It's being added for the next release. I am still working on the fan duct, which is complete but I'm making some changes to make it easier to print. Plus I've been sucked into Pokemon Go with my sons, so the last few nights I've been doing that.

 

Primarily flat or smooth rounded surfaces. I first noticed it when printing some very large cylinders and vases to test out the build volume. On more complex parts I do not notice it as much, if at all.

 

This honestly is some great detective work congratulations. Hopefully this will carry over to a lot of different printers that have a similar print pattern . Have been watching this closely and you guys did some great work..
I have my electronics coming and can't wait to get to printing on my C bot..
Man this can be a bear trying to get this thing Square. I have the z-axis going smooth but the x-axis needs a little work.

 

Indeed! Great detective work. I have another printer that is also direct drive that I will add a damper to as well to see if I can notice any difference in ultimate quality, though I've never noticed this particular issue on that printer (old Frankenstein-Bot).

As for squaring the frame - it can be a PITA, but I cannot stress the importance enough. You will be much happier when it comes time for functional testing if you take the time to square the frame correctly.

 

There's a mod for that...

 

So i have 2 questions, there appears to be different models/versions, which one is deemed the most uptodate/best at this point?

Secondly, i 'hope' im doing something wrong, because when i put the files into 3dhub, the checkout price is a WHOPPING $300 for PETG and $240 for PLA......thats for 92 prints.......that number seems high but who knows.......i seen the dbot thingyverse and it said like $550 total cost....but the freaking aluminum extrusion is like $200 alone + the plastic parts......thats something like $1k + with everything else........ so to be clear i dragged and dropped all the files from the cbot rework folders, except the 'optional' folder, which means it would have cost even more with those. thats straight cray cray

1 last thing, i see mini v wheels are being used, however i already have bought poly full size v wheels, will the printed parts still work with the bigger wheels? or are teh holes going to need to be in different spots due to bigger wheel?

i originally was aiming for the voxel ox, and was getting the parts little by little, then i seen this cbot and it looked like it might be better for 3dprinting, thought id just build a dedicaded cnc mill down the road instead..........but now im realizing this cbot is so much cray cray more expensive, $200 aluminum, $300 plastic the voxel is like $60 rails and like another $60-$100 for aluminum gantry plates/blocks/ect im just so confused now

 

The pricing for the printer neglects the cost of the printed parts, it assumes you'll be printing them yourself, or getting a friend to print them for you. For the c-bot rework, and probably the d-bot and others, mini-v wheels are required, but if you have normal wheels you could modify the files yourself to make them fit, but there might be a version floating around for the normal sized wheels somewhere. The largest expenses on the bot are probably the aluminum and wheels, and the hot-end (unless you get a cheap knock-off), and possibly the controller board depending on which one you choose.

3dhubs and such are meant more for one-off things than something like this, so it'll be very expensive to get them printed that way. If you print the parts yourself, or get a friend to print them, the cost is close to negligible compared to the rest of the components.

Also: For the C-bot, the most up-to-date one that I'm aware of is the C-bot Rework in the files here. There are some modifications floating around to it, but the files here for the re-work are all valid still and work just fine.

 

Do tell!!
What is there? Its seems like 100's of bolts to tighten then loosen then tighten etc. only to find its not square somewhere else.. Meh

 

He has made some motor brackets and bottom bracings that look rock-solid and because they're one-piece and everything is a tight fit, they come out square if they were printed that way. These pieces would replace the current motor mounts and bottom brackets.

 

@Star Crator already told you my sales pitch Here's the link: SpiffBot CoreXY 3D Printer by spiffcow

A word of caution though -- make sure the printer you're using to print these parts is 100% square. A small error can mean a lot of wasted plastic.

 

LOL I was thinking about how square my cheap i3 was and whether or not my printed parts were printed square... Will have to check.. would be cool to use your parts but , really? Just thinking of printing all the parts over and taking apart everything is killing me.... But....

Oh and on another subject... I opened @Chris Roadfeldt generator in Openscad and haven't any clue how to use it... What the heck Meh.... Obviously I have no clue how to use Openscad or the generator... Maybe someone could point out a youtube vid or explain to me what to do... I looked at the comments on Thingaverse but only got more confused..

 

The last time I looked at your design you were in the process of testing it... After further review I think you have gravely set back my time table for the completion of my printer..
Carls design is rapidly being improved by himself and others to the point I don't think I will every get this done.. lol..
definitely thinking of switching the corner pieces... I have a large format and think the added stability, rigidness and being square might help tremendously ..

 

im up to about $580 right now and im still missing the frame, frame connectors/hardware ect, and heatbed/plate

basically i went replicape which was about $170 + wifi $15 + memory card $10, then hotend $30, titan extruder $80 or so with motor and mount which was 3d printed which was suprising and flexes kinda strange,poly wheels x30 $54, flex couplers x4 $15, essentric nuts $19, power supply $37, 4 nema 17's $48, few other things.

anyways when i first started looking into this i asked on another forum about the plastic parts and they said 3dhub, i for some strange reason assumed the prices were reasonable, so now im kinda screwed cuss i mean i have no idea what these things called friends are. so i think im left with 2 options, go back to the voxel and buy all the openbuild parts which is still going to be fairly expensive i think as whats been happening to me is i 'think' i have a good idea what its going to cost eyeballing it, checking a few prices, but then once i start buying the price magically skyrockets.

or my other 'potential' option is to buy a monoprice printer and try it out for 30 days *hint hint*

or who knows maybe i will like that monoprice mini and just keep it and sell all this **** i bought. i wanted a bigger build plate cuss i had some neat ideas on things to print but im not sure its worth this headache

 

If you don't have access to a 3d printer already, the voxel looks like it may be easier to go with, you could always build a C-Bot later if you wanted to once you get a printer running. Another option for the printed parts, is to maybe look around on reddit or the reprap forums and see if someone can print you a set, sometimes there's people/groups that will help people with building a printer by printing the parts they need for free, or at cost or small markup, but I'm not sure of how common they are or their reliability.

 

Yeah, the bottom corners are by far the most important part of the mod.. If you decide to go with them, just print them at 20-30% fill and 2-3 wall layers with cheap PLA.. There's basically no advantage to high infill with these.

 

The Monoprice Select looks like a good i3 variant.

If you really don't have access to a printer and aren't in a huge hurry I wouldn't mind printing you a set of parts for the cost of plastic and shipping.

@Muh_3d likewise for the bottom corners if your i3 isn't square.

 

I bought a wanhao i3 last year and sold it a few months ago for £30 less than i paid for it. Definitely the cheapest way to print parts if you dont have a friend with a printer

 

I found the squareness of the X axis rely on the balance of the tension of the two belts. An adjustable tensioner on at least one belt would be handy.

 

That's certainly a factor, but if your frame isn't square you're going to have problems. I've found that if the frame is off you can get it to seem square for a small piece by adjusting the tension, but if you print a large square you'll see deformations near the corners.

 

For the amount of times you'll adjust the belt tension, the clamps currently used are fine IMO.
An adjustable one would be significantly more complex, and would add weight (albeit not much) to the toolhead.

Any type of tensioner that sits on the belt also couldn't be used (spring tensioner) as they can't pass over idler pulleys.

 

What I mean is even if your frame is square, if you are careless tensioning belt (as I was at the beginning), your gantry won't be parallel with the rear bar of your frame.

About tensioner, I was thinking of something that can be found on the carriage of the Mendel 90 : a set screw being able to move the anchor point of one belt. I wouldn't rely on spring tensioners put on the belt. Other CoreXY design allow moving motor back and forth to tension belt at the pulley.

 

BTW: today I managed to print something for the first time. Hooray !

 

Had that exact issue when I first built mine. Had the frame as square as humanly possible, but did not tension evenly. Result was a skewed gantry that would not go all the way to the front or rear without binding. A quick re-think of how to tension was the fix for me.

On my machine I found the easiest way to tension the belts and keep everything even was to start with the frame as square as possible (obviously), then slide the gantry all the way back until it stops, touching plastic on both rear corner brackets. I then used two soft-jaw clamps to gently hold the gantry against the rear of the frame. One on the left and one on the right (one from above, the other below so as to avoid the belts). With the belts bolted down on the rear extruder plate, loosen up the front clamps just enough to allow the belts to slip through a bit. Then I used some large needle nose pliers to grab both belt ends (assuming there is enough extra length in the belts) and twist. This applies fairly even tension to both belts while keeping the gantry in a square orientation. Once you have satisfactory tension, using your other hand tighten up the belt clamps. Worked great for me and solved the "How the hell do I do this with only two hands?" issue.

 

I concur on the belt tensioner issue. Maybe it's just because I keep testing mods that require me to disassemble and reassemble my printer, but belt tensioning has been a pain point for me.

 

if the belts are the exact same length then wouldn't all you have to do is measure the length of belt pulled past the belt crimp/holder and match it to the other side?


so as for my previous comments, ive been thinking about how to build this machine using openbuild parts, so i have a few questions. for the corexy movement to work, do the rails need to be a certain length? or can you just build a perfect square? the belts are the exact same length correct, just one is above the other so they don't hit? If all that is true, then i think i have it figuered out, but feel free to point out flaws, pros, cons, ect doing it the way im about to explain.

so looking at this picture http://www.openbuilds.com/attachments/img_1034-jpg.5946/

my idea is to build a square or rectangle frame, preferabbly using as much 500mm pieces as possible to limit the amount of cuts required to build it, which i assume will leave plenty of space inside for something like 300 or 400mm square build plate.

from there, the top 2 rails that the motors are next to, look at those rails, now imagine them longer, and going past where the motors are so they stick out.

then using these

mount on one side the motor ontop pointing down , and on the otherside mount it to the bottom pointing up! Both of these are mounted using the 'overhang' i talked about previously, towards the center. The only issue i can think of this is that the rails might not be tall enough and the belts hit since this is kind of just 'winging it' but if that becomes an issue then using spacers to distance the motors off the plate should work. getting the belts to ride inside the extrusion 'channel' might not be possible however, feel free to chime in on that issue and if thats a problem.

from there all that leaves is the pulleys and taking the idea from the voxel thread something like this might be an option mounted in the corners and on the gantry plates http://www.openbuilds.com/attachments/20160113_203936-jpg.13484/


i also seen this done with just those little corner cast aluminum things that are 1 screw wide, but i think those might twist under torque. Now the issue is it appears 2 pulleys are needed in each corner, so imagine the pulley in the picture, but with 2 pulleys on it. those pulleys are expensive like wheels so im thinking i will actually use the wheels i bought since i have extra, and use a tiny washer then a big washer on each side of the wheel to create the 'lip' that the real pulleys have to keep the belt on it. then of course spacers to seperate the pulleys, which might not be needed with the washers there anyway.

then lastly i could just use the open build gantry plates for the X and Y, and then basically just use the voxel Z design for the bed on this. i think that should get me running and then i can use the printer to print extra support like backplates to the gantrys.

does this all make sense? or is there something im missing that these plastic parts are doing? or some math that needs to be taken into account for it to work?

also are those gantry plates 'open source', like can people print them for each other? or are they protected/copy right?

and if your wondering why go through all this trouble , the answer is, if this works, then i don't need 92 printed parts, i need like 6 gantrys, 4 motor mounts, a few pully brackets, endstop holders, boom done.

 

Finally finished most of my C-Bot's build after over a year of sitting. Still have some tidying up to do, need a larger heated bed/glass(may CNC some mill spec aluminum for bottom too), new mount moving extruder to center of 2040 bar in back(or front), Cable chains or expandable sleeving and a new Z endstop.

I moved the Z endstop to max vs min because it just seemed easier/more familiar when using M206 to adjust the Z height offset since I am really used to my Kossel Delta Printer. So far I'm not real impressed with the mini wheels from openbuilds as one is terribly out of round/poorly machined on lathe and the others like to come loose or not roll if I have them too tight. If I were to keep the wheel based XYZ setup I would probably re-design the tension adjustment to be flat vs tapered and just use set screws to adjust the bolts up & down or make a upper + lower design that holds the wheels in a fixed position and use the clamping tension onto rail to adjust them. Really considering going to some MGN12 linear rails for all axis though....will post more when I get the time.


Thanks to all who have contributed!

20160721_180901 by hax0red posted Jul 21, 2016 at 6:31 PM

20160721_181021 by hax0red posted Jul 21, 2016 at 6:31 PM

20160721_180913 by hax0red posted Jul 21, 2016 at 6:31 PM

20160721_181251 by hax0red posted Jul 21, 2016 at 6:31 PM

 

This worked for me when tensioning. The key item being to keep the pliers as free from using any kind of leverage or device to hold it in place. The pliers with the belts has to be completely free so the belts equalize in tension on their own.

 

@Carl Feniak
Is there any chance I could get a diagram of the top back corners and the Y axis pieces similar to the ones you did for the motor mounts? (Those were really helpful BTW.. Thanks again for making them!)

Now that I have these steel reinforced belts that won't bend enough for the 16 tooth pulleys, I finally have sufficient reason to mod the other motion components.. I'm thinking I'll have pulley size as an optional component in the constants file

 

You need neoprene belts to fit the pulley profile on the C-Bot, and those belts tend to stretch.. So the answer to your question is "yes, but only the first few times you do it"

 

i thought of another idea , get one of those newton scales

wrap your belt around that hook, pull until u hit a certain weight, clamp. rinse and repeat on other side, tada both sides exactly the same