even come close to loading up the screws to even 1/4 their capacity.

 

thats interesting...which manufacturer were you looking at?

 

That's a great video from Thompson (one of the largest names in linear motion and bearings in general) but after watching that video he doesn't say anything about positional application of screws. yea he shows the difference between different styles of screws and the holding force of different pitches but that doesn't mean that it is not good for a vertical application. The unspoken meaning there is that there is a compromise between the load vs speed. A high load on a screw meant for high speed means that the holding torque of your motor and the resting position of your assembly become more important though out your design. You would need a motor that either has a built in brake or has the enough holding torque to keep your assembly where its positioned. Also means that your un-energized position needs to be at the end of the travel where the moving part is sitting on something solid and not hanging in the air.

In this case the entire assembly is light. I am using pre-loaded screws and they don't drift like the high helix one in the video. My challenge to overcome will be how fast i need the screws to spin to get the desired print speed. the nice thing about the ball screws is they will take the higher speed with no issue. I might need to add a 1:2 ratio belt drive to double the speeds of my screws, but until i get them and try out the printer i wont really know. The machine shop has had them for 3 weeks now. I was hoping them to be done in 1 unfortunately, but I can't complain too badly as its not costing me anything (owner is doing it as a personal favor).

I have the rest of the machine built and ready to go. I started doing some testing with using Pronterface and the Duet board to spin the motors as well as turn the hot end off and on and run the extruder to make sure the thermistors were reading correctly. I have to replace the thermistor on my heated bed. I think it may have been bad out of the box...new one should be here today. I know I have been slow to upload this build but I am working though some design iterations and don't want to clutter it up too badly.

 

It all has to do with the screw pitch. A finer screw and it wont do that. My plasmacam cnc plasma cutter has a screw to run the Z-axis and it doesn't move on its own when its not powered (granted there isnt much weight there). The exposed bearings are still a sealed system. that just means they have a ball return tube on the outside of the bearing nut. I believe that this style typically has a higher load capacity. they can use a larger ball because they dont have to account for a return passage within the body of the nut itself.

The angle of the screw pitch really is the key to it all. its the difference between rolling a car down a 1 deg incline vs down a 20 deg incline. with the belt drive system there is nothing to help the mechanical advantage of the motor and its 100% up to the motor to keep the carriage in its position. and most motors when turned off will free wheel. I am not sure how the belt driven delta bots stay up when powered down...or if they do at all. I cant imagine they just crash down but its possible that due to the friction in the slides they will drift down to the build plate.

 

yea that would be my concern.... mine has a glass build plate. i know my mechanism wont walk but I've thought if the screws are too problematic i would re-purpose them and go to belts but the drift wasn't something i thought about until now. you would need some sort of brake that would be locked when the power is off. or manually drop it to the deck when done. last thing i would want to do is shatter a $60 piece of tempered glass ?@!

 

well i have the electronics hooked up and had the motors turning... number of issues... first and the most damning to this project is the screws are not straight. they are bowed considerably which after removing the bearings that ride up and down the column its very apparent that I need to remove them and try to straighten them out. this may be the part that damns the ball screw idea...

I also am realizing that should have used a regular bearing on the bottom of the screws instead of a thrust bearing... I re-wired my 6 wire motors to half coil to help with torque issues but the duet board is still not able to feed the Nema 23's enough current to reliable turn them. they will spin for awhile and any little resistance they stall and sound really angry LOL.

I am not sure the rod ends on the delta rods were such a good idea. even after adding some spacers I am not sure they have enough movement in the ball joint to allow full movement of the delta configuration. I m working on trying to flash the arduino mega that I have with marlin or repetier since it has removable stepper drivers and maybe go that route for the electronics.

if i ditch the ball screw idea i might as well ditch the pvc columns and go to a better guide setup as well. ive been playing around in CAD the past 2 days and have a frame that will work with the makerslide extrusion and their v-guided wheels. However, when i start to look at the costs to rebuild this delta differently for belts and nema 17's, I am coming close to what it would cost to build the XY style printer I have been working on in CAD... so I am not sure what I want to do now...

 

cheap Chinese screws.... im guessing they came bent.

 

part of my issue is im working away from home most of the time so i dont have access to my shop. i can straighten the screws once i get home again... but the motor issue is a bigger fish.