Progress?

Progress?

Postby sane » Mon May 12, 2014 1:27 am

I've been looking at building a GUS Simpson for a while now. However I've held off because of the talk of arm redesign, and in particular the issue with the spool design and the fact that its not accurate enough to act as a reprap.

We come round to mid-may and these are still issues, and progress seems to have dried up.

Are we to assume that the focus is now on Lisa Simpson and the new CoreXY, and that GUS Simpson will languish - or is there any sight of a 'version 1.01' that encapsulates all the fixes that were deemed necessary?
sane
 
Posts: 12
Joined: Tue Jan 14, 2014 9:47 pm

Re: Progress?

Postby Niggle » Mon May 12, 2014 3:35 am

It is your choice, but you could be waiting a long time for an 'official' solution to all of the known problems.

If I was to start over at this point, I would build a SmartRap with the intention of using it to print GUS parts. With the plastic parts in hand, I would then have to choose between moving the electronics to the GUS or buying a new set.

For me, the critical part is whether or not I am having fun. For me, planning the next tweak to try to improve the device is fun. If it is not fun for you, you are likely to be happier building a more robust and reliable design.
Niggle
 
Posts: 84
Joined: Tue Feb 11, 2014 7:27 pm

Re: Progress?

Postby timrwilliams » Mon May 12, 2014 10:40 am

I 'second' all that.
I guess it's the interest/fun that gets most of us involved.
As for languishing - well apart from fitting stronger springs, I havn't done much to improve printing accuracy.
But I have been printing various small model creatures to entertain young ( and not so young) relatives over Easter.
So ---??
timrwilliams
 
Posts: 87
Joined: Fri Feb 14, 2014 10:32 pm

Re: Progress?

Postby sane » Tue May 13, 2014 2:00 am

Niggle wrote:It is your choice, but you could be waiting a long time for an 'official' solution to all of the known problems.

If I was to start over at this point, I would build a SmartRap with the intention of using it to print GUS parts. With the plastic parts in hand, I would then have to choose between moving the electronics to the GUS or buying a new set.

For me, the critical part is whether or not I am having fun. For me, planning the next tweak to try to improve the device is fun. If it is not fun for you, you are likely to be happier building a more robust and reliable design.

I just don't want to be in the situation of going full steam ahead with v1.0, only to find v1.01 comes along and negates it. In particular I'd like to build something with the accuracy to create itself and the base design faults fixed.

Since it seems to have gone quiet in here, with attention seemingly elsewhere, I wondered if those fixes were on the back burner or not.
sane
 
Posts: 12
Joined: Tue Jan 14, 2014 9:47 pm

Re: Progress?

Postby Nicholas Seward » Tue May 13, 2014 3:16 pm

I have been insanely busy lately. (I am a husband, teacher, and inventor in that order.) However, I should be back to extreme development in two weeks.

The CoreXZ is getting some of my focus now because I need to build 4 of them for my school so we can have some no-nonsense printers that kids can use. I will be done with them in a couple of weeks if all goes well.

Simpson style printers are the future of large scale printing IMO. I have lots of work planned this summer. I plan on working on the software a bit more for the current version of GUS to see if I can get him printing accurately. However, I will be working on other physical changes that should also help. Follow along it should be interesting. However, no one knows when we will get these printers to the point of self replication. I wish I could tell you what the future holds but all I can do is work on the problem and see what happens. Bottom line: I will be trying lots of different designs to fix GUS's problems. I won't be formally updating the design until I feel the changes are solid and worthwhile. I can't really hazard a guess on when this will happen.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm

Re: Progress?

Postby sane » Thu May 15, 2014 1:16 am

Nicholas Seward wrote:Simpson style printers are the future of large scale printing IMO. I have lots of work planned this summer. I plan on working on the software a bit more for the current version of GUS to see if I can get him printing accurately. However, I will be working on other physical changes that should also help.

Thanks for the reply.

I did take a quick look at the segmentize code and without really getting my head around the coord transform maths, it did seem as if there were scope for optimising and getting it running onboard the firmware. A number of places where I think certain quantities could be precalculated - plus it seems that Slic3r takes curves and turns them into small line segments, then segmentize takes the line segments and if it considers them too big, further subdivides them and increases the 3D maths overhead. Converting and optimising for ARM firmware would seem to offer the prospect of an order of magnitude processing overhead improvement (rough guess) and thus make it practical to just feed GCode to the printer.

I'd like to play in that, but to do so needs a printer, and that means having a design that's not obviously mechanically compromised. The off-centre/not round spools seems to me to not be one of those things that's going to be amenable to a software fix - the error maybe cyclic, but the relative rotational angles on the various spools will likely mean some seriously annoying calibration activities having to be done often as those angles migrate over time.

Oh, and by-the-by, has anyone looked at a a single arm robot, mounted above and on the central axis, rotated 360 by a servo and lifted in Z to create the necessary DoF? Slices wouldn't be flat but the working volume would be potentially massive and you'd only need one arm.
sane
 
Posts: 12
Joined: Tue Jan 14, 2014 9:47 pm

Re: Progress?

Postby Nicholas Seward » Thu May 15, 2014 1:38 am

Segmentize essentially does what is done in existing delta firmware. There is no question that we can do this in firmware but we need accuracy first or it doesn't matter.

I have a 3DOF single arm design but I am holding it back (it would be like Wally, very hard to calibrate with cheap parts.) because I have a 2DOF single arm SCARA with a linear z that I think will be more practical. For now that or GUS is my target for the 2015 Gada Prize.
Nicholas Seward
 
Posts: 738
Joined: Mon Nov 25, 2013 10:41 pm


Return to GUS Simpson

Who is online

Users browsing this forum: No registered users and 1 guest

cron