Grounded Experimental Delta Printer

Well, I know that this is one of the most unique and special 3d printers out there, but if you could make that thing working near to his maximum speed would be a hit in 3d printing... just imagine that, one cheap, precise, easy to build and printable 3d printer that is also fast as hell.

So you are saying...


The weak link by far is the filament drive. I am actually considering doing a direct mount. (After NY Maker Faire of course.) If a hobby machine breaks down it is almost always the filament drive. I have a bunch of ideas. I want a better solution before I KickStart in December or there is no way to use all the potential. (Let's not get side tracked on good/bad filament drives here. I will eventually start a new thread when I get to that point. )

From what I understand of filament drives, it comes down to the interplay between the hobbed bolt and the filament. This guy claims his design is good up to 350mm/s, and, it's by far the best write up on hobbed bolts I've seen.

From the article:

"Preliminary bolts... My best results were with 22 teeth, and less than 18 give bad results.
The internal groove must be sized so that the pairs of teeth square edges dig well into the filament sides.
Hence, it depends on the filament diameter and hardness... so there is probably no single 'best' bolt."

This is awesome to see the new design in action. I am a bit confused as to what advantages the new design holds over the previous iteration? It looks like the only advantage is more stability at higher print speeds. How does the volume of printed material compare between the two versions?

Either way I can't wait until the design is finalized and released, I'm already saving my pennies so I can get the parts to make one of these.

The advantages are almost too numerous to count.

The first machine was pretty much just a test bed. I never got it to print nice straight lines. I am sure I could have but I learned what I needed to know that the geometry works and I needed to work on making everything more rigid.

Advantages:
*More printable
*Larger macrocreation factor.
*Easier to calibrate (I did 1 calibration point for this version. Before I would do 30 to get a level print.)
*Quicker to assemble
*Cheaper
*Much less bearings. (I am down to 22 bearings and there is a movement to make a completely bearingless Simpson in the works.)
*Faster
*More rigid
*Cuter
*etc.

The volume is half of the previous but you can scale this design to suit your needs. This one is more in line with other printers.

nicholas.seward Wrote:
-------------------------------------------------------
> The weak link by far is the filament drive. I am
> actually considering doing a direct mount.

Ok, I suppose that you are referring to the filament drive used to feed the extruder, instead of the ones I suppose you are using for the motion, I know you are a busy man right now, but I would thank you if you could describe which part broke, how, and in which circumstances.

My PLA filament got turned into an oval and pressed out the bowden tube. I needed that much grab on the filament for high speed but I can let up a bit on the engagement and lower the speed. I am using a modified QU-BD MBE.

Could you do something like this between the fillament drive and the bowden tube? Something as simple as two bearings spaced 1.75mm apart in line with the major axis of the ellipse might be able to squeeze it back into shape, or at least enough to fit in the bowden tube.

It will certainly increase the force that have to be done, but maybe it works.... I was thinking right now in another way of transmit the force, doing that in a larger area would certainly decrease the deformation.

It appears the filament drive gear QU-BD uses is this, which looks to be their custom creation. The shape of the curve pretty much explains why It's getting squeezed into an oval shape. Extruded brass with the oval indent milled out. It'd be awesome to see a version with a square indent milled out, like the one in my previous. That should allow more grip with less pressure. Nicholas, you're in good with QU-BD, see if they can whip something up; they already have the extruded metal and the milling wouldn't change that much, just a much smaller bit.

How about 3mm filament. It has more surface area to 'bite' into and may give more pressure for a given drive type?

Solving the filament drive issue (it works just not at speeda Simpson is capable of.) is not a problem. Time is the only hurdle.

I was brainstorming/looking for solutions to the extruder speed issue. I had a memory floating around of the atlas rope ascender, but that's overly complex for this situation. Then I was thinking some sort of worm drive might work, but seems it's been done. Also, BFB did it in their BFB-3000 printer a few years ago and had issues with it, plus I don't think it would be fast enough. But somehow, that led me to the kraken, which looks rather impressive, even more impressive if you could drop it down to a single motor. Open air extrusion of 400 mm/s, burst at 900 mm/s, and able to perform at 20 microns (not sure of the print speed there).

Quote
nicholas.seward
(Let's not get side tracked on good/bad filament drives here. I will eventually start a new thread when I get to that point. )

Sorry, it happened.

However, I'll try to bring it a bit back on point. Are you going to be injection molding for the kickstarter? I assume the beta is all printed (way too small of a run, and too much potential for change), but I figure the kickstarter might be big enough for a production run. I know molds are expensive but it definitely would fulfill orders much more quickly. Although...you do want to make a quad-morgan farm, so who knows how you'll go about it.

(This relates to the arm idea shown in the September 11, 2013 10:47AM post. That's the post I clicked "reply" in!)

I don't think this 2-joint arm scheme works as-is, because the string headed for the second joint will float off the left-hand bolt when the first joint bends a bit to the left.

I assume you can't stand a full wrap around the bolt, to keep the string in position, because the friction would be too great.

You could use a pulley instead of the bolt to reduce friction, but when the string goes more than one turn the incoming and outgoing are going to rub and likely cause some erratic behavior.

Perhaps some guides to keep in and out apart could do the trick. The bolt pulley could be just a flat roller, and little guide pulleys could force the string to follow a very slight spiral to avoid interference between in and out strings.

What you'd really like is some frictionless guide to get the string past that first joint. Bicycle brake cable...

Edited 1 time(s). Last edit at 09/17/2013 09:55PM by DaveGadgeteer.

@Matt M: I am opposed to injection molding at first. I want an evolving design. I also want to make each machine to the customers spec. Eventually as we figure out the popular specs we could injection mold as it become cost effective to do so.

@DaveGadgeteer: I don't think a full string wrap is a problem. The fishing line is so slippery and the speeds are so relatively slow that we should be good. (The gear joint has a 270 wrap. I will let you know how that works out.)

What kind of fishing line works best?

I use 100+lb braided fishing line. Spiderwire, Spectra, etc.

Gear-arm Simpson #2 is alive and printing. The calibration is off due to a silly metric conversion error on my part so this print started poorly and eventually failed, but that will be tweaked easily enough. This squirrel is fast becoming the official first print for new Simpsons... with a little more work we should get a whole one to print.

[youtu.be]

I would like to second having the Simpson Squirrel be the official first print.

BTW, I finally settled on a naming scheme for this iteration.

GUS Simpson

Geared Union (the mathematical symbol) Simpson

Thanks owens. I kept on calling the machine Gus in my head and I couldn't figure out why I should have a U in the name.

It is so exciting to see two printers come together. GUS #1 and GUS #2 will both be at the NY Maker Faire along with Wally.

Edited 1 time(s). Last edit at 09/17/2013 11:56PM by nicholas.seward.

Nicholas, do you know where you booth will be at the New York Maker Faire??

I want to make sure and meet you and to see Wally and GUS Simpson.

I'll be there on Sunday.

Ralph

My maker number is 11494. I don't know if that will get you to my booth. The name of my booth is ConceptFORGE.

I can't wait! I look forward to meeting you.

Wish I was there. Can't make it because the distance is too great. Any1 able to take some photos and post them?

I will make a time lapse like or two. I will also be getting a pictures of all the Gada winning printers together. There will be a Morgan on my table and Chris is bringing his 3DPrintMi.

Sorry for the beginner question, but is it feasable for me to build me one of these as my very first 3d Printer? (once you've released the design)

i absolutely adore the design, also it's propably cheaper to get together than buying a prusa i3 kit, but i see quite a few problems, beginning with the fact that i propably won't get an instruction, also it'll be more annoying to get the printed parts together. Currently i wouldn't even know where to begin getting a precise base-plate, to mount the printed-parts onto...

One option would of course be to buy and build the i3, print myself the simpson parts and build that with my gained experience, although it would obviously be a lot more costly... but who can put a price to fun learning?

If you're interested in a Beta-Tester who is a completely blank (but quite enthusiastic) page in 3D-Printing... I'd be willing

Edited 1 time(s). Last edit at 09/21/2013 11:23AM by TheYang.

@TheYang I'm exactly in the same situation. I have been following the 3D printers world with interest for a couple of years, but i never have been convinced of building one for myself.... until now, Wally and GUS really caught my attention.

@TheYang: A few things, this would be a less than ideal first printer. I plan on having awesome instructions but the learning curve for 3D printing in general is a little steep and to add Simpson problems on top would cause some frustration. I would suggest a cheap bot with a large community and then move to the exotics after you get that down. I am planning a Kickstarter in December and hope to have it more beginner friendly. Especially after I get feedback from 20 beta tester.

@harguinxx: Good luck. Let me know if you decide to build one and need any help.

Hey nicholas, check out the XWinder in booth MP10 if you get a chance. [www.xwinder.com]

Thinking this is a toy that I could use.. Large format Simpson with carbon fiber tube arms and 3D printed end-cap joint/gear.

Matt M Wrote:
>But somehow, that led me to
> [url=http://www.3ders.org/articles/20130822-first-
> dual-pinch-wheel-extruder-for-3d-printers.html]the
> kraken[/url], which looks rather impressive, even
> more impressive if you could drop it down to a
> single motor. Open air extrusion of 400 mm/s,
> burst at 900 mm/s, and able to perform at 20
> microns (not sure of the print speed there).

Looking at the kracken I don't understand why they didn't just use a geartrain so that the motor can drive both wheels without any chance of slippage.


Also, I was just at the design and manufacturing show in chicago 2 weeks ago, and small run casting (plastic, and even metal) has gotten amazingly cheap. I think one company I was talking to was saying as low as $1500 for molds to a very complex nickel heatsink, and the piece parts were like stupidly cheap after that. I think if the volume was enough (potentially as low as 100-200 orders) nicholas could get molds made for a mini-simpson (maybe the longest part is 6") without running the cost through the roof. From this mini simpson, the customer could then "grow" any sized simpson they wanted.

Personally, I love the simpson design because there are so many potential variants due to the fact that you don't need to buy linear bearings, and threaded rods. Theoretically, you could print out some bering holders for each end, and then replace the length of the arm with a metal dowel, or bar. This would reduce the RepRap-ability of the design, but it would make it easier to change the size of your 3-D printer on the fly, and parts would be faster to replace. Come to think of it, I think that this is going to be my first project once I get a 3D printer.

@pjoyce42: One word of warning, torsional stability is important so small diameter rods would be the problem. Carbon fiber tubes would be lighter and have better torsional rigidity.

Actually my original thought would be to use some of the aluminum extrusions from 80/20 or rexroth. I don't know the exact numbers, but I would expect that the torsional stiffness is, at a minimum, comparable to the 3d printed plastic arms of the same cross-sectional area. I was trying to avoid round rods because then you would have to find a way to fix the rod. With the aluminum extrusions, you could print the female profile into the plastic end piece so that it just slots into place. I haven't done any math or modeling yet, so I'm just spitballing right now, but I can't think of a reason that the aluminum wouldn't have the necessary torsional rigidty.