I too tried the Twister Blocks and then took them out. I switched to the Reptar V2 "blocks" with GT2 belts and pulleys.
I too tried the Twister Blocks and then took them out. I switched to the Reptar V2 "blocks" with GT2 belts and pulleys.
Those of you who had problems with the twister blocks might leave a comment on the TV page. Chopmeister might be glad to give some advices how to print or to provide an improved version if necessary.
I feel a bit unsure right now as I was planning to use the twister blocks (I already printed them).
Thanks for the report guys!
I guess I'll try out both the twister and reptar blocks. The twisters just seem to make assembly a lot easier (I can't really say because I haven't printed any blocks yet) which would be a big plus. But optimal function is of course the most important thing.
I will be using 303 tooth GT2 belts, already bought everything.
I'll have to update the list in the first post, but I want to try out some things before I decide. Hopefully I'll get somewhere this week - Just bought black color for the frame and will start painting the panels today.
The E3D hotends are already somewhere in Switzerland - I hope to get them on Tuesday when I should have the frame ready.
/edit:
Done sanding the wooden parts. Starting to paint now...
I sanded the parts with 400 grain sandpaper only. I know that's not optimal, but it does a good enough job and I don't want to take off too much material.
I use water-based acrylic paint, applied with a paint roller. Doesn't smell bad, doesn't stick everywhere like spray paint. I'll do one or two thin layers, then coat it with two-component protective finish to make it strong, glossy and prevent it from warping with changing humidity.
I don't really know about the two-component finish and heat compatibility, but I guess it shouldn't be a problem as you don't usually heat up the chamber above 70°C.
Jonny,
Don't want to discourage you but I can pretty much guarantee the twisterblocks won't work unmodified with the 303 GT belts.
I sort of got them to fit be heating a knife and slicing off the "tab" of this part (shown towards the back of the photo):
https://s3-eu-west-1.amazonaws.com/files.youmagine.com/uploads/image/file/42380/export.png
Unfortunately, the belts then slip in the block so you need to had some thin friction material. I used dry wall sanding screen:
http://ecx.images-amazon.com/images/I/51KS4JXE2NL.jpg
cut to fit between the belt and block. It locked the belt up quite well. However, it was all a pain to assemble requiring a slip wrench to compress the tab to put the screws in.
IMO, not worth the trouble.
The reptar blocks might work. The only thing I don't like about them is they require non-metric or non-machine screws and standoffs to assemble. I really wish someone with more CAD skills would remix them to use the leftover hardware from the UM1. Seems kind of strange to have to go out and buy hardware and standoffs that could be replaced with printed parts and the metric machine hardware.
Good luck.
+1 to the twister blocks not playing nice with gt2 belts.
I love the reptars, they are by far the easiest and work beautifully with the 303Gt2 belts. Ordered the hardware for a few dollars.
Ok, so I'll try the Reptars first, and if I'm happy with them it'll be the Reptars
Afaik they were designed to be used with 303 tooth GT2 belts anyways.
Plastite screws aren't that bad actually. It's not easy to get the dimensions right for sinking hex nuts into the plastic. Melting them in is a bit brute, but works of course.
I've used plastite screws myself for "the big project" I'm working on atm. They work well, but you have to be very careful not to overtighten them.
Machine screws and hex nuts are worse. The nuts sink deeper and deeper into the soft plastic without ever being really firm.
Imho, the best way is to use machine screws, hex nuts AND large washers on both the screw and the nut side. This works really well for me and gives strong connections that you can tighten with 0.5Nm (torque tool) without worrying about the threads. And 0.5Nm isn't that much...
The only drawback is, it takes up quite a bit of space... So for small parts plastite screws are what I'd use. Buy a bag of them and have a supply for all your upcoming projects
/edit:
Actually I only wanted to give an update:
Second painting session done & drying now. At least two more (last corrections & protective finish) ahead, and then I'll start printing all the hacks on my first machine.
Hey there! I wasn't really looking at the forums for a while now, too busy at work and stuff... But, there should be no problem modifying the TwisterBlocks to work with the GT2 belts.
They were originally designed to work with stock belts which generally come a bit loose, especially over a longer period of time, so they provide crimping and tension via the same part.
All that is needed here is to change the little tab a bit and the rest of the blocks stay completely the same, so if you already printed them out, all you need is to print out those 4 small parts again.
I will do the design right now, I have some GT2 belts laying around somewhere, so expect an updated part sometime today. In the future please don't hesitate to contact me here, or on TV, or YM if you need help with some of my designs, I'm always glad to be of assistance if possible.
Sorry for the thread hijack Jonny!
When you redesign for GT2, you need to indicate what length you are supporting. 303 GT2 belts have a center-to-center distance of 283.00mm (for 20 teeth pulleys) vs the MXL 300 c-to-c distance of 284.48mm. That missing 1.48mm seems to make a big difference. 305 GT2 belts have a c-to-c distance of 285.00mm or .52mm MORE than the 300 MXL belts.
Also, for some reason, possibly just my machine or the way I printed them, there wasn't enough room in the left-right direction to fit the 6mm rod without binding the bushings on the front-to-back 8mm rods. The back to front distance between the blocks seemed fine with 1-2mm of play.
I used this for the distance calc: https://sdp-si.com/eStore/CenterDistanceDesigner
@chopmeister: Don't worry, that's just what I want if it helps with the UM black edition!
Imo, ideal blocks should be designed for 303 tooth GT2 belts as these seem to be just the right length. Additional tensioning might become necessary over time, but probably just by a very small amount...
On the other hand, it's convenient if you have blocks where you can just cut your belts yourself, clamp them together and tension them as needed.
The 6mm shaft length seems to be a critical part. These shafts are made with a pretty large length tolerance, somewhere around +-0.5mm, maybe even more. It doesn't matter if the shafts are slightly too short in the end - axial play on these shafts doesn't generate any problems. Maybe just give a little more space for the shafts.
Maybe the x blocks have to be different from the y blocks?
The blocks can work with 303 GT2 belts without a problem, I just need to make the appropriate tabs (which I wanted to do yesterday, but couldn't find the time :( ), which have an inverse GT2 tooth profile in them so they keep them in place (like the Reptar blocks have), instead of the raised area which tensions and clamps the belts now.
As for the 6mm rods, the blocks were designed to provide 269 mm of space between them for the rods on both axes. At least according to the latest UM drawings. It is possible that some printers print them a bit too thick due to overextrusion, and when that combines with imprecisely cut rods, problems start appearing. A bit more room could be added, but I fear that would compromise the rigidity of the blocks. On the other hand, situations like that are exactly why I uploaded the STEP files. :mrgreen: There's just too many different setups, and while I try hard to cover as many as I can, you can't get them all.
No hurry with the tabs (at least on my account)! I'm still painting..
I'm also still waiting for my E3D hotends to arrive. Royal Mail is a bunch of #@°§¦tards, too stupid to find my house it seems - internet tracker says "undelivered/unclaimed", but I was waiting all day for the doorbell to ring.
Still nothing today, I hate it when people don't do what they're paid for... (Yes, I will vote against a general minimum wage in Switzerland!).
Sorry for my mood here, hope to get rid of that habit some time o.O
Jonny thanks for the great list, chopmeister's blocks have made a massive difference quieter and it seems more accurate. One thing I did notice was the small belt pulleys where jamming against the wooden case due to the play in the rods. Should there not be a spacer between the bearing and the pulley? ( I bought it secondhand so something could be missing )
The list is far from complete, actually
I just got some very nice G11 fiberglass plates to make my new build plate with. More updates on the complete Z-stage will follow soon. I can do only one thing at a time..
The Original Ultimaker actually doesn't have any spacers between the pulleys and the bearings. The UM2 does have these. I'm thinking about a solution for that.
Actually I don't think that having spacers is necessary. But I'm going to make a little tool to help with the installation, so that all the pulleys can be mounted with the same distance from the walls / bearings.
If your slide back and forth inside the bearings, you should think about stabilizing them. That's not supposed to happen, but it's possible and it is a known problem (I heard something like that from the first UM2 users).
I'll make some rod-endcaps which integrate a pointed setscrew to hold the rods in place. Will take some time though, but it's one of the next things on my list.
I'll make some rod-endcaps which integrate a pointed setscrew to hold the rods in place. Will take some time though, but it's one of the next things on my list.
If you're not too much in a hurry, you may use a design I made this week. It uses a bolted end cap from Thingiverse and a ball clamping screw.
That sounds very interesting!
I was thinking about those screws, too. The good thing is you can choose either pointed or ball clamping setscrews... I already have pointed setscrews, but if it improves things, I'll get some of these as well.
/Edit:
Are these ball clamping screws really suitable for constant movement? These are not like ball bearings, or are they?
Are these ball clamping screws really suitable for constant movement? These are not like ball bearings, or are they?
They have a ball inside which can move if you mean this...
Hmmm, yeah of you look closely at the link, the ball clamping screw has a ball in the tip. It's not just a rounded tip. It has a ball bearing in it.
That might work. What you really need is a small thrust bearing and a hardened race....
Yeah I know it's a ball, like a ball pen. But is this suitable for actual rotation? Won't it wear out quickly?
I don't know what these screws are normally used for, but I woud have guessed they're just for fixing stuff in place (like any other setscrew)...
Anyways, it's most likely better than a pointed setscrew. The screws shouldn't be tightened so much that they clamp the shafts anyways, it's just to hold them within their bounds...
You could of course add another standard ball bearing, put in a piece which is made so that it can't slide out of the bearing and holds the setscrew... But that probably goes too far
Working on the new super-sturdy z-stage & build plate right now. If all goes well, I'll machine my new build plate today. Received really nice material for it: https://shop.maagtechnic.ch/ishop/product/catalog/01/product/183735.xml (site is german or french only...)
I haven't tested the ball clamping screws yet. I'll keep you informed.
I've been toying around with the z-stage, and I've noticed something:
The Ultimaker 12mm linear bearings are totally crappy?!?
I noticed that I wil sacrifice a lot of build height if I try to use two linear bearings per side instead of just one (in order to increase stability).
Incidentally, I compared the Ultimaker bearing against a bearing of the same dimensions from Misumi. While the Misumi bearing (actually low-cost model from Vietnam, not even Japanese made) slides without any friction and with almost no wiggle at all, the Ultimaker bearing feels like they used cubes instead of balls, and it also wiggles a lot more.
Note that I used the Misumi precision shaft for that test, but it seems like this Ultimaker bearing is a real no-go. I wouldn't have thought it were that bad...
Did anyone have a similar experience?
My impression was that it's quite important how you embed the linear bearing into the wooden panels. Mine do actually have so few play that I had the problem of a sticking z stage because the two z rods very not perfectly parallel. Looks as if tolerance variation is quite large...
I'm wondering on how to attach the bearings as well...
I'll definetly have to redesign the entire z-stage, so there are several possibilities. One would be to use flanged bearings. The problem is, I want to have two bearings per shaft. So if I attach the bottom one with a flange, ok. But how do I mount the top one perfectly in the same alignment...
...
Just had an idea
Will put it to paper and report back...
/edit:
Seems like this might just work. It's not going to be the cheapest solution ever, and it's probably going to be heavier than the whole rest of the printer, but at least it'll be nice and solid.
Found an easy way to connect 2mm steel plates using square nuts and short bolts, I hope this will work
The only bummer is, I have to buy new 12mm linear bearings again, need flanged bearings...
If you're not too much in a hurry, you may use a design I made this week. It uses a bolted end cap from Thingiverse and a ball clamping screw.
I have heard of drilling a shallow hole into a shaft (smaller then the ball bearing), so a ball bearing dropped into a set screw hole will have something to index with. Besides having a precise index, the surface of the shaft is not deformed like it would be if you clamped directly to the shaft, which makes it easier to take things apart later.
I imagine this screw is for that sort of application. The ball might be separate to make it easier for the set screw to turn during the final tightening and when the set screw is removed.
Recommended Posts
Top Posters In This Topic
85
22
21
13
Popular Days
Jan 16
21
Jan 17
12
Jan 21
11
May 7
10
Top Posters In This Topic
jonnybischof 85 posts
Dim3nsioneer 22 posts
aviphysics 21 posts
zumfab 13 posts
Popular Days
Jan 16 2014
21 posts
Jan 17 2014
12 posts
Jan 21 2014
11 posts
May 7 2014
10 posts
aviphysics 9
I got good results from the banana blocks. However, the one that runs on the 8mm rod at the back of the machine needs to get shaved down a little or it touches the threaded rod. Doing it again, I would consider using the Reptar XY blocks and clothespin springs.
Adjusting tension with the banana blocks worked really well, but assembly was a pain.
In regards to crimping the bushings, I think it is important to examine the fit of your printed blocks carefully before assembly. I had to scrape off some defects in the print to get a proper fit. When tightening the blocks, they only need to be tight enough to keep the bushing from sliding out. They don't need to be clamped on super hard. If they are like the banana blocks, they are designed for there to be a gap between the blocks when properly secured.
Link to post
Share on other sites