3D Printer Build

[Peter Randerson]

Bob
I am following this with interest as I have built my own Cartesian style printer using 20/20 extrusions and am now onto my second build of a Core XY style machine, again using 20/20 V slot.

Not sure if you are aware but one of the many problem areas on these things is poor eccentricity on the stepper pulleys. Biggest offenders are the cheap items out of China.

Genuine Gates belts and pulleys are the best option. You can read about them here:
https://e3d-online.com/blog/2018/11/29/gates-belts-pulleys-and-idlers-now-available/

Cheers
Peter

 

[bmac2]

Peter I have to agree with you. Ordering the cheap stuff from China (like everything in this build) can be a crapshoot. I’ve had good luck with most of the stuff I’ve gotten but a couple of items have been absolute garbage. I don’t know if it’s a lack of quality control or someone passing off factory seconds and rejects. I’ve never had a problem getting a refund or replacement but with one month lead times it’s a pain.

 

[bmac2]

So I finally came up with an idea for the X belt tensioner by mounting it on the back on the X carriage.

I dug out a cut off piece of the 3/8” flat bar with the rounded edge I used for the Z nut holders. Cleaned up a couple of other bits from the scrap bin for its base and the belt clamp.

Brought the rounded piece down to size then mill an 1/8” slot in it and spot milled a shallow hole in the flat end.

Chucked up some .625 brass in the lathe, drilled it clearance size for #4-40 and parted off a thin washer using the good old bamboo meat skewer to catch it.

Milled the base to shape, drilled and tapped the holes for the jack screw, the guide screw, and the two 3mm screws that will attach it to the carriage plate.

After I got it assembled I clamped it into place to check the fit and . . . . . it was not going to work. I don’t have enough space between the bearing blocks for the belts.

The solution that I came up with was to mount it on the other side of the plate and to capture the belt on the lower end. This meant having to mill a small bite out of the lower bearing mount to allow the tensioner to move its fell travel but I can’t see it having any effect on the bearings. I may have gotten a little overzealous with my “small bite” but again I can get just over 3/8 of an inch of travel out of it.

The way it works is the bit with the rounded end slides on the 4-40 screw when the jack screw is tightened. It can’t rotate on the screw because the jack screw sits in that shallow hole in the end. When the tension is set tightening the 4-40 screw with the washer locks everything in place.

 

[XD351]

So I finally came up with an idea for the X belt tensioner by mounting it on the back on the X carriage.

View attachment 107517

I dug out a cut off piece of the 3/8” flat bar with the rounded edge I used for the Z nut holders. Cleaned up a couple of other bits from the scrap bin for its base and the belt clamp.

View attachment 107518

Brought the rounded piece down to size then mill an 1/8” slot in it and spot milled a shallow hole in the flat end.

View attachment 107519

View attachment 107520

Chucked up some .625 brass in the lathe, drilled it clearance size for #4-40 and parted off a thin washer using the good old bamboo meat skewer to catch it.

View attachment 107521

View attachment 107522

Milled the base to shape, drilled and tapped the holes for the jack screw, the guide screw, and the two 3mm screws that will attach it to the carriage plate.

View attachment 107523

After I got it assembled I clamped it into place to check the fit and . . . . . it was not going to work. I don’t have enough space between the bearing blocks for the belts.
View attachment 107524
The solution that I came up with was to mount it on the other side of the plate and to capture the belt on the lower end. This meant having to mill a small bite out of the lower bearing mount to allow the tensioner to move its fell travel but I can’t see it having any effect on the bearings. I may have gotten a little overzealous with my “small bite” but again I can get just over 3/8 of an inch of travel out of it.

The way it works is the bit with the rounded end slides on the 4-40 screw when the jack screw is tightened. It can’t rotate on the screw because the jack screw sits in that shallow hole in the end. When the tension is set tightening the 4-40 screw with the washer locks everything in place.

View attachment 107525
View attachment 107526

 

[XD351]

You could also try ptparts for 2.5 pitch pulleys .
https://www.ptparts.com.au/products...ming-pulleys/12-T2.5-6--t2-5-6-timing-pulleys
These pulleys are only centre drilled each side so you can fit them to what shaft size you like by boring the hole yourself .

I had to adjust the amount of steps on my x&y axis as i was printing 0.5 mm undersize over 25 mm ( half of that is probably shrinkage ) So i set up a dial gauge on each axis and used the controller to jog 0.1 , 1 and 10 mm and noticed some inconsistencies , not huge ones but if i jog 1 mm then another and so on one will read spot on the next over by 0.1 mm and the next under by the same then is will go back to being spot on with the next jog so i would say the pulley isn’t the best . Strange thing is if i do the same using a 10mm jog it is more consistent usually within 0.025 or so .

 

[bmac2]

The loose end of the belt will just clamp down on the top of the bearing block with the hardly noticeable little notch I milled into it.

I started by trimming a piece of 1/8 inch flat bar to size and setting it up in the mill to cut the slots.

The closest cutter I have to match the belt teeth is 1/32, not perfect but it will work. Set the speed on the mill to “15 year old on his second Red Bull” and slowly cut out the slots.
With slow feeds and lots of WD40 I got them done without hearing that nasty “Tink” sound that small end mills make when they snap.

As luck would have it a good old 3/8” end mill worked out to be the perfect width for cutting the hole for the idler pulley.

With everything back together I was able to tension the belt to the point that I started to worry about the 5mm shaft on the stepper and could see no appreciable deflection on the 8mm rods using a dial indicator.

 

[bmac2]

I spent the past couple of days making the Y axis carriage (the Frog). In retrospect I can’t think of why I made it to match the Prusa I3 when It would have been much simpler to just bore some big holes in it like Myfordboy did to reduce the weight but done is done.

The only aluminum plate I have is .375, .25 and .1875 so 3/16” it will be. The carriage is too big to layout with the DRO on the mill and I can’t see where anything other than the holes for the bearings is all that critical so I just printed it out full size and taped it to the plate.

If you’ve never used an optical centre punch they’re pretty cool. I don’t use it as much as I did before I installed the DRO’s on the mill but for something like this it’s perfect.

The base has an “O ring on the bottom that keeps it from sliding around and the optical rod has a magnifier polished into the top and a “Cross” in the bottom.

Once you have the cross on the layout mark you hold the base, slide the optical rod out and replace it with the punch and give it a light tap. The optics are much better than in the picture, I guess the camera just couldn’t’ decide where to focus but you can’t get much closer than that.

Once I had it marked out I took off the paper, gave it a once over with an automatic centre punch and marked them for the appropriate hole size then drilled it out.

 

[bmac2]

I hate having to use sketchy setups. Whenever I mill on a piece of wood it just never feels like it’s clamped down as tight as it should be. But after many setup changes and a bit of cleanup and deburring it was done. Somehow I got the one hole way off centre but I’m just using it to anchor the wiring for the heated bed so it’s ok.

 

[bmac2]

I got the bearings mounted and sitting on the bench the carriage slides smoothly without the rods wobbling around.

This is where I once again my plan to cut the frame down came back to haunt me. Cutting it down would only shorten it by an inch and a half but at this point I think it’s going to be easier to tweak the rod mounts than disassemble the frame, cut it down and get everything lined up again. And I just might need that extra inch and a half someday.

 

[bmac2]

Originally the mounts for the Y rods where going to basically be an upside down “T”. The new mounts have a bit more style have a rounded top and step back to catch the rod behind the 2020 V slot.

Started on the rod supports by cutting down some 1.25” aluminum bar, squared it up and milled the basic L shape of the mount.

I left a small shelf at the bottom and cleaned it out with a 3/8” round nose bit to help maintain my “style” points.

Brought the height down and milled out the “tombstones” then drilled the 5mm screw holes.

After cutting them apart on the bandsaw and cleaning them up I indicated them in the mill to drilled and reamed the 8mm holes for the rods. The only critical dimension on these it the distance from the center of the hole to the bottom of the mount.

A quick test fit and I’m happy with the way they turned out. The rods feel ridged even without the locking screws. I’ll finish them off by rounding the top and back on the belt sander and drill and tap the ends of the rods for m5 screws.

 

[bmac2]

For the idler mount I started with a piece of 1x1 bar and milled out the front keeping with my “rounded” theme. Flipped it over and milled a notch in the back for the overhang then milled down the sides and drilled the hole for the axle before milling the slot in the top for the idler.

Mounted to the V slot it is solid and shouldn’t flex with any load I’ll be able to put on it. Cleaned it up and another part got checked off the list.

 

[bmac2]

Ok. Most of the mechanical is done on this thing. I only have the "Y" belt clamp and the build plate left but that involves going out to the garage to cut stuff on the bandsaw and it was just too damn cold out there today.
So instead of doing anything productive I hooked it up to a 2 axis (so no Z) controller I had around and just played with it.
The horrible grinding noise is one of the bearings on the X axis. I swapped it out and it’s much quieter now.

 

[ddmckee54]

Bob:

What kind of bearings are in the idlers? They are going get a lot of mileage put on them.

Don

 

[bmac2]

Hi Don. The pulleys I’m using are an aluminum body with 2 ball bearing races pressed in similar to the Openbuilds drawing below. I’d never really thought about it but ya over time they won’t ever see high RPM’s but they will be racking up the mileage. Luckily even good quality ones can be had for $6 or $7.00.

 

[bmac2]

Well it finally warmed up enough to get out to the garage and cut some metal. The stuff gets damn cold even with gloves on. I’ve found that when its colder out if I don’t let my saw run free for a few minutes to loosen up it likes to throws the blade.
I’m using 7/32” aluminum for the build plate and it was shear cut so I trimmed a bit off each side to relieve any stresses in the metal (see Myfordboy vid).

I cleaned up the edges and put it on the surface plate to check it. At first I thought it was warped but I discovered I’d managed to scratch a sizable burr on the bottom. After a light cleanup with a file it looks good.

I screwed the build plate solid to the carriage and with a dial indicator in the head mount it doesn’t look terrible. Now I just have to come up with some springs, from what I’ve read the stiffer the better?

 

[ddmckee54]

Bob:

The stiffer the better - within reason. You still want the springs to give if your print head crashes on the build plate. When you're homing the machine and the Z limit fails you don't want bad things to happen.

Don

 

[bmac2]

Bob:

The stiffer the better - within reason. You still want the springs to give if your print head crashes on the build plate. When you're homing the machine and the Z limit fails you don't want bad things to happen.

Don

Good point. I guess it could go from “bad things” to OMG! OMG! terrible things in a hurry with a glass build plate. We have a couple of shops locally that mostly just have machines and filament but I’m going to make some calls to see what else they have in the way of parts.
I have a decent stash of small/medium springs but this thing is getting to the point where it would be one less variable I’m going to have to deal with.

 

[bmac2]

For the Y axis belt clamp I started by sandwiching a piece of .125 and .25 together and trued them to size.

Set up in the mill I drilled the 6 holes #19 (close as I can get to m5 tap size) and then milled the square hole through the .125 piece.

With a little heat I got the two pieces separated and used a transfer punch to mark the location on the Y carriage and spot drilled it.

I milled a shallow recess in the .25 piece to match up with the square cut into the .125 plate. The middle two holes where then tapped m5 and the outer four drilled clearance size. On the .125 plate the outer four are tapped m5 and the middle two are drilled clearance. All of the holes in the carriage are clearance.

Once assembled the two middle holes mount the (.25) top plate to the carriage. The belt runs in the recess in the top plate and comes out through the square hole in the (.125) bottom plate. The other four screws pass through the carriage and top plate and are threaded into the bottom plate to clamp the belt in place.

I think I went a little overboard on what should have been a simple clamp but it works great. Just feed in the belt, pinch the two ends and tighten the screws.

 

[bmac2]

I woke this morning and while I was enjoying my coffee the weather man on TV was telling me that the wind chill was -40c.

Sounds like a good day to stay inside and start on the electronics.

Over the past few of days I have been reading up on the pros and cons of the various types of end stops. I have the components for the three most common, mechanical (micro switch), optical, and Hall Effect. From what I’ve read Hall Effect can be less susceptible to noise and false triggering, and I like playing with them so Hall Effect it is.

The sensors I have on hand are A3144. There a unipolar (only one pole will trigger it) and nonlatching. I going to use small neodymium magnets for the triggers and I’m thinking I should be able to get good repeatability out of them.

The circuit is pretty simple just the sensor, a resistor, a small LED for debugging (has to have LEDs), and some veroboard. Back when I use to do a lot of prototyping many of the circuits did nothing visually so we’d add LEDs to a data bus or clock and call them the “Managerial Circuit”. That way when the boss came around he’d have some blinking lights to look at. Amazing how that worked.

These behave like a normally open switch so go high when triggered.

The mounts are pretty simple and just attach to the V slot with m5 screws. I Installed Pronterface and so far it’s been really handy for testing and sending G-code to the controller.

For now the controller is just screwed to a handy sized piece of aluminum on the side of the frame. This thing is going to need some serious “cable management”. I use to hate having to do harnesses but these days with age and a good audio book it shouldn’t be that bad.

I’m running it on a current controlled bench power supply to help keep the magic smoke in. But everything is moving when it’s supposed to move, stopping when it’s supposed to stop, and hot things are going to a set temperature and staying there so it time to start the software configuration in Marlin.
This could take a while. Like a lot of cheap 3D printers this thing doesn’t come with customer support.

 

[ddmckee54]

Don't forget about lights for the build plate. They really help you see what's going on/wrong with the print.

I'm still trying to figure out how to get some decent light at the nozzle without the part cooling fan ducts getting in the way.

Don