3D Printed 32 Line Mechanical Television System

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Wed Mar 08, 2017 10:32 pm

Here's a 3D Printed Baird Inspired 32 Line Mechanical Television System!
It uses an Arduino, just like my current build.


youtu.be/a6dxp6cTvq0
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 9:36 am

Nice, the problem with printing Nipkow disks must surely be creating the tiny apertures? He did well here to get them down to 1mm which, of course, is far too large for a disk of this size.

HOWEVER, it seems to me these would be ideal for "bead" disks, where the aperture is the size of the optical quality acrylic balls (I use 1/8", 3/16", and 1/4").

What is the (working) size of your 3D printer bed Andrew?
gary
 

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 10:03 am

gary wrote:Nice, the problem with printing Nipkow disks must surely be creating the tiny apertures? He did well here to get them down to 1mm which, of course, is far too large for a disk of this size.

HOWEVER, it seems to me these would be ideal for "bead" disks, where the aperture is the size of the optical quality acrylic balls (I use 1/8", 3/16", and 1/4").

What is the (working) size of your 3D printer bed Andrew?


Not difficult to create small holes, actually. I could probably do rough holes which are 0.1mm diameter positioned to 0.05mm accuracy and then round 'em off with a drill bit.
I use a Flashforge Dreamer... here are the claimed specs...

Positioning Precision: 0.0001” on Z axis, 0.0004” on XY plane
Resolution: 0.004”
Build Volume: 9.1” x 5.9” x 5.5”

That claimed 0.0004" seems a tad optimistic to me!
But you see the volume - about 15cm max diameter. You could print a disc in multiple parts with a jigsaw-puzzle-like joiner and it should go together pretty accurately.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 11:08 am

Andrew Davie wrote:
gary wrote:Nice, the problem with printing Nipkow disks must surely be creating the tiny apertures? He did well here to get them down to 1mm which, of course, is far too large for a disk of this size.

HOWEVER, it seems to me these would be ideal for "bead" disks, where the aperture is the size of the optical quality acrylic balls (I use 1/8", 3/16", and 1/4").

What is the (working) size of your 3D printer bed Andrew?


Not difficult to create small holes, actually. I could probably do rough holes which are 0.1mm diameter positioned to 0.05mm accuracy and then round 'em off with a drill bit.
I use a Flashforge Dreamer... here are the claimed specs...

Positioning Precision: 0.0001” on Z axis, 0.0004” on XY plane
Resolution: 0.004”
Build Volume: 9.1” x 5.9” x 5.5”

That claimed 0.0004" seems a tad optimistic to me!
But you see the volume - about 15cm max diameter. You could print a disc in multiple parts with a jigsaw-puzzle-like joiner and it should go together pretty accurately.


Interesting, I have no doubt about the positional accuracy (it is, after all, a CNC machine) but I would have thought the width of the extrusion would be the limiting factor (as is the diameter of an end mill in a CNC machine) but I suppose the fact that it is an additive machine rather than a subtractive machine (CNC) means that can be compensated for?

None-the-less I would still have thought it unlikely that you could get a consistent enough extrusion width to be that accurate (~.01").

Impressive stuff if you can do it - I suppose you have done a test?
gary
 

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 11:39 am

gary wrote:
Interesting, I have no doubt about the positional accuracy (it is, after all, a CNC machine) but I would have thought the width of the extrusion would be the limiting factor (as is the diameter of an end mill in a CNC machine) but I suppose the fact that it is an additive machine rather than a subtractive machine (CNC) means that can be compensated for?

None-the-less I would still have thought it unlikely that you could get a consistent enough extrusion width to be that accurate (~.01").

Impressive stuff if you can do it - I suppose you have done a test?


It's not going to be really accurate in width which is why I said use a drill bit. Nonetheless, since I said it could be done and haven't tested - I actually have one printing right now. Here's the OpenSCAD source code if you want to see how it's done and have a play...

Code: Select all
$fn=32;

SCANLINES = 32;

DISC_RADIUS = 75;
DISC_THICKNESS = 0.5;


HOLESIZE = .25;
OVERLAP = 0.8;                              // 1.0 = no overlap, 0.9 = 10% overlap, etc.
OUTER_EDGE = 10;                            // distance from outer edge to first scanline hole
INNER_EDGE = 10;                            // distance from inner scanline to sync holes


// Auto calculated - do not change these!
SPACING = HOLESIZE*OVERLAP;
POSITION = DISC_RADIUS - SCANLINES*SPACING - OUTER_EDGE;

SYNC_POSITION = POSITION - INNER_EDGE;
SYNCHOLE_SIZE = 1;                           // radius



difference() {
   
    // the disc itself
    cylinder(r=DISC_RADIUS,h=DISC_THICKNESS,$fn=64);
   
    // a hole in the middle
    translate([0,0,-1])
        cylinder(r=1,h=2+DISC_THICKNESS);

    // do the scanlines and the sync
    for (i=[0:(SCANLINES-1)]) {
   
        // The scanline hole
        rotate([0,0,i*360/SCANLINES]) {
            translate([POSITION+i*SPACING,0,-1])
                cylinder(r=HOLESIZE,h=2+DISC_THICKNESS);

            // The sync hole
            translate([SYNC_POSITION,0,-1])
                cylinder(r=SYNCHOLE_SIZE,h=2+DISC_THICKNESS);
        }
    }   
}   


Attachments
nipkow.scad
(487 Bytes) Downloaded 632 times
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 12:18 pm

There are a number of factors affecting "creating small holes". The actual 3D model is easy to create as accurate as you want, but when printing you run across the physical characteristics of the printer and materials. First, after creating the 3D model, you run it through a "slicer". The slicer cuts the model into many very thin slices, and sends instructions to print each slice stacked on top of each other to recreate the original model. Firstly a slicer will put an outline around all the edges of the layer it's working with. So holes are created by printing a circle around the hole. All good so far.

The material being printed needs to be melted, and the properties of that material control how viscous it is, and also how rapidly it cools/sets. That is also affected by the heat of the surface onto which you are printing. So, if the material is a bit too hot, and a bit too runny, then drawing circles around tiny holes will tend to "fill in" the holes. They will be precisely located, just not really holes. But the big advantage is, you know *exactly* where they are because you can see the pattern in the surface of the printed object, and the circles around the holes are immediately apparent. So, heat up a fine needle and just push it through at the right spot.

Other factors - the first layer tends to "squash" a bit, because you want it to adhere to the printing surface... particuarly when printing fiddly bits like the fine hole circles. Those tend to lift off the print bed unless you have correct temperatures and distances. So it's a balance - we want to squash the first layer down so it sticks, but if we squash it down, then the molten material tends to spread and flatten out, and reduce the tiny hole sizes. Upper layers don't have this effect/problem.

Finally, if you do a print you can just adjust the hole size to account for the above factors and print again. If I select an 0.1 mm hole, and they all get filled in, then I'd up the diameter to (say) 0.2 mm and expect to see something like an *actual* 0.1 mm hole in the final product. If you're familiar with your printer and how it performs with different situations, then it's possible to get good accuracy... but sometimes it's tricky.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 1:12 pm

Here's the first prototype. I have a close-up alongside a mm scale. You can see the locations of the holes - quite apparent - but the smaller holes have been filled-in. Nonetheless, their center points are very precise and would be relatively easy to poke a pin through. These were printed with synch hole radius = 1mm, scanline hole radius = 0.25 mm. You can see the effect of material flow filling in. I'll do a second one with scanline hole = 0.5 mm and see how that goes.

nip2.jpeg
nip2.jpeg (83.36 KiB) Viewed 17073 times


nip1.jpeg
nip1.jpeg (55.07 KiB) Viewed 17073 times


In summary; yes you can print small holes, but their size depends on material and machine properties. You can fudge it by printing larger holes which print to smaller holes. Regardless of hole size, the hole LOCATION is quite accurate.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 1:24 pm

Nice work Andrew, they certainly look precise *to the eye*, of course the proof of the pudding... It would be great to see one in action down the track.

Edit:

BTW I suppose the bottom layer could be wider than the next etc to produce a cone rather than a cylinder, that would improve the viewing angle a little. It would be even better if the bottom could be "counter bored" but then you wouldn't have a layer on which to lay the top most layer(s) (you could, of course, do it the other way around but then you still have the problem of the bottom layer being "squashed".
gary
 

Re: 3D Printed 32 Line Mechanical Television System

Postby Harry Dalek » Fri Mar 10, 2017 2:02 pm

The pulley system that has to be adjustable.... to tight or to loose you will have speed problems..if he had made it thus the speed control would of worked better with he's speed control electronics .If you put the motor on an arm you can test this and lock in place when you have the correct position .
You need to watch the sync on a scope frequency meter you can tell if its drifting or not then.
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 2:21 pm

Prototype 2. The scanline holes were printed with radius 0.5 mm. Most of the holes are now holes. A few are filled-in. The close-up shows some of the scanlines and pretty much inconsistent size as expected, but I'm betting the centerpoint location is quite accurate. I'd be going over these with a small drill bit, by hand, and rounding them out to the same diameter.

nipb2.jpeg
nipb2.jpeg (72.49 KiB) Viewed 17071 times


nipb1.jpeg
nipb1.jpeg (55.36 KiB) Viewed 17071 times


Gary it would be nice to see how it looks in actual use. I'll send you a sample disc to try, OK :) ?
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 2:37 pm

Great! Can you send me one for my beads as well? Then I only need one LED :-).

Edit: I would want the apertures a little closer to the disk edge though!
Last edited by gary on Fri Mar 10, 2017 2:40 pm, edited 1 time in total.
gary
 

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 2:39 pm

gary wrote:Great! Can you send me one for my beads as well? Then I only need one LED :-).


So, two of the same? Not sure if you're making fun or not. If not, PM me your address and I'll do some for you.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 3:40 pm

Andrew Davie wrote:
gary wrote:Great! Can you send me one for my beads as well? Then I only need one LED :-).


So, two of the same? Not sure if you're making fun or not. If not, PM me your address and I'll do some for you.


No, not making fun, just laughing at my cheekiness in asking for another :roll:

The other thing is that all of my "set ups" these days are single led systems, great for bead disks, not so good for for aperture disks.

The bead version would have larger holes (I need to check my supplies but I think I have enough 1/8" beads) - but I could just as easily drill out the smaller apertures, it's just that I have a gut feeling that 1/8" apertures would be more accurate and it would be interesting to see if that is correct.

I will be in touch.
gary
 

Re: 3D Printed 32 Line Mechanical Television System

Postby Andrew Davie » Fri Mar 10, 2017 5:57 pm

gary wrote:The bead version would have larger holes (I need to check my supplies but I think I have enough 1/8" beads) - but I could just as easily drill out the smaller apertures, it's just that I have a gut feeling that 1/8" apertures would be more accurate and it would be interesting to see if that is correct.


If the beads are glass, then you could heat them up to about 200C and then just drop 'em in place; they would melt the plastic subtrate and possibly even weld into position.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: 3D Printed 32 Line Mechanical Television System

Postby gary » Fri Mar 10, 2017 6:45 pm

acrylic, they just sit in place with a force fit.
viewtopic.php?f=1&t=1179&p=9184&hilit=bead+disk#p9184
gary
 

Next

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 30 guests