Key Things

Zach Smith and Bre Pettis have created a web site called Thingiverse, which is designed to make it easy for people to share digital designs of real objects. I have put most of the things I have designed for printing on a RepRap on there. I even created one especially for it, a key for reading utility meters :-





You can get these free from the utility companies, but not a quickly as you can RepRap one for a few pennies worth of plastic, and they are easily lost. The files can be found here :- www.thingiverse.com/thing:88.

I also keep a gallery of all the things I have made with HydraRaptor here: sites.google.com/site/nophead/Home/hydraraptor/ThingsMade. They all now have links back to Thingiverse for the files.

A lot of the objects on Thingiverse are for a laser cutter, but I downloaded this twisted star box designed by Marius Kintel and printed it.





One trick I have learnt is that you can shrink objects a little by heating them with a hot air gun. When I first made this the top was too tight, so I shrank the bottom a little with the heat gun. It is now a perfect fit. You have to be careful not to get it too hot, or it will sag.

New Toy

The company that I have worked at for 25 years gave me a long service award recently. I could choose anything worth £500 so I chose this small lathe / drill / milling machine combo. The tiny watchmaker's lathe I have been using up to now is not really big enough for RepRap parts. This should be just about right.



A combo like this is a bit of a compromise and only recommended if, like me, you have limited space. It gives me a lathe, pillar drill and milling machine in a small footprint.

I also bought accessories with my own money, which came to about another £300 :-

• Cutting tools - essential.
• Compound cross slide - for cutting tapers.
• Drill vice - to hold things on the cross slide when milling or drilling.
• Tailstock chuck - to drill down the axis of round things.
• 4 Jaw chuck - for turning square and irregular shapes.
• Milling chuck and collet set - to hold milling bits.
• Wiggler - for finding centres and edges when milling and drilling.
• Die holder - for tapping threads.
• Headstock centre - for turning between centres.

I have also made a couple of accessories with HydraRaptor. The first is very simple: a t-slot cleaner for removing swarf from the cross slide's t-slots.



That saved me about £2.80.

The second object is the biggest thing I have RepRapped so far. It is a cover to go over the chuck to make it easier to turn by hand when tapping. It took over 7 hours to build and weighs 77g.





Here it is installed :-



I haven't used it yet but it feels like it should work well.

Sky Hook

I have found that the most convenient way to provide the filament feed is to throw the whole reel into a large hanging basket and suspend it over the machine with a ball bearing. The machine pulls the filament from the centre of the coil through a hole in the bottom of the basket. As it does this the basket rotates to prevent the filament being twisted.

When the basket contains a full 5lb or 2Kg coil then it does take some force to pull the filament from under the weight of it. Since I sharpened and angled my drive screw and moved to an HDPE filament guide I have not had problems with grip, so this works flawlessly.

I have even started stacking the baskets two deep to save on storage space for different colours of plastic. The top filament can be used simply by passing it through the centre of the lower basket. Obviously only one can be used at a time.



It was easy to attach a frame over HydraRaptor but harder to do over a Darwin. I decided to attach it to the ceiling instead with a "Sky Hook", as suggested by Adrian Bowyer here.



I designed a ceiling mount to hold a skate bearing. These seem to be the cheapest ball bearings to buy. My first idea was to put an M8 bolt through the bearing and attach a plastic hook, but I didn't have any bolts long enough, so I made some plastic parts instead.



The barbs just go through the centre of the bearing, but only when they are twisted over each other.



More by good luck than design, but that means that they can't be pulled out because they pull towards each other and meet in the middle before they can clear the M8 hole.

Just to make sure I made a small locking piece which slots in afterwards to keep the barbs fully splayed.



The bearing then drops into the housing and is held in place by gravity and the ceiling.



The eye and barbs are tensioned along their strong direction, so can be quite thin, but the housing is stressed in the direction where de-lamination can occur, so I made it 4mm thick. It would be tragic if it gave way and dropped a few Kg of plastic onto the machine from a height, so I tested it with a lot more weight to get some confidence.



These 13 wine bottles weigh 16.5Kg, which is 8 times more than the plastic I intend to suspend. It held up for 24 hours with no sign of stress to the parts. Using plastic ties with barbs on the end could be a useful technique for making RepRappable fasteners.

When I came to mount it on the ceiling I realised that the three screw holes were too far apart to all screw into the ceiling joists, which are only 1.25" wide, so I made a two lug version. I made this with 50% fill rather than 25% to ensure it was as strong as the three lug version.



Here you can see a trick I did to get support for the overhanging lip. I just put a one layer thick diaphragm across the hole and cut it out with a penknife afterwards.



Here it is installed : -

Black Heart

I made this tiny heart shaped box as a little gift for my wife's birthday.



Being such a simple shape, with no extruder off moves, it came out perfect, just like my wife.

Pot black

Having tried green ABS and found it a bit disappointing, I had a go with black ABS and was even more disappointed. I got it to try and make objects a bit more aesthetically pleasing but it was even harder to get working than green and has aesthetic problems as well.

The temperatures seem to be the same as those for green, i.e. higher than I have used for plain, but it is harder to get it to stick to the raft and it de-laminates more readily.

To make it stick I had to make the first layer a bit lower than normal. With plain ABS I can have it at the normal layer height above the raft (filament diameter * 0.8) but with HDPE and green ABS I had dropped it 0.1mm. With black I had to drop it 0.15mm as 0.1mm does not stick enough and 0.2mm sticks too much. 0.05mm makes all the difference and has about the same effect as changing the temperature by 10°C.

Another problem is that the extruded filament is not smooth. Close examination reveals that it has small craters in the surface where it has out gassed. It is particularly noticeable when laying the thick filament for the first layer of the raft.



It could be the effect of water absorption but, as all my ABS is stored in the same room, I think it is more likely a volatile component of the black dye. Another effect is that if the filament is stretched while molten, so that it is drawn into a fine thread, then it looks like a string of beads. I think that is an indication the black dye does not mix well with the ABS.

Fortunately, for some reason I don't understand the crater effect is less noticeable when making objects.



These are Darwin y-belt-clamps, a nice small part good for a quick test. The small one is made with 0.3mm filament and shrunk by 3/5. As you can see it is far less shiny. All I can think is that is related to the fact the filament is being stretched more.

Another downside of coloured ABS is that ABS turns white if it is stressed. This shows up far more when it is coloured and particularly with black. Because the base of the object is weakly welded to the raft then it gets bruised when it is peeled off.

This is the bottom of another screw topped pot.



As you can see it has white highlights where the welds to the raft have been broken. One way of fixing it is to wave a hot air gun over it to relax the stress points. That also flattened the base, which was a little convex due to warping.



I expect rubbing it with a little solvent like MEK would also solve the problem.

I now have a little family of pots!



I would not recommend coloured ABS as the dye introduces more variables and generally seems detrimental to strength and aesthetics.

Brassed off

My first attempt at an extruder had stainless steel bearings and a stainless steel drive shaft, more by accident than anything else. I wondered at the time how the bearings would last compared to the recommended brass ones. Obviously stainless steel is harder than brass, but brass should have less friction, so less wear.

The first drive shaft got retired because the bearing lands were off centre. At the end of its life the bearings and lands were still in good condition.



I ran it from 10/07 to 01/08 but up to that point all I had made was lots of HDPE test shapes.

I replaced it with a plain steel shaft that I bought from BitFromBytes. That had the big advantage of being solder-able.

It worked well for a long time but eventually the bottom land on the shaft wore down so much that the pump halves closed together when using undersized filament (2.7mm).



The bottom land has worn down from 3mm to 2mm. The bearings show a little wear but still have some life left in them. This is after running 5lbs of ABS through the extruder.



In the last few weeks I replaced the shaft with a new zinc steel one and switched to brass bearings in an ABS extruder with HDPE filament guide. That worked well until I noticed the pump halves closed together again. When I opened it up I found that the brass bearings had rotated in the ABS, but they had also worn down a lot, considering the short time I had used them.



The drive shaft lands are still fine though.



So it would appear that the best combination is stainless steel bearings and a stainless steel shaft, but I would have to find another way of attaching the nut.

Most metal bearings I have recovered from old equipment are bronze. I don't know how that compares to brass but it seems to be the thing to use.

Maybe it is time to look at ball bearings and an offset shaft like Ian Adkins' design.

Brain dump

I have added a page to the new Reprap Builder's wiki recording all I know about the original MK2 extruder. I make no apologies for including lots of links back to this blog and recycling some photos, but there are also many new ones.

Although the design has moved on, I have stuck with the original and tweaked it to fix the problems I have encountered. I don't think any of the later, more complex designs have better performance, except perhaps the anti-ooze version. That seemed like a nice idea but rapidly got too complex with a double nozzle and two heaters plus some tricky machining. I think instead stringing can be mitigated to a large extent by increasing the speed of moves between extrusions paths. I am hoping to go from 32mm/s on HydraRaptor to 150mm/s on Darwin. Also I think reversing the extruder motor a fixed amount measured by the shaft encoder, plus Enrique's oozebane and comb modules in Skeinforge should help a lot.

Spurred into action

Forrest's success at milling spur gears with Tommelise 2.0 got me wondering how well RepRap can make them with Fused Filament Fabrication. An obvious test case are the two gears on Ian Adkins' alternative extruder design. These are laser cut, but look just about achievable with FFF. I couldn't find a picture of Ian's, these are a similar design by Vik Olliver.



Here is how they came out :-



The infill does not meet the toothed edge very well but they seem very strong never the less. They mesh well even though the involute profile seems a bit rounded off.

In an attempt to get the infill to work better I tried running my machine with 0.3mm filament, 0.24mm layers. I didn't make the nozzle hole smaller, I just changed one number in my software so the filament is being stretched finer. The result was interesting :-



The infill fills the teeth and is also very flat. The hole for the motor shaft has come out smaller and the teeth are a bit asymmetrical. They lean clockwise, the direction the outline is laid down in. I think this is because I am extruding 0.3mm filament through a 0.5mm hole so the point it exits from can move about but 0.2mm. I expect using a 0.3mm hole will solve that.

I also had a go at making a smaller version of my screw top pot with 0.3mm filament. I used the same g-code, I just scaled all the coordinates by 0.3/0.5.



It came out well and the thread still works.

With the finer filament it takes about 3 times longer to make anything so I upped the speed from 16mm/s to 32mm/s. That had an unfortunate side effect: The wall of the pot is quite thin so after the outline and inner wall have been extruded, the infill is a very narrow zig-zag. That makes it a high frequency movement. Speeding it up made it more like an audio vibration which seems to resonate the frame of my machine, shaking anything loose off the desk.


HydraRaptor growling from Nop Head on Vimeo.

Going green

Getting a bit bored of natural ABS, which is cream coloured, I ordered some black and some green from New Image Plastics in the USA. Despite the high cost of shipping it still worked out cheaper than buying it in the UK. I chose green because it should be easier to photograph than black or white and it the RepRap logo colour.

I was surprised to find that it does not behave the same as the natural that I got from the RRRF. It does not seem to stick to itself, or to the base material, as well as the RRRF ABS did at the same temperature. I have raised all the temperatures by 15°C and, although it is better, I think it still needs bit more. It's appearance is more glossy, so that might account for it being less sticky, or it may just have a slightly higher melting point. I have no idea if the green dye has an effect or whether the base ABS is different. It will be interesting to try the black.

I made a new version of my screw top pot in order to perfect the thread before posting the STL files on the RepRap Objects wiki. It turned out to be nightmare to get it to print. Besides the problem of it not sticking to the bed, my extruder kept breaking GM3 gearmotors, but that's another story.



The ABS version came out about 1mm bigger in diameter than the HDPE version due to lower shrinkage.

I modified the thread to make it easier to start. The original thread started abruptly at the top of the pot and the bottom of the lid. I modified it in CoCreate by continuing the helix past the top of the pot and then milling it back to 2mm below the rim. I did the same to the lid's thread. That allows the lid to overlap 4mm before the threads engage, making it easier to line up. The thread now tapers out to a thin sliver to make it easy to engage. The result works a lot better than the previous attempt.

Screw Top Pot

A day before going on holiday I decided I needed a container to store the lens adapter for my camera. The lenses have caps on each end, but they require and adaptor which is a bit delicate: -



I knocked up an HDPE pot with a screw top lid that just the right size to hold it: -



The outside diameter of the adapter is about 48.5mm so I made the pot I.D. 50mm to allow some clearance. It actually shrank to 48.5mm, so it is a snug fit. Lucky it didn't shrink any more!



I made the thread using the helix tool in CoCreate. You draw a 2D profile and then use the helix tool to spin that round an axis, specifying the pitch. The dimensions were just a stab in the dark: I made the crest of the thread 0.8mm as that is two layers (Nyquist sampling theorem) and made the sides 1mm long 45° slopes, so that made the crest height 0.7mm. The minimum pitch with this profile would be about 3mm so I made it 3.5mm to give some clearance. I also made the the lid 1mm bigger radius so there is 0.3mm clearance from peak to corresponding valley.



A couple of things I missed which would make it easier to engage: -

1. The thread starts abruptly, but it should have a tapered lead in.
2. Both the pot and the lid should have a few millimetres with no thread to aid lining them up before the thread engages.

Despite this it works surprisingly well for a first attempt with an arbitrary profile and dimensions.



I am slowly homing in on getting rafts peel-able for HDPE. I made this with the first layer outline hotter than its infill :-

HDPE.raft_temp = 215
HDPE.first_outline_temp = 230
HDPE.first_layer_temp = 205
HDPE.layer_temp = 240

Most of it peeled with a little encouragement from a chisel at one side.

Bearing fruit

My wife has been asking me to make something to prop up the overladen branches of our dwarf apple tree for a few weeks now. I put it off while I was set up for ABS because I knew I did not have enough to finish my Darwin. Now that I have switched the machine to HDPE it is no problem, but it is now a few days late as one large branch has already snapped off!

We have lots of plastic covered metal poles so all I needed to do was make some Y-shaped end pieces. My first attempt went a bit chaotic while making the arms: -



I wasn't watching it but I figured it got too hot when doing the small pieces so I made the arms thicker.



Better but still very rough, it should look like this :-



I cleaned it up with a penknife and it was functional but it felt more whittled than extruded.



I made a couple more with even thicker arms but I was around to observe what was going wrong this time: -



When building the curved arms Enrique's software switches to 100% fill because it decides part of the layer is two layers from an outer horizontal surface, which a thin sliver down each side is. That would not be a problem in itself but because I have the infill overlap option set it ends up with slightly too much plastic on the 100% layers. As the height increases this excess builds up until the nozzle is actually submerged in the object while it is building it. Amazing that it manages to make anything resembling the correct shape!

What really needs to happen is that if the infill overlap parameter is set then the head needs to lay down the infill slightly faster so that the amount of plastic is still correct. I ran into the same problem with ABS when making an object with 100% fill.

I made a fourth version with the infill overlap set to zero and it was a lot better: -



Still very blobby but all the blobs are down to extruder overrun and easier to carve off. Overrun is worse with HDPE because it seems to be a more non Newtonian liquid than ABS. I.e. it compresses and expands more than ABS does, so when the extruder stops it oozes for longer.

I haven't tried anything to stop ooze yet. Simply stopping the extruder before the end of the line like the RepRap host does should improve it and is easy to do. Reversing the motor drive should also help. Simply stopping causes the extruder flow rate to fall exponentially but backing up a little should stop it completely in a finite time. The shaft encoder can then be used to go back at full speed to where it was before it backed up. There will still be some ooze without a valve but I think it could be a lot better.

Here is the final version cleaned up: -



And here is the tree with four crutches installed although only three are visible from this view though: -