GUS v2 Goal Setting/Discussion

Re: GUS v2 Goal Setting/Discussion

Postby Jfriesen » Mon Dec 07, 2015 5:21 pm

I agree string on the inside is cleaner aesthetically. This thing is kinda cool in a showing how it works way but ultimately it feels a bit too much like the mechanical guts are exposed. I considered perhaps a small cover piece or something of the like but I haven't thought of one that would be simple/low profile enough.

Functionally I think this arm is a bit stiffer and smoother (probably helps the gears were printed with z-aligned axis). I don't want to build the arm with more than two pieces as I worry it might be tedious/misaligned. A careful design could probably find a way for the mating screws to pull the thing into good alignment. I think you'd have to print 1/2 of the gear as a seperate piece then build the thing like a sandwhich.

I'm going to finish this iteration and see what I learn. I know in the internal string iteration we could make inserting all the pulleys a lot easier without much redesign. Just make a larger hole on the nut side of the pulley axes so you could pop all the pulleys/washers onto the bolt then screw in a cork-like piece at the end that would provide support for the end of the bolt and seal up the hole. That would already make building the actuators a fair bit easier.

The other problem with the previous one is strings tend to overlap on the drive pulley leading to uncertainty in how much string you're letting in/taking out because it changes the effective diameter of the pulley. I also had a string break during a print and I'm fairly certain overlapping strings on the drive pulley was the culprit. There was a characteristic popping noise inside the actuators that broke which suggests this.

This new iteration addressed that. The key was having ample distance between the guide pulley that feeds onto/off of the drive pulley and the drive pulley itself. This allows the string to produce that clean helical spooling without requiring the string to stray to far from the pulley. Again, with a small modification to the previous internal design we could probably incorporate this. I think about 2-3cm is the minimum length where things start overlapping.
Jfriesen
 
Posts: 60
Joined: Fri Nov 06, 2015 5:06 pm

Re: GUS v2 Goal Setting/Discussion

Postby ekaggrat » Tue Dec 08, 2015 1:44 am

yes i agree that to make a the stringing easier modifications need to be done but those would be only minor like making hole for accessibility and making the gap between the pulleys and the arms on the back side a little bigger. One thing i noticed was that the motor pulley was not centered to the pulleys , this makes the string returning on the lower side wrap on itself a bit i think .

So what modification you think needs to be made to the pulley as i think that is more important . ( any napkin sketches would do) .. I will have to stop the build for now then.. anyways i was waiting on the filament and the smoothieboard ( clone ) . :( :(

I think iteration 2 is better in the regard that it can be printed on the side . One thing that can be done is that if the bearing axle is shifted to the side like the original then the arm can be printed in one piece , simplifying it more.
Attachments
IMG_20151208_081800.jpg
IMG_20151208_081800.jpg (797.54 KiB) Viewed 78185 times
IMG_20151208_081745.jpg
IMG_20151208_081745.jpg (695.12 KiB) Viewed 78185 times
IMG_20151208_081716.jpg
IMG_20151208_081716.jpg (829.76 KiB) Viewed 78185 times
ekaggrat
 
Posts: 56
Joined: Tue Mar 04, 2014 3:24 am

Re: GUS v2 Goal Setting/Discussion

Postby Jfriesen » Tue Dec 08, 2015 2:16 am

Looks nice! Love the teal. I'm making my next ones out of opaque white t-glase (if it ever ships :/). Should look nice and shiny. I think it works as is, only one of my three actuators was making that sound before one broke and it was making it from the beginning. My guess is that one just was strung poorly, but part of the problem is it's very difficult to see how the cable is spooling since it's tucked down in there.

I'm almost done with my arm revision I think. This arm gets a few centimeters more throw (31-32 cm compared to 29ish) has cleaner cable routing, can be strung in just a few minutes and has a much larger mechanical advantage which means better resolution. Also When fully extended at 32cm it is much further from the undesirable singularity so it remains much stiffer even when fully extended. The layout of the cables also produces a much stiffer actuator because they cover a larger width.

I actually already had that thought this morning on condensing into a single part, but I don't like offsetting to the side as it introduces unbalanced forces. I opted for some breakaway support material. I'll report back tomorrow with a vid of how to string it up and showing it working.
Attachments
2015-12-07.png
2015-12-07.png (285.31 KiB) Viewed 78185 times
20151207_175909.jpg
20151207_175909.jpg (536.21 KiB) Viewed 78185 times
Jfriesen
 
Posts: 60
Joined: Fri Nov 06, 2015 5:06 pm

Re: GUS v2 Goal Setting/Discussion

Postby ekaggrat » Tue Dec 08, 2015 5:33 am

looking cool.
i see you have upgraded to solidworks 2016 :( i will have to upgrade to look at the files. I will continue with this version . and finish it in some weeks( i am not as fast as you ) M.. meanwhile i will wait for your new iteration to be complete.



One more upgrade that is required is to change the 608 bearings to 605 or 625 . that way things can be slimmed down. whats your opinion on that?
ekaggrat
 
Posts: 56
Joined: Tue Mar 04, 2014 3:24 am

Re: GUS v2 Goal Setting/Discussion

Postby Jfriesen » Tue Dec 08, 2015 5:11 pm

If by slimmed down you mean mass wise, then I agree wholeheartedly. All the m8 bolts and bearings start to add up quick. However profile wise, I think we need to keep a lot of the width for rigidity sake. I worry going below 25 mm might start lowering the stiffness significantly.

Spent a lot of time yesterday and have almost completed a new clean assembly. I'm going to modify my github repository to include all CAD in various formats there so I can do better version control if I make any future changes (the bearing thing should be quick and easy). Perhaps you could add a folder/branch whatever with your version of the CAD for those who prefer internal stringing.

Sorry about the 2016 thing they upgraded our campus computers. :roll:
Jfriesen
 
Posts: 60
Joined: Fri Nov 06, 2015 5:06 pm

Previous

Return to GUS Simpson

Who is online

Users browsing this forum: No registered users and 2 guests

cron