26 (edited by pirvan 2015-01-04 08:18:02)

Re: Alternate belt tensioning method

Don't forget the model is accurate "within itself".  By that I mean the parts have either no clearance or minimal clearance between them. 

When you print them, they are not exactly the same size as the 3D model, they are slightly bigger because of the thickness of the extrusion, so they may require minor filing or cleaning to get them to fit just right

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

27 (edited by mdrVB6 2015-01-05 14:35:08)

Re: Alternate belt tensioning method

Yup, the tolerances are nice and tight.  I printed them in ABS yesterday and they turned out well.  If you have the E3D 0.25mm nozzle, I would recommend using that to print the clamps to get the max resolution out of the teeth grips.  The 0.4 nozzle is a little wide and you don't get the teeth quite right.  I've got some more nylon on order and once I dye it I will try a set in that too and use which ever seems to have a better grip on the rods.  I've been meaning to convert to direct Y-drive too so I'll get moving on that once the parts come in.

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

28

Re: Alternate belt tensioning method

pirvan wrote:

,
http://www.soliforum.com/misc.php?action=pun_attachment&item=7218

That is a great design! Any chance you could also please share the extruder, hotend, and fan carriage stl files?

29

Re: Alternate belt tensioning method

I had already posted the extruder a while back.  You can find the info and files here:

http://www.soliforum.com/topic/7038/sd3 … placement/ 

The fan duct is something new, and is for the standard hot end, it won't work on the E3D.  I'll post the fan duct STL later this evening.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

30

Re: Alternate belt tensioning method

Here is my take on the cooling fan duct.  It's small, detachable and quite efficient.

http://www.soliforum.com/misc.php?action=pun_attachment&item=7305

It is designed around a 30x30mm fan, and it attaches to the front of the extruder by means of magnets embedded on the mount.  I used 6mm neodymium magnets, which are very strong, yet if the duct were to hit something , it will just come off without doing any damage to the print or the machine.

The duct is was sliced in two parts so that it can be printed flat.  Simply glue the 2 parts together.  The magnet mount offsets the duct from the hot end, and provides a good gluing surface for the duct.  The duct can be rotated about 35-45° in either direction before gluing.  Just make sure that you don't have anything in the way (like the heater wiring).

One note:  I usually use black anodized steel screws, but the last batch of screws I bought were stainless steel.  Depending on the type of stainless steel used, it might not be magnetic, so you need to make sure the screws you use on the extruder are magnetic or the whole concept fails

Post's attachments

Extruder & Fan Duct.jpg
Extruder & Fan Duct.jpg 148.81 kb, file has never been downloaded. 

fan duct (layout).stl 3.43 mb, 17 downloads since 2015-01-09 

You don't have the permssions to download the attachments of this post.
To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

31

Re: Alternate belt tensioning method

Thanks @Privan

32 (edited by pirvan 2015-01-21 21:03:13)

Re: Alternate belt tensioning method

Finally Finished.

Yesterday I finally got my GT2 sprockets, and I assembled everything and ran a couple of tests.  I was able to tighten the belts pretty well, so I didn't need the tensioning springs I bought.  After adjusting the steps in the configuration file and saving them in the EEPROM, I ran my calibration piece and it looks nearly perfect.

I also seem to have gotten rid of the Z ribbing, although I need more testing on that.  When I was replacing the Y rod bushings with the flanged bearings, I noticed that the right side bushing was really loose and wobbly, so that may be what was causing the ribbing.  We'll see after I run a larger print.

So, without much ado, here is the final product...


http://www.soliforum.com/misc.php?action=pun_attachment&item=7496


...and here is what the whole printer looks like.


http://www.soliforum.com/misc.php?action=pun_attachment&item=7497

Mods List:

  • RAMPS 1.4 w/ DRV8825  Extruder stepper driver (1/32).

  • 750W Server PSU w/ front, lighted Power Switch.

  • Graphic LCD controller.

  • Ball Screw and nut for Z drive.

  • Aluminum Print platform w/ full Copper Print bed and MK2A Heater.

  • GT2 belts, sprockets and idler pulleys.

  • Chrome hardened 8mm shafts.

  • Chrome hardened 6mm "Y" drive shaft w/ flanged bearings and Pillow Block.

  • Side opening wire Chain carrier.

  • Custom "X" carriages w/ linear bearings.

  • Custom extruder carriage w/ linear bearings.

  • Custom "Y" motor belt tensioner.

  • Larger Extruder motor (Kysan 1124090).

  • 40W Heater cartridge for Hot end.

  • G-Code controlled Cooling fan & duct

Post's attachments

SD3_GT2_001.jpg
SD3_GT2_001.jpg 433.32 kb, file has never been downloaded. 

SD3_GT2_002.jpg 257.76 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.
To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

33

Re: Alternate belt tensioning method

Very nice work!  Hopefully my GT2 hardware gets here this weekend and I can begin. 

PS: check your PM's, I need a CAD favor and I'm willing to pay

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

34

Re: Alternate belt tensioning method

mdrVB6 wrote:

Very nice work!  Hopefully my GT2 hardware gets here this weekend and I can begin. 

PS: check your PM's, I need a CAD favor and I'm willing to pay

I saw it late last night.  I'll take a look at it later this evening.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

35

Re: Alternate belt tensioning method

Very nice, thank you.
A question: when your extruder goes all the way to the front, does the fan "poke out" of the case? By that imagine if the front of the frame was covered by plexiglass.
Something Solidoodle 4 owners should be aware of.

Solidoodle 4

36 (edited by jagowilson 2015-01-22 13:27:01)

Re: Alternate belt tensioning method

pirvan wrote:

I also seem to have gotten rid of the Z ribbing, although I need more testing on that.  When I was replacing the Y rod bushings with the flanged bearings, I noticed that the right side bushing was really loose and wobbly, so that may be what was causing the ribbing.  We'll see after I run a larger print.

How would a wobble in the Y axis cause Z ribbing? However this does explain why you witnessed such slop in the Y stepper belt on your machine and it is no doubt the reason you have had such a bad time with circles. Those bushings are mechanically sloppy as hell even if they are tight.

37 (edited by pirvan 2015-01-22 15:26:16)

Re: Alternate belt tensioning method

redbarret wrote:

Very nice, thank you.
A question: when your extruder goes all the way to the front, does the fan "poke out" of the case? By that imagine if the front of the frame was covered by plexiglass.
Something Solidoodle 4 owners should be aware of.

Yes it does, by about 10mm.  I have an open frame, so I don't really care about that.  Besides If I ever were to cover the front, I could always restrict the movement, or simply remove the fan.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

38 (edited by pirvan 2015-01-22 15:30:14)

Re: Alternate belt tensioning method

jagowilson wrote:

How would a wobble in the Y axis cause Z ribbing? However this does explain why you witnessed such slop in the Y stepper belt on your machine and it is no doubt the reason you have had such a bad time with circles. Those bushings are mechanically sloppy as hell even if they are tight.

Well... it didn't.   I did a larger print with smooth vertical surface, and while it does look better, the ribbing is still there.  I'm really at a loss on how to get rid of that.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

39

Re: Alternate belt tensioning method

Don't wanna send this thread too far off topic but have you tried an enclosure? The IR images of the mk2 show it's barely ok near the edges and maybe an enclosure will help keep the heat in. Given that your system seems pretty rigid I can only assume it's a temperature issue.

40

Re: Alternate belt tensioning method

jagowilson wrote:

Don't wanna send this thread too far off topic but have you tried an enclosure? The IR images of the mk2 show it's barely ok near the edges and maybe an enclosure will help keep the heat in. Given that your system seems pretty rigid I can only assume it's a temperature issue.

I don't think it's temperature related. The bed itself is not an issue, it only needs to be hot enough to hold the piece in place.  Nor do I have any large fluctuations in extruder temperature, my temp curve looks pretty flat throughout the print job.

Then there's the fact the error is extremely periodic, there's exactly 4mm between the "rib peaks".  That corresponds to the pitch of my ball screw.

I've been going back to check some of my early prints, and they didn't have this.  Best I can narrow it down to is during the time frame when I made the change to the RAMPS board.  I wonder if that has anything to do with it. 

Don't know, maybe some day the solution will present itself.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

41

Re: Alternate belt tensioning method

pirvan wrote:
jagowilson wrote:

Don't wanna send this thread too far off topic but have you tried an enclosure? The IR images of the mk2 show it's barely ok near the edges and maybe an enclosure will help keep the heat in. Given that your system seems pretty rigid I can only assume it's a temperature issue.

I don't think it's temperature related. The bed itself is not an issue, it only needs to be hot enough to hold the piece in place.  Nor do I have any large fluctuations in extruder temperature, my temp curve looks pretty flat throughout the print job.

Then there's the fact the error is extremely periodic, there's exactly 4mm between the "rib peaks".  That corresponds to the pitch of my ball screw.

I've been going back to check some of my early prints, and they didn't have this.  Best I can narrow it down to is during the time frame when I made the change to the RAMPS board.  I wonder if that has anything to do with it. 

Don't know, maybe some day the solution will present itself.

Did this ribbing by any chance coincide with the installation of the MK2 and are you using PID control for it?  Reason I ask is my PCB draws about 20A and I had to go to a mechanical relay (with BangBang enabled) temporarily while waiting for my SSR. During that time, due to the slow cycling of the PCB Heater, what I thought was Z-ribbing suddenly appeared.  Short story is don't discount thermal issues as even a few degrees +/- cycling of the heat bed can cause pretty noticeable banding.

SD2 - Stock - Enclosure - Heated Bed - Glass Plate - Auto Fire Extinguisher
Ord Bot Hadron - RAMPS 1.4 - Bulldog XL - E3D v6 - 10" x 10" PCB Heated Build w/SSR - Glass Plate
Thanks for All of Your Help!

42

Re: Alternate belt tensioning method

pirvan wrote:

Besides If I ever were to cover the front, I could always restrict the movement, or simply remove the fan.

Since I'm using the stock SD4 extruder, I think I'll just edit the mesh to be screwed on the bottom screw holes of the left side of the extruder aluminum holder, where the side fan currently is. That won't get in the way of anything on stock SD4.

Solidoodle 4

43

Re: Alternate belt tensioning method

IronMan wrote:
pirvan wrote:
jagowilson wrote:

Don't wanna send this thread too far off topic but have you tried an enclosure? The IR images of the mk2 show it's barely ok near the edges and maybe an enclosure will help keep the heat in. Given that your system seems pretty rigid I can only assume it's a temperature issue.

I don't think it's temperature related. The bed itself is not an issue, it only needs to be hot enough to hold the piece in place.  Nor do I have any large fluctuations in extruder temperature, my temp curve looks pretty flat throughout the print job.

Then there's the fact the error is extremely periodic, there's exactly 4mm between the "rib peaks".  That corresponds to the pitch of my ball screw.

I've been going back to check some of my early prints, and they didn't have this.  Best I can narrow it down to is during the time frame when I made the change to the RAMPS board.  I wonder if that has anything to do with it. 

Don't know, maybe some day the solution will present itself.

Did this ribbing by any chance coincide with the installation of the MK2 and are you using PID control for it?  Reason I ask is my PCB draws about 20A and I had to go to a mechanical relay (with BangBang enabled) temporarily while waiting for my SSR. During that time, due to the slow cycling of the PCB Heater, what I thought was Z-ribbing suddenly appeared.  Short story is don't discount thermal issues as even a few degrees +/- cycling of the heat bed can cause pretty noticeable banding.

I see where you're going with this, you think the MK2 current draw is affecting the current output to the extruder.

I don't know if the MK2 installation coincided with the ribbing.  I do know I have PID enabled on it, and the bed temperature doesn't deviate by more than +/- 0.5° C.  I also looked at the current output curve and it's steady.  I use a 750W PSU and all the current is going directly through the RAMPS board and out to the extruder and bed.  I'm not using any relays to power the bed directly.

I'm going to go through the firmware configuration tonight to se if there's some discrepancy there that I didn't account for.  I'm also wondering if the DRV8825 driver may be at fault.  I'll replace it with a standard 4988 Palolu and see if the problem goes away.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

44

Re: Alternate belt tensioning method

So earlier today and last night I tore apart my printer to do the first major rebuild since I first installed the Lawsy carriages.  I went direct drive externally mounted motor on the Y (accousedmonk's piece), dual pillow blocks and new 6mm Y drive rod, all GT2 belts and these new carriages.  I kept the Lawsy center carriage and still using the E3D V6 mount I made.  One picture here, then see imgur album for several large photos:

http://i.imgur.com/gEDAYOO.png

http://imgur.com/a/VRoPE

A couple comments:

  • I also used your (Pirvan's) adjustable Y belt idlers that you posted in my other thread.  These were awesome!  The carriages seem to hold the belts very tightly but I still felt I need just a little more.  With just a couple turns of the wrench, I had great belt tension.  It was easy enough to knock out the PEM nut, just take your scraper and tap down on the front of the nut and it pops right off.  Then just hold the part of there as a template for where to drill the adjacent hole.

  • Nylon was an excellent choice for the belt clips.  Its slightly flexible and it helped to not have a brittle ABS piece that would crack easier.  But ABS will also works, but if you've got nylon then go for it.

  • The bearings are VERY tight in the holes.  I printed them at 0.3 layers in nylon.  I actually had to tap them in with a hammer.  Nylon does not sand as well as ABS.  However, the nylon is less prone to cracking with I shoved the nuts into the traps.  I was not able to get a good alignment between the two bearings and when I had two bearings in there I broke two of them by grinding and several balls fell out.  So I put a single bearing in the middle and there was no alignment issues and it still works great.  So if two bearings do not glide smooth, just use one.  Or get a better print and/or sand the carriages to get a good alignment.  It is less forgiving than the Lawsy carriages, which leave slight wiggle room between bearings for them to align on the rods. 

  • I wanted to use bearings on all the idlers but the ones I got were too big on the outer diameter an hit the walls, so I'll look for some smaller ones.

After installing and changing steps per mm, I threw on a calibration cube and it came out good!  I just also did a vase with 80 mm/sec perimeters and it seems to have no problems going that fast.  All the belts are nice and tight.  Overall, I think this is the best mod since the Lawsy carriages.  Thanks Pirvan!

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

45 (edited by mdrVB6 2015-04-20 14:52:22)

Re: Alternate belt tensioning method

Bumping this for a detailed review since I have now been using these for a couple months and have them on 2 printers now.  Additional comments:

  • You do lose about 10mm of build space on the X motor side if you use these Y carriages with the lawsy X carriage.  The X carriage hits the X motor mount before the nozzle gets to the edge of the build platform.  The only way to fix this would be to move the motor mount further to the outside, which would present other (big) issues.  Just be advised if you need every bit of the build space.

  • While the Lawsy carriages have the bearings "clip" in, these carriages are snugged into place.  This makes it very hard to align 2 LM8UU bearings.  If you don't get it perfect, they will be noisy or bind, which causes the balls to fall out and ruin your bearings.  TO solve this, I just used one bearing in each carriage.  If you get it just right, you can still have a very rigid axis like this.  However, I have ordered some LM8LUU bearings, which are twice as long.  I'm going to try this and will report back.  You must be very careful to not oversand the bearing holes or the bearing will be loose, and that's  reprint because you can't have that.

  • Additionally, having the bearings fully covered all the way around seems to put pressure on it and make it run louder.  My Y axis, with only 2 bearings, is louder than my X axis (Lawsy) that runs with 4 bearings that clip in and are not fully enclosed.  IMO, these carriages could be improved if the bearing mount was more like the Lawsy carriages, but doing that would make them bigger.

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

46

Re: Alternate belt tensioning method

Interesting.

I doubt many users here used these anyway, but here are few "rebuttal" comments on mdrVB6 review.

1.  Loosing 10mm of build space.  That's interesting.  I don't use Lawsy's printhead (X) carriage, so I don't have this problem.  Is the 10mm lost from both sides (5mm + 5mm), or all on one side? Which?

2.  I did use Lawsy's "Y" carriages at one time, and I simply wedged some pieces of folded paper (from a business card) under the bearings to prevent play. 

On these, it's true they are tight, but the idea is to keep them tight.  To keep the 2 bearings aligned, I put them on a rod, then pushed them together into carriage.  Your idea of using a single LM8LUU long bearing works too.

Alternately, we could print the hole slightly lose, then use a set screw that aligns with the groove in the bearing.  That would probably be the better design, as it would also make installation and removal of the motor easier, since you could easily remove the bearings.

3.  I personally never noticed the bearing noise, probably because I have nothing to compare it too, other than the "X" carriage noise, and they both seem to be about the same.  Once it start printing, I think I hear the motor itself more than the carriages sliding on the rods.

Anyway, I've had these installed on my printer since the day I posted this, about 3 months now, and they seem to be working pretty good.  The belt tension is also good, and doesn't appear to have slackened any.

I've been toying around with replacing the standard hot end with the all metal one I got.  I even have a complete extruder assembly (carriage w/ bearings, hot-end, motor, fans, etc) ready to drop in place, but you know the old adage:  If it ain't broke...

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

47

Re: Alternate belt tensioning method

1.  The lost print space is on the X motor side of the carriage.  I could get a picture later tonight if you were really interested.
2.  I really like the set screw idea.  Make the holes just big enough to slide in a long LM8LUU bearing without pushing it to the point of making it too tight on the rod and then secure it with 2 set screws (with nut traps).  That would be awesome.
3.  Agreed that the bearing noise is not a big issue, but this would be addressed with the set screw mod above.

All in all, these carriages are a great design.  Thanks again, Pirvan!

SD4 w/ RUMBA, E3D Volcano, all bearings, glass bed

48

Re: Alternate belt tensioning method

Have you ever tried to create an x carriage(Left & Right) out of lawsy's setup along with the sd4 setup and your's combined? I like the design, I'm kinda leary about stuffing an lmuu bearing into a print and making it stay there. I like your design, just afraid of the bearings popping out of printed carriage.

SD3/SD4 Mods completed: glass bed, changed zrod to 5mm, slop nut, aluminum arm-originally wood, plexi-glass case, Z Wobble Preventer,  Lawsy Carriages, X\Y motor fans, control board fans, extruder cartridge heater, MK5, Esd Lite
Mods working on:Direct Y Drive
Remember.......All you touch and all you see is all your life will ever be....Pink Floyd...Dark Side of the Moon.

49

Re: Alternate belt tensioning method

blarz wrote:

Have you ever tried to create an x carriage(Left & Right) out of lawsy's setup along with the sd4 setup and your's combined? I like the design, I'm kinda leary about stuffing an lmuu bearing into a print and making it stay there. I like your design, just afraid of the bearings popping out of printed carriage.

If that is the case, just apply a glue to the bearing housing before you put into the carriage. Most of the designs I have seen however require some force to get the bearing in and once it is in, it will go nowhere. You would want to have to really remove it to get it out.

Printing since 2009 and still love it!
Anycubic 4MAX best $225 ever invested.
Voxelabs Proxima SLA. 6 inch 2k Mono LCD.
Anycubic Predator, massive Delta machine. 450 x 370 print envelope.

50

Re: Alternate belt tensioning method

Geez... Someone was actually looking at these... The design is nearly 3 years old. 

I haven't used my SD3 for the least year or so, but I can tell you that the bearings never "popped out".

As for Lawsy's carriages, they are great as well, and another alternative.  But when I set out to do the carriages and the extruder, I wanted a unique design of my own, so I didn't mix the two. 

There were other changes that were made after this, like a front print bed stabilizer, an e3D style extruder and others.

Alas, it all sits here on my desk unused, ever since I got into SLA printing I haven't found the need for FDM printing

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.