Wally-0

Wally-0

Postby Nicholas Seward » Mon Jan 20, 2014 3:28 am

I thought I should start a thread for the original Wally. I am in the midst changing out the filament drive to my new design and printing a new XY UBIS MOUNT ARM. (I flipped the orientation that it prints, added kossel mount like GUS, integrated washer, and flipped the nut traps to the other side. I would have pictures if my printer didn't have a tangle in the middle of the spool. Reprinting now.)

Here are some filament drive pictures.

WALLY FILAMENT DRIVE 2.jpg
WALLY FILAMENT DRIVE 2.jpg (304.29 KiB) Viewed 64580 times


WALLY FILAMENT DRIVE 1.jpg
WALLY FILAMENT DRIVE 1.jpg (313.07 KiB) Viewed 64580 times


After this, I have some software changes and some calibration fixes that I am anxious to verify. I hope to get Wally up to snuff.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Wally-0

Postby Nicholas Seward » Tue Jan 21, 2014 1:18 am

More progress on getting Wally-0 back up and running...

I added my modified "xy ubis mount arm".
WALLY NEW ARM.jpg
WALLY NEW ARM.jpg (299.79 KiB) Viewed 64548 times

It uses the same Kossel compatible mounting scheme that I am using for LISA and GUS. You can use this to mount fans, level probes, etc. Anything that has been made for Kossel. I did some other small changes. Flipped the nut traps to the other side, integrated the plastic washer, flipped print orientation, etc.

I error mapped the bed position
WALLY Z CALIBRATION.jpg
WALLY Z CALIBRATION.jpg (334.37 KiB) Viewed 64548 times

Wally Position Error.jpg
Wally Position Error.jpg (46.87 KiB) Viewed 64548 times


It is surprisingly good. I would not have expected the error to stay under 1mm. You can see that the printed spool adds a lot of error and then the triangulation error is a bit more well behaved but more substantial. Anyways, it will be easy to plop this data into my preprocessor to get perfect z positioning.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Wally-0

Postby tommythorn » Tue Jan 21, 2014 1:44 am

How consistent is this when you run it again?

I'm looking forward to how the milled pulleys work out. (You ordered them for GUS, but I assume you could use them here as well?)
tommythorn
 
Posts: 90
Joined: Tue Dec 03, 2013 12:14 am
Location: Silicon Valley, CA, USA

Re: Wally-0

Postby Nicholas Seward » Tue Jan 21, 2014 1:56 am

I did some spot checking. I was usually within 50 microns. Better yet, sets in one spot check run where usually about 20 microns +/- of being off from the average offset. (You have to remember that this is with zeroing by just placing the bed at the bottom and making sure the string isn't slack. That manual zeroing is bound to skew different runs in different directions. To see that skew under 50 microns is pretty unexpected.) I think I got the z whipped. I need to figure a way to get the xy just as good.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Wally-0

Postby tommythorn » Tue Jan 21, 2014 2:21 am

If the calipers where integrated into the design (effectively turning it into a close-loop system) then all the messy calibration problem would evaporate (with a "small amount of firmware"). And no drift to worry about.

EDIT: I'm naturally assuming one would tap into the debut port on those calipers, but I just realized that it wouldn't be fast enough for real-time control. It could still be used for automatic and continuous background calibration of the (existing) open-loop control. We would have to find or fabricate the connector so people wouldn't have to mess with fine soldering.
Last edited by tommythorn on Tue Jan 21, 2014 2:28 am, edited 1 time in total.
tommythorn
 
Posts: 90
Joined: Tue Dec 03, 2013 12:14 am
Location: Silicon Valley, CA, USA

Re: Wally-0

Postby Nicholas Seward » Tue Jan 21, 2014 2:28 am

@tommythorn: I can raise my glass to that. Closed loop machines are coming soon to a cheap hobby machine near you. I actually want to one up that. I want to have a machine which is initially unaware of its geometry and may only only have repeatability over a small time scale. The theory goes that you could build your approximate work space map when you start and it can continually refine/correct its internal model during a print. Cardboard 3D printers with $1 DC motors anyone?
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Wally-0

Postby see3d » Tue Jan 21, 2014 4:51 am

Nicholas Seward wrote:Cardboard 3D printers with $1 DC motors anyone?

Ok, I'm in... :D
see3d
 
Posts: 67
Joined: Tue Dec 03, 2013 1:24 am

Re: Wally-0

Postby Nicholas Seward » Tue Jan 21, 2014 1:35 pm

Well, I am making lots of progress on my Wally refinement. I added guitar tuners to my build so that I could tension the system for a run and then quickly detension. I can tell you already that it is a godsend.

http://github.com/NicholasSeward/ConceptFORGE/tree/master/WALLY/STLs
There is some folder naming issues with Git that I need to figure out so the necessary files are weirdly located for now.

WALLY GUITAR TUNER 1.jpg
WALLY GUITAR TUNER 1.jpg (300.75 KiB) Viewed 64517 times

WALLY GUITAR TUNER 2.jpg
WALLY GUITAR TUNER 2.jpg (14.98 KiB) Viewed 64517 times


Interesting note: When I tension the hot end gets pulled down about .5mm. I expected a bit more. I think I will have the bed come up at the beginning of a print and then use the tensioners to perfect the offset before I go into hot end heatup mode. I will get my 3 point software level when I am at tension otherwise this scheme doesn't work.

Well, off to school. I will try to squeeze in some time today between classes to finish refining the preprocessor. I feel like I should be getting some good results from him tonight.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Wally-0

Postby Cozmicray » Tue Jan 21, 2014 5:27 pm

Are these tuners for tensioning easily obtained?

Seems like the design is getting MORE complicated
moving away from a production type printer.

I thought I read somewhere the string was going to be replaced by timing belt.

So should all string type and timing belt mechanisms be un tensioned when not in use.
I wonder if my inkjet printer it's belts when it sits there waiting to print?
Cozmicray
 
Posts: 126
Joined: Tue Dec 03, 2013 10:23 pm

Re: Wally-0

Postby Nicholas Seward » Tue Jan 21, 2014 6:17 pm

Tuners

Feel free to PM me if you would like me to send you some tuners. They come in packs of 6 for $10 so unless you have 2 more Wally's hidden there is no need to to buy 6.

I don't know if I would say this is much more complex. In my book it is simpler for the user and it only cost $3.33.

I don't know if detensioning helps. I assume it will keep the other flexible parts in my design from exhibiting creep. I am sure your ink jet printer doesn't need to be detensioned.

I do plan on playing with belts. That will be a way to make calibration of the xy arms easier. I will maintain both a belt and a string version if I do so.

I personally have no plans on mass producing Wally in its current form. My first foray into production will most likely be completely lasercut. I have yet to disclose the specific form factor.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Next

Return to Wally

Who is online

Users browsing this forum: No registered users and 2 guests

cron