RepRap: Blog

While searching for some useful gears and motors inside a buggered CD drive, I noticed a few things of note to RepRap. First off, they're using a rack & pinion to move the head on rails that look pretty similar to what we've got, just smaller. The rack and it's driving motor are anti-backlashed as you can see from this photo:



The anti-backlash is the springy thing inside the big gear, for those going "huh?" The gear wheel is split, and the two halves propelled in opposite directions. This pinches the gear on the neighbouring shaft, and cuts out any slackness in the movement. A similar trick is done with the rack.

They've also only used the inner quarter contact surface of their rails, which we might try to reduce friction further.

Keith, Ash and myself have mated the 3D axis assembly with the frame and dangled an extruder over it for the first time, to size up the gantry that we'll need to hold the extrusion mechanism:



After some discussion we thought we might try mounting two parallel rails over the top of the linear axis, and allow the head units to suspend themselves there. By fitting the head assemblies with adjustable feet, we'll be able to reposition each head individually to align with the centre of the turntable at the apropriate height.

In short, this proposal will allow multiple heads to be integrated at an early stage, and define a method of implementing interchangeable heads that can be exchanged with other users.

I informally suggest a gap for the rail spacing of 100mm with enough error to accomodate people who like to think in terms of 4 inches :)

What do others suggest?

The photo shows my freshly-made example of Adrian's extruder clamped by the now obsolete pinch wheel assembly. It is at least serving as a space model!

This may not be an original idea, but I haven't been able to find a reference to it.

Objects made by RP are anisotropic. In particular they tend to be stronger in the layer directions (say X and Y) than in the Z direction. This is particularly true for the extruded filament technique that we use. But it should be simple to program an RP machine to leave many small blind Z-axis holes in solids and then to inject-fill them using the write head when they reach the top of the solid. This should increase strength significantly, and also not slow the process down (the time taken to fill the holes should be less than that saved by not making them solid in the first place).

Just a thought.

This page as a useful disccussion on making RS232 and I2C interfaces work under the SDCC compiler on the PIC. Some of this will no doubt be useful to us:

http://hawthorn.csse.monash.edu.au/~njh/electronics/pic-tricks

Vik :v)

Robert at Webtec has kindly hosted the video of the Polymorph saddles being stress-tested. The videos are identical, but are spread around to balance the load (it's a 620K MPEG):

link1

link2

link3

In this video the saddles where the carriage contacts the rails, the brackets holding the rails down, and the attachment point for the belt drive to the carriage are all made from Polymorph and are held in place only by physical fit and bolts threaded directly into the Polymorph.

The recoil visible is not caused by the carriage slamming into the end of the rails, but by the motor rapidly reversing as the carriage lines up with the vertical bars.

I left this going for a couple of minutes, and nothing warmed up or fell off. It won't be going this fast in the final thing, and a sturdy frame is under construction by Keith.

Vik :v)

Based on the shortcomings of the Meccano turntable, I've drawn this proposal for an auto-lowering turntable. It relies on using 3 vertical pieces of M5 studding driven equally by a gear train. Play in the gear train should not be significant for the purposes of raising or lowering, but the fit of the initial worm drive onto the turntable shaft will need to be a good one.



As the centre of the turntable rotates, the M5 studding rotates at 14.4:1 . As the thread pitch is 0.8mm, every revolution causes it to drop relative to the static nuts approx 0.06mm. So it can rotate back and forth through one revolution to do the fabrication, then rotate twice to drop 0.12mm and build the next layer.

Not sure what to do for bearings yet. I might smooth things off, or I might pack the studding into cheap skate bearings.

I learned a lot about gearing designing this, but feel free to enlighten me further :) Worst case, we might have to use separate motors but this would be simpler.

Vik :v)

If RepRap could use the aliphatic polyester polylactic acid (PLA) as its primary working polymer that would be good. The advantages would be that it:

1. Has a low melting point (150 C; glass transition: 60 C),
   
2. Is mechanically pretty strong [yield: 70 MPa (nylon: 80 MPa); modulus: 3.2 GPa (nylon: 3 GPa)],
   
3. Is biodegradable, and
   
4. Can be made from biomass.

Actually, as Vik and I have pointed out before, having a polymer that is biodegradable is a disadvantage, as it means that products made from it don't lock up atmospheric carbon (but all these recycling directives that insist on it are written by politicians, so what do you expect?). But if the polymer comes from biomass, then at least the whole process is carbon-neutral.

In the case of PLA, the polymer can be made by the fermentation of corn (maize), a crop that will grow in poor soils in a very wide range of climatic conditions. This allows the possibility of having the RepRap machine make its own fermenter, and then having a manufacturing process for high-technology goods that could be bootstrapped straight from an agrarian base with very little capital expenditure. Rich World look out: here come the African farmers :-)

A good source of information on PLA is Rafael A. Auras's Powerpoint file available here.

A colleague of mine at Bath is working on the stuff too, and has promised me a sample. So experiments on it will be blogged in the near future...

P.S. - an excellent paper on polylactic acid production by Narayanan et al. can be found here.

Ash and myself have assembled our first rod-based slide, using the Polymorph saddles described earlier. I've also made a set of unsplit Polymorph clamps to hold the rails, held in place with Meccano screws (dunno what they are, but they're close to M4). We'll see if they need grub screws later - they're easy to melt in with a soldering iron and tweezers. Meccano came to the party and donated a Centenial Crane Kit to the cause; though it had no gears in it there were plenty of long Meccano bolts and little pulleys.

A two-part Polymorph clamp is used to hold the drive belt to the carriage without damaging it, and the clamp is attached to one of the first FDM brackets Adrian made for me. The 150g assembly can be seen here, dragging an additional 800g:



With an additional 900g payload, the drive belt starts to slip occasionally. So we can set our rough estimate for the maximum mass of the carriage assembly as 1050g. I'll whack off 250g for errors and safety margins for the moment, and figure on a total carriage mass of 800g.

This carriage, when fitted with a turntable, looks like weighing about 750g. This only leaves 50g for the payload - our extruded artifact. But Meccano is a lot heavier than Polymorph, so we'll see how we go with the gradual replacement of parts. Besides, 50g of Polymorph actually goes quite a long way.

The switch on the side of the carriage is for later trials. We're going to see if we can get it shooting back and forth at maximum speed, and then we'll see if anything warms up or falls off.

Vik :v) & 4sh

The first tests have been done on the Polymorph carriage, oiled lightly, with the feet trimmed to 75% of the circumference and de-stressed with a hair dryer. Smoothly overcoming the friction between the four saddles requires a force of 2.6N and 1.2N keeps it moving:



The carriage itself weighs 140g and I started to wonder just how much power we're going to want to expend on moving the carriage around at this point?

As an example, the tiny little 3V-6V Meccano motor I have here consumes 200mA startup current at 3V and will move the carriage. Assuming that's around 50% efficient, thanks to the gearbox, and a load on the carriage that is expected to be about 300g, the dilligent student should be able to work out the maximum speed and acceleration. It's been a good few years since I did physics at school...

Would anyone like to tell me:

a) The likely top speed.
b) What's the accceleration, and
c) Is this good enough?

Vik :v)
PS Spot the cameo appearance of the FDM'd pinch wheel assembly.

Having seen Adrians wondrous system using the pinch bolts - and incidentally having just got hold of my pinch-wheel design out of the Stratasys - I've had another bit of activity in the brain cell. Why not use the bolt thread?



Trap a Polymorph rod next to a bolt in an enclosed space, turn the bolt and keep it steady - Polymorph moves along the thread. There's an inherent high gearing and ratchet from using the thread, but we might need to put a bit of spring-loading (easily FDM'd) in there to hold the Polymorph firmly against it and make up for any imperfections in the feedstock.

Also, the bolt can then hold a drive gearwheel, keyed onto a nut (or two if you want to be really sure, and want to locknut it without compressing the gear). The stress on the gear wheel might be low enough to enable the use of extruded plastic gears.

No prototype photos yet - I only had the idea this morning. Funding permitting (Suz is still very budget-conscious) I might get a cheap electric screwdriver on the weekend and see if it can drive the thread directly.

Vik :v)

For the far future of the RepRap, we might be able to make our own microcircuitry in a way that borrows principles from the current silicon manufacturing technologies.

If we have a photosensitive organic semiconductor ink that hardens on exposure to light, and project the image of a circuit (possibly using a PC screen or some other emissive display) onto a substrate coated in it, then the conductive polymer is fixed in place. Wash it and put a layer of another ink on, expose that to a different set of tracks, rinse and repeat until the circuit is complete.

RepRap should enable the production of equipment that can automate this process. Like I say, far future, but we have to get these ideas down in case some antisocial tries to patent them...

Vik :v)

I've done a different design to Vik's for a feeder head to go with the polymorph extruder in the blog of Saturday 4 June. Mine's more complicated than his, but is a bit stronger. Here's a close-up of it working:



It consists of two pinch wheels driving a 3mm diameter polymorph rod into the top of the PTFE tube (see the previous blog). The pinch wheels are 40mm M4 cap screws being turned by gears at the back. The knurling on the cap screws gives a good grip on the polymorph, though it still slips a bit, so this is not the final design. However, if you want to download the STL file for it (1.6MB), click with the right mouse button here. Here's a close up of the feed mechanism:



The cap screws run in bushes embedded in the head. The two M6 screws that you can see the heads of on the right serve to clamp the pinch-wheels together, and the one half out of the bottom of the picture clamps the PTFE tube of the extruder. Ignore the blue scribble - that's just me making notes to myself...

Here is a view of the gearing at the back:



Just to the left and above the small gear you can see the shiny end of the bush insert.

The whole device works quite well, but needs a few improvements. These are:

1. Double or triple up the pinch wheels in a totem-pole up the polymorph rod to reduce slippage; they're only cap screws with gears on the back, so that's not going to cost much more.
2. Use Vik's much better solder nozzle design (see the blog for 8 June; improved further by his making it detachable on 11 June). If one used the higher temperature solder that would allow the polymorph to be worked at a higher temperature. One could also bury a loop of one end of the heater nichrome wire in it and use the brass extruder barrel to make one of the heater connections. This would automatically make a thermal fuse to prevent temperature getting to the point where PTFE starts to decompose (250 C). The problem Vik found with the solder not wetting the brass may resolve with additional flux. More experiments needed...
3. Switch to steel or brass gears. The nylon ones are plenty strong enough, but I had to drill and pin them to the shafts. It would be much simpler to file flats on the shafts and to attach the gears with grub screws. This'd need metal gears to hold the grub screws strongly enough.
   
4. Fit two optical sensors, one to measure pinch-wheel speed, and one to detect when the polymorph rod runs out so the machine can go fetch another.
5. Control the whole thing (including the extruder temperature) with a local PIC, rather than by hand...

Here's the parts list:

1. The RP part from the STL file,
   
2. 1 off: 40mm M4 cap screws with 20mm unthreaded shaft at the top,
3. 1 off: 30mm M4 cap screws with 20mm unthreaded shaft at the top (or saw 10mm off Part 2 above...),
4. 5 off: thin (2mm) M4 nuts,
   
5. 2 off: 25mm M6 cap screws,
6. 1 off: 30mm M6 cap screw,
7. 4 off: M6 nuts,
8. 4 off: M6 washers,
9. 1 off: 35mm M6 brass screw,
10. 2 off: 5mm M3 countersunk screws (for attaching the motor),
    
11. 45mm x10mm diameter PTFE rod,
12. 400mm 0.2mm diameter nichrome wire,
13. A reel of PTFE plumber's tape,

Plus these parts from RS Components:

1. 1 off geared 12v motor - 245-6118,
2. 1 off 80-tooth gear - 745-270 (this needs an internal diameter of 4mm, which it doesn't have so you'll need a sleeve to make it fit),
   
3. 3 off 20-tooth gear - 286-2355,
4. 1 off thermistor - 484-0149,
5. 4 off bushes - 262-1939 (two in each hole).
   

My friend Keith was round, kindly offering to make up a wooden base for the new prototype here in New Zealand. It was during the process of explaining what we wanted it for that I tried to demonstrate a carriage mechanism, and walloped a saddle together out of Polymorph that sat around a ground steel bar. It worked beautifully. The low friction and close fit formed the perfect saddle on the carriage guide bar.



The trick was to slit the underside of the encircling Polymorph so that it would slide freely on the bar. Chopping out a bit of the centre section further reduced friction. As you can see from the photos, we also melted a Meccano bolt into it, and can remove it to attach various fittings. The insertion was best done by warming and pushing the bolt in with a soldering iron.



This method has the added advantage that, as long as the carriage bars are parallel, it does not really matter what size they are. Polymorph supplies permitting, we'll try to build an entire carriage mechanism based on Polymorph and Meccano - which we can replace with manufactured components as we progress.

We've also produced a detachable 0.32mm solder nozzle made from a couple of bits of M6 bolt and an M6 nut. The nozzle part is soldered to the M6 nut, and PTFE tape ensures a leak-free join onto the main bolt. We did use mostly lead-free solder here (melting point 220C), but it would not tin the brass and we had to resort to 60/40 solder to do that.



Vik :v)

This could be useful: an amorphous (i.e. non-crystaline) metal alloy with a glass transition temperature of only 68 C. It was developed by Wei Hua Wang of the Institute of Physics in Beijing. See here for details.

I had a successful run on the 'Afghan Lathe', and managed to bore a beautiful 3mm hole in an M6 bolt, right down the middle. To avoid the use of fine drills for the nozzle, I tried out an idea we'd had on using solder. I propped a piece of 0.32mm nichrome wire vertically in the hollow brass bolt - I'd only drilled the last 2mm out to a 1mm hole on that end - and soldered around it. When the solder was cold, I pulled it out with pliers to leave a 0.32mm hole in the solder. Nichrome is very handy here as it can't be wetted by standard solder.



Tests with a syringe show it squirts water in a straight, true jet. If it works with Polymorph, it'll be a very simple way of making nozzles of varying sizes. This first one will be incorporated in a Polymorph extrusion nozzle similar to the one Adrian just made, and we'll see how well it ejects Polymorph.

Adrian is currently printing the first draft of the pinch wheel feed mechanism for the extrusion head. It consists of two interlocking pinch wheels that push the 3mm Polymorph rod down a guide and into the opening in the 10mm PTFE rod:



Small holes in the pinch wheels can take metal points if additional traction is required, and the symmetrical clamp should hold the PTFE rod centred.

The strange lumps near the axles of the pinch wheel are where ArtOfIllusion seems to fall over when trying to poke a hole using its boolean operator. It was much faster to add drill centring points and drill manually than fix the obscure bug! I did actuall try to fix the bug, but no joy. Still, onwards and upwards we go.

Encouraged by my previous polymorph extrusion experiments, I designed a much simpler extruder. Though I used a lathe to make a part of it, with care it ought to be possible to build this design using just hand tools as the only operations were drilling holes and tapping threads (which I did by hand anyway).

I started with a brass M6 countersunk screw:



I cut the head off, and drilled a 3mm hole right down the middle:



One could instead use M6 brass studding, or one could cut a thread on the end of a piece of brass tube (only the last 5mm or so actually needs to be threaded, as will become apparent, though the rest of the thread is useful to guide the heater wire, as will also become apparent). I drilled out the end, tapped it M4, and fitted the nozzle described in the blog for 28 May (q.v.).

Then I wrapped some PTFE thread-sealing tape (obtainable from all good plumbers' suppliers...) round the thread...


...and wound a 300 mm length of 0.2mm diameter nichrome wire into the PTFE-insulated valleys of the thread as a heater. The PTFE needs to be thin to allow heat conduction, but thick to prevent shorting with the brass. About two or three thicknesses seemed to work well. I then put a thermistor (RS 484-0149; cut and paste that part number into the box top-left on the RS main page) against the thread at the nozzle end and wound more PTFE tape round it to keep it warm.

I drilled a 3mm hole down a 40mm length of 10mm diameter PTFE rod, then drilled out and tapped M6 the last 5mm or so:



I then screwed the whole assembly together:



The thermistor is the bump at the right hand end; this was positioned past the end of the heater windings to get temperature readings from the brass, not the nichrome heater wire; before putting the thermistor on I cleared a patch of PTFE away using a wall file to expose the brass, so the thermistor is resting right on the metal; I wrapped a little PTFE tape round the thermistor leads to stop them shorting. The heater wires are the ones heading off top right.

I rolled up some 3mm diameter polymorph rod, and switched the heating current on. 300 mm of 0.2mm nichrome wire has a resistance of about 12 ohms, and I was running it at 12 volts (i.e. 1 amp / 12 watts). This brought it up to temperature (about 130 degrees C, or a thermistor resistance of about 350 ohms) so quickly (less than a minute) that I had to turn the voltage down to prevent overheating. This is good, as it means it should be easy to do PWM temperature control with a PIC and a power transistor.


The multimeter on the left is measuring the thermistor's resistance, and the bench P.S. on the right is supplying the heater.

I pushed the polymorph rod through. A force of between 20 and 30N gave extrusion at 2mm per second.



The PTFE rod stayed nice and cool, and should be easy to clamp into the pinch-wheel feed that Vik is designing. It may be necessary to put a hot-air heating jacket round the extruder, not to keep it hot, but to direct hot air down to soften the polymer that has already been deposited to get good adhesion.