3D printable Visco fuse machine

[Simoski]

Ok I'm going to print the section of the funnel come die in the image below ( in the next few hours ) to see if my printer is capable of dealing with the intricacy of it

 

 

Edited January 29, 2018 by Simoski

[Simoski]

Ok so I did the test print and even though as the holes enter, every thing is fine, by the time its printing the holes within the interior of the funnel come die, its a mess. ( see images below )

There just does not seem to be enough space!

 

What I'm gonna do:

1 reduce the number of bobbins on the top spindle down from 15 to 12, this will give 20% more space to work with

2 reduce the complexity of the holes so each is basically a simple tube probably at a 30 degree incline

3 each hole will enter the funnel higher up, this is so there is more plastic when the hole meets the interior funnel hole.

4 widen the base of the funnel from 20 mm to 25mm, this will give more plastic to bore the holes through

 

 

Edited January 30, 2018 by Simoski

[Simoski]

And I remodeled the funnel / die with 12 holes instead of 15 and did another test print with much better results...

 

Edited January 30, 2018 by Simoski

[fredjr]

As is typical when first playing with 3d printing, your print quality is pretty bad. I highly recommend Simplify3D. The software makes it fairly easy to adjust all the different settings to improve your print quality. On the website they have a quality troubleshooting guide to point you in the right direction even if you don't use their software.

[Simoski]

Hi Fred, I've got Simplify3D now... what a good idea, its a much better slicer than CURA.

 

Any idea about the friction coefficients of PLA and ABS plastics? Is it viable to spin them against each other without much heating?

[Simoski]

Ok, I've finally got some time this weekend to start printing and do more modelling.

Expect and update during the weekend.

[Simoski]

Hi

 

Someone pointed out that having a huge single print of say the entire top tray with all the spindle uprights will work but will require an entire reprint if say one of the spindle uprights breaks.

Now I know 3D printing takes a LOT of time, so @ first I thought I'll "thread" them ( the spindle uprights ) so you could just screw out a broken one, print a brand spanking new one and screw that... in.

But I doubt my printers ability to print the thread, so I've gone wth a pin based system as you can see in the following image...

 

 

over the next day or two I'll print the three parts and put them together to see if the idea will work.

 

8 )

Edited February 12, 2018 by Simoski

[OldMarine]

Why not print the holes a bit small and tap them? Wouldn't take but a couple of minutes to do all the pins on the platen.

[Col]

I`d just taper the holes and pins slightly to provide a tight friction fit.

[MrB]

Sooooo many ways to skin a cat, as someone said above.

 

I'd just make the peg with a square "fot", and with a slot in it, protruding under the plate. Lock it in place with flat toothpics. (or purpose printed replicas.)

Threads will be more "complicated" and after you break a pin a few times, they might get damaged.

Friction fit takes a lot of precision, and if a pin starts rotating the hole could possibly become larger over time, making new pins a "special order".

I like the slotted pin approach in the test sample. Possibly since it was my suggestion from the start, and Simoski took it, and made enough changes to it to make it his own.

I think it's probably over engineered, but then again, a bit more mass in the rotating assembly will just provide more RPM stability, so that the threads are spun evenly. A positive, not a negative.

[Sleipner]

Just found this interesting thread. I'm working on a design myself but chose to take a different route with a stepper per plate, pulling mechanism etc as then thread sizes, fuse diameters etc is just a parameter change away in a control box. Of course a bit more expensive and complicated, but I like the challenge.

 

In any case, I saw you were going for a ball bearing which is nice, but I found a lazy susan bearing to be fine as well. Dirt cheap and easy to install. Did you consider that? Have a look here: https://www.aliexpress.com/item/32841516387/32841516387.html

[Twignberry]

Mate,

 

I've been off the forum for a year due to an overseas posting but I'm heading back soon.

 

I've made countless generations of designs, can't say I've perfected it but I have found a lot of pitfalls.

 

Couple observations

 

Print Quality

Judging by your build volume I'm guessing you have a CR-10. You should be able to get muuuuch better than what you are getting, takes a bit of practice to optimise settings but there are guides all over youtube. I would recommend PLA, it is infinitely easier to print and get great results. You should be able to print that 15 groove die flawlessly. Added benefits are higher strength, wear resistance, rigidity and flame resistance, at the cost of a little brittleness. No chemical incompatibilities found for standard comps, just don't get it too hot.

 

Bearing

The ball bearing will be fine, deep groove ball bearings are not meant for axial load but if they are small like this case the only issue is a little wiggle. I chose lazy susans because the through bore is bigger (19mm) this allows you to put gears and pulleys below more easily.

 

Manual Winding

Make sure manual wind is on the collector, it is the most likely to bind. If you have the crank on the spools, the pulley might slip on the collector

 

Die Sizing, angle, ratios thread counts

Pretty sure its hiding in my previous posts, if not, I'll try dig up notes

 

Funnel

Would strongly recommend making it removable. Its a PITA threading the die, a lot more so if you cant get fingers/tools in

 

Tracers

Make sure you can adjust the width between tracer threads. This will allow you to adjust the flow of powder into the first die. Needs to change as you move between BP and Chlorate based comps.

 

Capstans

My Alu extrusion MKI machine, while not a 'capstan' did use a direct drive winch that kept a constant feed rate. The collector spool deliberately was underpowered and stalled to simply take up slack

[Simoski]

Cool I feel like I'm on the right track...

[Simoski]

@Sleipner nice idea with the lazy susan thing, if only I'd thought of that earlier.

 

I'm gonna run with the ball bearings for now..

 

@Twignberry, thanks mate, I'll definitely be asking you more questions and as for the die/funnel... yes definitely removable

[chuckufarley]

Hey twig. Good to see you back around again, any updates on your laser cut wooden machine?

[Twignberry]

@chuckufarley

 

Still a few months out from move date and getting settled but I'll keep you guys posted.

 

Had a few issues I'll have to sort out for gen II

[OldMarine]

Hey! Twig is back!

[chuckufarley]

Awesome! Glad to see you back around again, a few of us thought you might be gone for good. Hope all is well, and look forward to your future designs, and creations.

[Sleipner]

@Twignberry

 

I looked up your work on the visco machine, impressive work! Shares a lot of ideas with the design I'm going for. I've opted for i2c controlled steppers to allow for as many moving parts as one could ever want (modular design with as many plates one wants feeding thread, paper etc). Also, with one of those extra steppers I'm working on a controlled powder dispenser mechanism that gives a more deterministic amount of powder.

[Simoski]

@Sleipner are you gonna share your design too? Id est provide the STL / GCODE with us?

[Sleipner]

@ Simoski

 

Yeah, sure I will share what I have eventually, not ready for prime time yet though. If you're in a hurry and you need a machine design now, don't wait for me. From looking at Twignberry's design I think it will satisfy most people. My design goals are 1) for me to have fun building the machine , 2) not minimise machine production cost, just reasonable, see point (1), 3) flexible, modular, extendable, hence a build around i2c which comes at a slight cost increase. A build around typical 3D printer components (like twignberry's) is great cost-wise and sufficient in the vast majority of cases, but in my design I forfeit those positives buying me somewhat greater flexibility of having many more actuators that can give me more layers, controlled powder delivery, direct drive pulling the fuse at constant speed, controlled receiver roll etc. I can also hang various sensors off the i2c to measure various things. Did someone say over engineered? oh well, for me building the machine is almost as fun as using it.

 

 

Two thread layer visco fuse is the first proof of concept goal maybe with a controlled NC-coating step, but my main goal is various forms of time fuse (three layers of thread with tar (or perhaps beeswax) in the outer two layers, or paper reinforced visco). I'm looking for input on what kind of paper one could put in a middle layer, and where to source strips of that in sufficent lengths( >500m per roll). I'm also looking for good sources of cheap hemp thread (0.5mm and 1mm diameter). The time fuse I've reverse engineered had an inner layer of 10 x 1mm, then two tarred layers of 7 x 0.5mm. Will need a lot of cheap hemp to make any length of time fuse I saw in "The Agora" forum a group buy for hemp, but I understand it was some time ago, and shipping to Oz would kill the good price.

[Simoski]

@Sleipner, great! Sharing is caring.

Its sounds well over engineered mate yours will be vastly more configurable than mine. I wish I could help you with ideas for the time fuse but you're ahead of the game on that one.

 

My goal is to build a cheap, simple machine, where most parts are 3d printed or widely available and the resultant fuse will be waterproof, will not blow out in the wind, have a constant furn rate, look professional and be suitable for sale.

 

Here my first major component printed... the top spindle base plate

 

 

What do you think?

[Sleipner]

Looks good!

 

I don't have good access to a workshop where I am at the moment, so by necessity I'll be working mostly with 3d printed and off-the-shelf parts as well. But I am trading some $$$ for configurability as to me that's what makes it fun I added some equations to the gallery and pointed to them in Swimmer's parallel thread that can help with designing speed ratios etc. Will be adding to those in the next coming days as well to make it easier to apply.

Edited February 20, 2018 by Sleipner

[OldMarine]

Sleipner, Lloyd recently parted with a Bickford time fuse machine and could probably offer some good tips on materials and processes for that area of adventure.

[Sleipner]

@ OldMarine, thanks for the tip!