Tuesday, December 6, 2016

Curta 3D printing source files released

This is just a quick post to say that I have released a BOM and the STL files to allow you to print your own Curta! I also made the CAD source publicly available in OnShape in case you are interested in those.

I have not yet released build instructions which are pretty vital to a complete solution to building your own Curta. Those were taking a significant amount of time and are not near done yet so I decided to go ahead and release the files for those intrepid among you who wish to make a go at it anyway.

The BOM includes a list of tools and non-printed parts as well in case you want to pad your Christmas list.

I also want to thank my wife, Adam Savage, The Charlotte Maker Faire, /r/3DPrinting, and the thingiverse community for inspiration and support along the way.

Look forward to one more part to my build as I finish and paint the Curta!

Monday, August 29, 2016

3D Printed Curta Part IX

Completing Assembly of My 3D Printed Curta

Completed Curta on my assembly workspace
I applied for and received an invitation to the Charlotte Mini Maker Faire! I look forward to seeing readers of my blog and other makers there! If you're in the area, please stop by and tell me you've seen my work online.

After having completed the main portion of the Curta, I started to focus on the upper carriage portion. At first I thought this would be a breeze as it is mostly the digit dials which spin on axles connected to the carriage casting. I encountered multiple issues which I will detail below.

Digit Axles

The digit axles were one of the easier prints out of the carriage assembly, but I did end up increasing the infill because the axles broke off in the carriage casting a few times when I put too much pressure on them. The main time this happens is when screwing the carriage ring onto the carriage cage.

I printed these with the wider section down on the print bed with a small amount of support because they have a cutout that helps them sit in the carriage cage. I ran them through a rough sanding to smooth them out some and then sprayed them with the PTFE lubricant I have been using.

Digit Dials

The digit dials were pretty straight forward and almost as easy as the digit axles. They did require support for the geared portions. I printed them with the numbered dial downwards and the geared portions facing upwards. There were two types, some with a one-piece geared segment and some where it was two pieces. This is so that the teeth on the clearing ring don't interfere with normal operation of the Curta while the clearing ring is in one of its resting positions.

The only issue I had was that the hole for the pins that depress the carry levers were so close to the edge of the dial that it caused the outside diameter of the dials to not quite be a complete cylinder. I also sprayed these with the PTFE lubricant. I tried to keep the lubricant on the inside of the shaft only.

Carry Pins

The carry pins exert pressure on the carry levers and I didn't want to have problems with them breaking off so instead of printing these vertically, I printed them horizontally. Due to their small size and their function, I printed them at 100% infill. I printed most of the carriage parts at 0.1mm layer height, but I took special care to print these at that layer height so they were rounder when printed horizontally. A small bit of filing was necessary to fit these. I estimated the distance to insert them into the digit dials and then later fine tuned it when the dials were placed onto their axles and seated into the carriage cage. 

Clearing Disc

The clearing disc was interesting. I printed it upright which meant lots of support underneath it, but I wanted to keep the top surface clean. The difficulty is in the groove for the decimal markers. The Simplify3D support didn't work as well printed on a slanted surface.

I printed this with 30% infill at 0.1mm layer height because it is an outside facing part. I want the sanding and filling to be easier later prior to painting.

Carriage Casting

The carriage casting was a real trick to print. This one had me stuck for about two months. It wasn't terribly difficult to get the initial print done: I printed it upside down to avoid support material. The problem was once I added in the digit axles and digit dials, it didn't align properly with the lower portion of the Curta. The digits would seat onto the transmission shaft crown gears, but they would not align with the carry levers properly. The alignment problem was also obvious from the spider spring whose fingers did not perfectly match up to the holes on the top of the casting.

Here the misalignment is visible by looking at the spider spring. It's not much, but it is enough to prevent the number dials from triggering the carry levers properly.
Another view of the misalignment on a segment of the carriage casting printed in black. Notice that left and right the spider spring matches up, but top and bottom it does not.
I checked my printer, my models, and even the slicer (by trying another slicer). I also put in a support ticket with Simplify3D to see if they could find a problem with what I was doing or with the software itself. They were very helpful, but couldn't point me towards a definite cause.

Finally, I flipped the model over and printed it with the support I was avoiding. Something about the rotation re-aligned the ball bearing holes and digit axles to the rest of the Curta and to the spider spring. At first I thought rotating the model to eliminate support material caused some rounding error in the model, but after reviewing the CAD I realized that I had designed the part in the same orientation I originally printed it (there was no rotation for the original print). If anything, it seems like any rounding error encountered when rotating the part upside down should have thrown the reprint off further.

Finally, everything matched up. More importantly the number dials matched the lower portion of the Curta so that the pins on the number dials would contact the carry levers.
At this point I'm not 100% sure why flipping the part over helped, but I'm glad I discovered the workaround.

Carriage Cage

The carriage cage had the same issue as the carriage casting as far as alignment goes. It also had the same solution. However alignment wasn't the only issue I had with the carriage cage. The engineering drawing was somewhat difficult to interpret so it took a few iterations to get the right output. The threading on its lower outer edge where the carriage ring connects also required an extra iteration. The carriage ring didn't thread on nicely at first and I had to decrease the diameter of the threads about 0.2mm before it would.

I printed the carriage cage at 30% infill with a 0.1mm layer height. The original print required no support, but the reprint to correct the alignment issues required a lot of support.

Carry Levers

After getting the carriage assembly mostly complete I connected it up to the lower portion of the Curta to try out some simple carry operations and found that the carry pins on the number dials now met the carry levers, but did not press the levers enough to trigger them.

Initially I worked on fine tuning the carry levers to smooth out their action and improve their positioning. I was able to get many of the levers to carry properly, but there was a very small window of alignment where they would both trigger a carry and reset afterwards. For some of them the window was so small that simply triggering a carry and reset would push it out of position far enough to not work (this was on the order of 0.1mm or less). Another fault I encountered was that sometimes the tens bell would not push the lever high enough to prevent friction from causing an errant carry during the next operation.

When I realized just how finicky they were, I went back into CAD and altered the levers to be just a little bit taller. This way the levers would trigger more easily and could be adjusted a little bit lower so that the tens bell could engage them more completely to reset their original positions. That change helped significantly, but I had to adjust the length of the levers more than once before I found the sweet spot in size. If the lever is too tall it can interfere with the number dial's ability to turn during the carry operation (and potentially break something -- like the main shaft).

I think I printed each lever about three times. They were printed at 100% infill in a vertical orientation. Each one (there are 15) had their sides and edges filed to smooth them out. The final round of recalibrating the levers went significantly easier because with a better height, they were less finicky about their position. I did find I had to adjust the bends in the carry lever springs after extending their height.

Breaking the Main Shaft ... Again

I broke the pin next to the main shaft (which drives the tens bell) while testing the carry levers. I reprinted the main shaft, but during the long print I decided to experiment with using resin to repair and strengthen the part. I mixed some resin and hardener and poured some into the hole where the pin broke off. I then poured some more resin into the pin itself at the break point. Since printed PLA isn't water-tight, the resin seeped throughout the part. Using a piece of toothpick to hold the parts together, I let it cure.

Disassembled in preparation for repairing the step drum.
I believe the main shaft got stressed when the tens bell pin broke off. Shortly after repairing the tens bell pin, the main shaft broke during normal operation. I tried swapping the tens bell out for the one I printed when the tens bell pin broke, but I had accidentally printed an old copy which had its teeth misaligned (see Part VII). Instead I repaired the main shaft in the same way I repaired the tens bell pin. When I get more filament I will reprint with the correct model.

As with last time, I will print the main shaft portion of the model with 100% infill, but this time I will also extend that solid infill to the top few layers of the step drum.

Other parts

I also printed the spider spring and spacer parts, clearing cover, clearing ring, clearing plate plunger, clearing tooth racks, the rivets for the clearing ring, the main bearing sleeve, the collar, the carriage ring (the knurled part you grip), and the lower housing. These parts all printed pretty easily and assembled easily. For the most part I stuck to 0.1mm layer heights and 30% infill. The plunger was printed at 100% infill and printed horizontally for strength. It required some filing / sanding to round it out some.

Curta Function

The Curta now functions well. I have been able to run different operations, but I haven't been able to do much actual math without numbers on the results dials. I plan on printing some temporary paper wraps for them soon so I can record some videos on it.

The carriage does lift and rotate to allow multiplication and division by tens. The clearing ring rotates, but I have noticed that it is tough to turn and doesn't always leave every digit on zero. I'll have to work on tuning that. Subtraction seems to work beautifully, but with more friction than addition. The reversing lever works nicely as well.

Two printed parts just weren't living up to their intended functionaly. These were the spider spring and the stepped drum positioning spring (it connects to the bottom of the tens bell). I figured I might have some trouble with those parts so I had already been working on a plan to beef them up to improve their springiness in plastic.

For the stepped drum positioning spring, I thickened up the arms and added a half-cylinder to each leg to increase rigidity. That did the trick quite nicely -- the step drum now pops up and down between the addition and subtraction positions. I reprinted this the same as the original at 100% infill on its side with support material.

For the spider spring, it has a couple of washers that sit over it. I combined those with the spider spring to increase its thickness and tapered off the fingers to the original height. That stiffened up the design, but I also decided to increase the infill from 30% to 50% with the intention to increase it further should I continue to have issues. So far it seems to be working quite well. We'll see how it does over time.

So far these are the only two parts I have redesigned. I have combined parts and divided parts in order to get prints to work more easily, but these two parts have actually changed form slightly due to the attributes of the material and manufacturing process.

Video

Update: Here is a short video demonstrating a simple add and subtract 2

Curta Logo work

While I was doing all that work (particularly in the two month period I was stuck on the carriage casting alignment issue), I got a head start on the numbering and lettering on the Curta. I searched quite a bit and couldn't find a font which matched the text on the Curta exactly.

I ended up deciding to use OnShape to actually CAD the text. I worked from the engineering drawings which had some dimensions, but not all of them (the curves in particular were missing dimensions).

When I was done, I wanted to know just how close I came to the engineering drawings I was working from so I exported the CAD sketch as a .dwg file. I used a section of the engineering drawing and the exported sketch to create an overlay. The CAD sketch was only scaled (uniformly) and positioned. Neither the letter spacing nor the aspect ratio were altered. Below are the results which came out pretty darn close.

Curta logo from the engineering drawings
Curta logo I designed in CAD largely eyeballing curves
Overlay of CAD designed logo over logo from the engineering drawings

Next Up -- Paint (Suggestions Welcome)

I plan on painting the Curta -- it isn't very functional without being able to see the math being done. Most of the parts needing painting are external parts where changes in dimensions don't matter, so I have that going for me. However I also have moving parts and parts that will need quality lettering and numbering. My current plan is to use a laser cutter to cut masks out of masking tape for the lettering. I don't know what type of paint to choose either. I am open to suggestions here as painting isn't something I have a lot of experience in.

More Curta pictures

Nearly ready to fully assemble.
Bottom of the Curta (minus the base plate) after assembly.
Curta minus the lower housing and crank
Another view of the Curta minus the lower housing and crank
The carriage with ball bearings ready for the spider spring.
Note the carriage spring and associated parts above it.
The carriage with the spider spring in place.


Clearing ring with the clearing teeth installed.
I printed those in much the same way as the Curta was made.
They are created flat and then curved and pressed into the bottom of the clearing ring.
They had a good bit of friction, but I ended up using super glue to keep them in place.
The main part of the Curta with the collar added with M3 screws.
I did have to widen the holes slightly for the heads of these screws.
Close up of the top of the Curta fully assembled.

Selector shafts on completed Curta.

Reversing lever on completed Curta.
Top view of completed Curta.

Thursday, April 28, 2016

3D Printed Curta Part VIII

Transmission shafts

Printing

I printed the transmission shafts in groups. There were three different shafts some with with grooves at different heights along the shafts, but most had no grooves. The grooves allow for clips to keep a gear or rotation lockout at a certain height.

I printed all shafts at 80% infill. I also printed them combined with their crown gear tops which added supports, but eliminated the need for a tool to help me offset the crown gears properly. If I were to do it again I would try printing them separately, but it might be better the way I did it. The crown gear tops were kind of fragile -- I had to reprint a few transmission shafts due to breaking them off.

Fitting

The transmission shafts needed some sanding to smooth them out. The holes in the upper main casting and the bearing plate also needed their holes widened for a sliding fit to allow for easy rotation. To widen the holes I used a set of needle files I got at my local hardware store. The Amazon link is to the set I found. There are cheaper sets on Amazon, but I like this set since the files can be used without the handle which helps for tight spots.

You might notice in the top-down picture in the next section below that some of the shafts look different from the top than others. When I initially scaled the parts up, the offset of the shaft into the crown gear top did not get scaled alongside the parts so the shafts were not inserted into the crown gears as far as they should have been. That wasn't a problem except that the shafts were now too long. Since FDM printed parts are weakest along the joint between two layers, I was able to measure out the extra length, then mark and snap off the extra smoothly at the without having to reprint them.

Count gears, lock gears, and sleeves

Once the transmission shafts were printed and in place, I printed the gears needed for the transmission shafts. They are supposed to be press fit together on sleeves that slide fit onto the transmission shafts.

Press fitting might be necessary for machining, but I printed them as one piece to avoid press fitting small pieces that would likely break in the process. That did mean support material, but it turned out well on these small pieces to use Simplify3D's supports without a higher density interface layer.
The first group of gears I printed came out nicely, but I had to do a lot of work to fit them onto the transmission shafts. To reduce that, I modified the sleeves to widen them a small amount. It was easy to do them all at once in OnShape because I based them all on one sketch.
Transmission shafts added and locking ring in place

Transmission gears installed

Lubricating

Once I had the transmission shafts gears assembled into the frame, I manually positioned the gears and test ran the Curta. It worked well, but plastic on plastic was squeaking.

To resolve this, I used a spray-on dry PTFE lubricant. I disassembled the entire Curta and lubricated each part that spins or slides against another. I didn't lubricate everything because certain parts of a Curta should not be lubricated. The results were incredible. It eliminated the squeak and an amount of friction I didn't know was possible to remove.

Tens levers

Printing


The tens levers were a trick to print. They are normally formed from sheets of metal that are cut and then bent.



I tried printing these lying flat, but the bends cause weak sections where the layers have a small cross-section of adherence.

I ended up printing them upright which came with its own challenges. Upright they have a small footprint so keeping then stuck to the bed was an issue. I first tried lining them up along the x axis to reduce y axis movement and vibration. That did not work -- they still fell over. My next attempt worked. I turned them aligning them along the y axis which gave them batter adhesion along the moving axis.

I also printed the carry lever bearing blocks they require a small amount of supports. There may be a way to re-engineer them to eliminate the need, but it was easy enough to print and remove the supports. Simplify3D showed it's value here and throughout the project.

Fitting

The carry levers did require some filing to fit their bearings, but that isn't too surprising given the narrow parts. The amount was minimal and I got pretty quick at it once I knew which areas needed work to get smooth movement.

In addition to needing to fit the levers into the bearings, the bearings needed fitting into the upper main casting and custom springs needed to be made to support the proper motion.

Carry Lever Springs

The springs have their own engineering drawing depicted below. In order to get an accurate shape consistently, I decided to design and print a jig to make the springs. I used a music wire near 0.6mm in diameter I found at a local hardware store. The jig is simply a block with a few holes in it with a diameter near that of the music wire. The holes are of the lengths necessary to make bends at the desired offsets. Music wire will spring back from the bend some so I used a pair of needle nose pliers to finish off the springs.

Adjusting; Eliminating bend


The first spring I made was scaled directly from the engineering drawing. It was a little bit too long to properly snap the tens carry lever into position.

I altered and reprinted my jig for the springs and had a better length, but I had problems with the legs of the spring catching on the shoulders of the tens carry bearing. That problem was solved by eliminating the 10° bend at the opposite end of the spring from the legs. I guess scale changed the requirements of the spring. 

Calibration

Once I could crank out springs that would consistently work for the carry levers, I had to actually calibrate each carry lever. The tens bell has a ramped section which resets the carry levers from their lowered carry position to their raised no carry position. The springs help keep the levers from being between the upper and lower positions, but each lever will have slightly different friction and each spring will have slightly different pressure. This means that the amount the tens bell has to press upwards on a carry lever to get it to reset may vary a little bit per carry lever.

In order account for that variation, the carry lever bearings allow a range of height positioning in the upper main casting. Each carry lever is adjusted height-wise until it can easily be triggered then reset by the tens bell and the lower position aligns the carry gear with the tooth on the tens bell.

This step took some time and patience. I had to adjust everything multiple times to get them right. Once each carry lever was working properly, I tightened down the screw next to it to prevent it from shifting.
Tens carry lever install complete

Selector shaft assemblies

Printing -- knobs, shafts, guide screws, bearing pins, and plates

The selector shafts are comprised of multiple printed pieces. There is the shaft itself, the knob, a guide screw which facilitates the shaft's rotation, a bearing pin, and a couple of plates which hold a set of four shafts on the Curta.

I printed the shafts in two parts in order to reduce the support needed and to avoid having to force fit the number roll onto the shaft. I added a key to the two parts in a way similar to how I did on the main shaft in order to make sure they mate properly. I printed both at 100% infill. I also added a raft to prevent them from falling over during the print.

I considered printing the selector knobs and guide screws together as one piece, but decided to try printing as separate pieces and cutting thread into them. I printed the selector knobs at 30% infill and the guide screws at 100% infill. I also printed the ones, thousands, and millions digit selector shafts in red. The rest were printed in black.

I printed the bearing pins at 100% infill for strength (the pin tips are narrow and must keep the selector shafts in place) and the bearing plates at 30%. There wasn't anything special with these pieces -- they printed well with no special settings needed aside from infill settings.

Threading

Each selector knob needed thread to accept the guide screws. Each plate also receives two M4 screws to hold it in place. The bearing plate needed threading there to accept the M4 screws. The holes were easily threaded. The guide screws were more difficult. I sheared two off inside the die before I figured out I was using too much pressure and moving too quickly. It also helped to turn the die backwards more frequently to clear out the shavings which helped eliminate unnecessary force.
Tapping the selector knob
Threading the guide screw
The guide screw with thread

Engineering drawing error

After completing my first selector shaft print and adding it to my Curta, I realized that it did not align the transmission gear to the step drum properly. I looked around the internet at pictures of Curtas and at vcalc.net's disassembly images and came to the conclusion that at the zero position, my selector knobs were aligning too low by about 1.2mm.

I wasn't sure at first why the knobs were too low. It could be cummulative error in the sizes of the parts of the curta causing misalignment or a mistake in one of the models I made or even a problem with the engineering drawings (I had already seen one in Part VII with the main shaft alignment).

In order to figure that out, I got measurements made directly off of a real Curta by Cody Brocious who is working on a project to fabricate a 1:1 scale Curta by machining the parts. His measurements showed me that the distance from the top of the selector shaft to the first divot where the selector knob sits was different on an actual Curta from the engineering drawings by ~0.4mm which corresponds to my misalignment of 1.2mm when scaled up to 3:1.

In OnShape, this was an easy fix -- I altered a couple of parameters to the model and was able to print copies of the selector shaft with the correct dimensions.

Assembly

I assembled the selector shafts using springs from McMaster Carr and #TT sized steel balls from ballisticproducts.com using the method I described in an earlier post. One notable difference is that I added the use of the PTFE lubricant.

Assembled selector shafts

Reversing lever

Threading

I threaded the top of the reversing lever shaft in much the same way as I did the selector knob guide screws. I had more trouble keeping the shaft normal to the die than I did the guide screws. I was concerned that this might cause the threads to be torn off as the shaft turned deeper into the die, but that fate was avoided. The M4 nut that holds the reversing lever on doesn't sit perfectly flat, but it seems to do the job.

Assembly

Assembling the reversing lever and shaft went smoothly after having done the same for all of the selector shafts. I used a narrow hex wrench to press the ball bearing and spring into the reversing lever knob and then just slid the shaft on to the point where the edge of it covered enough of the bearing to hold it in place. I then removed the hex wrench and slid the shaft the rest of the way through. On the reversing lever there is no guide slot or guide screw so a little bit of toying with it was necessary to find the right spot for the ball bearing to engage one of the two divots on the reversing lever shaft.

Breaking the Curta's main shaft

The reversing lever's knob engages all of the turns counter gears. The gears need to spin freely in the portion of the reversing lever that holds them in place. I did not check that thoroughly enough and combined with other gears getting caught up in the device, I managed to break the main shaft. The shaft broke right at the hole for the pin that connect the crank handle. It broke there because that is where I was applying pressure to turn the main shaft (I was using a small hex wrench to turn it).
Boo!


The main shaft was printed as one piece combined with the step drum and all its toothed segments in a twelve hour print. I have already reprinted the entire thing, but disassembly and reassembly of everything I have done so far is still in my future. While I am in there, I will be adding the parts of the Curta back one at a time and checking friction at every step of the way, filing and re-lubricating any part that shows trouble.

The device should run smoothly and it wasn't doing that when it broke. In fact, there were a several times the Curta got hung up and I couldn't immediately see why it was seizing. It ran pretty smoothly before I added the digit selectors and reversing lever, so I am pretty sure the problem is that the gears don't spin freely when mated with their respective knobs. In some cases I was lazy and pushed harder instead of doing what I should have which is to find what was catching and fix it. This caused stress where the hex wrench pressed against the edge of the hole it went through that weakened and eventually broke the main axle at that point.

I decided to improve the strength of the main axle while I was reprinting it. Instead of the 30% infill I printed it at originally, I was able to tell Simplify3D to print the main shaft portion of the model containing both the step drum and the main shaft at 100% infill. I shouldn't need the added strength once I fix the issues as I reassemble it, but it will be nice to have that piece of mind.

Monday, February 29, 2016

3D Printed Curta Part VII

Much more printing

I have done a lot more printing since I posted last.

Step Drum and upper portion of the main axle

The main axle is just a little bit taller than I can print. I might have been able to modify my printer a little bit to print that tall, but the print bed on my printer is the y-axis so the parts move back and forth as I print. That could cause a tall print to dislodge from the print bed. Instead I divided the main axle into two keyed parts. It is divided such that the parts mate inside the middle of the step drum. The upper part I combined with the step drum and printed as one piece. The lower piece will slide into the bottom of the step drum with a tight fit -- no glue necessary.
I also could have tried printing the main axle horizontally, but the z-axis resolution is limited by the layer height you specify. I might have been able to get a pretty round axle, but the first layer is often slightly squished when printing (it takes a lot of time to tune a printer to be 100% perfect on the first layer and the tuning must be repeated often) and the footprint area that the first layer adheres to the print bed would be very small which also could lead to the print dislodging from the print bed.
After printing the step drum and main axle I found that I had a mistake in my models. The step drum teeth were misaligned by 20 degrees. I had to fix the model and reprint. That was a painful mistake as it took 13 hours for each print.

Main axle bottom portion

Since I split the main axle, I needed to print the bottom part separately. The keying of the two parts is important. The bottom portion of the axle must be oriented properly and the key ensures that it is. After having printed it and carefully fitting it into the step drum, I had a completed main axle. Only I discovered later that the engineering drawings are either incorrect or from different versions of the Curta Type 1. The orientation of the lower portion of the main axle was 90 degrees off. That was easily fixed by altering the direction of the key I added after dividing the main axle into two parts.

Other printing

  • bearing plate 
  • anti-reversal plate, zero positioning lever, and anti-reversal lever 
  • body support columns 
  • some of the transmission axles and gears

Making springs

To make the zero positioning and anti-reversal mechanisms work, I needed to make a couple of springs. All other things kept the same, a torsion spring scales up the force it exerts with the diameter of the wire used to make the spring. That made things easy for me -- I simply multiplied the wire diameter from the original engineering drawings by three for my 3:1 scale Curta.

In order to make springs, you need to have the right kind of metal tempered the right way to make it springier than it is malleable. To this end I got some music wire from a local hobby shop. To actually form the spring, I printed out a couple of mandrels -- basically cylinders with a hexagonal end that I could fit into the chuck of my drill. I made the mandrels a little bit smaller than I needed the springs to be because the wire will want to relax after winding.

This can be very dangerous, so take all of the necessary precautions if you try this (eye protection, gloves, thick clothing, etc)!

I placed the wire and mandrel into the drill so that the chuck grabbed both and very slowly turned the drill while using pliers to hold the wire taught at a tangent to the mandrel. My drill chuck would not turn without the drill turning. Ideally I would have turned it by hand. As the mandrel turns, it will begin wrapping the wire. Keep the loops tight together (it may take some practice to get this right). Once I had the number of loops required, I put the drill in reverse and slowly let the springy wire unwind as much as it wanted to.

It's very important to do let the wire unwind slowly. If the wire is released too early, it will violently unwind while whipping around and possibly causing injury. I accidentally had this happen to myself while winding the thicker wire. When the spring unwound it whipped into my hand. The force broke the skin and left a nasty bruise. I thought for a moment I might have broken it. I've broken a finger before and it actually hurt less than that wire.

Once I had wound my springs, I put them into a toaster oven for as hot as it'd go (450-500 degrees Fahrenheit) for 30 minutes and let them cool slowly inside the oven without opening it. That helped temper the wire so it would keep its shape.
The springs I wound in the toaster oven getting tempered.

Making thread

Many of the parts on the Curta have either external or internal thread so that they can receive screws or nuts. There are some shoulder bolts I couldn't find as well so I set out to cut thread into 3D printed parts using a metric tap and die set I have.

After several tries, I gave up on the 3D printed shoulder bolts. They were M5 bolts and I printed them head down on the print bed so I wouldn't need supports, but that meant that the layers aligned normal to the length of the bolt. The weakest part of a 3D print is between the layers, so as I tried to thread the bolts, they sheered off leaving plastic bits in the die I was using. I eventually decided I could produce those bolts by printing a sheath for a normal M5 bolt to create a shoulder at the size I needed.

After solving that issue I still needed some external thread. The support columns are threaded at the ends so they can be connected to the body and the bearing plate of the Curta using hex nuts. I tried a lot of different methods, but what worked in the end was printing the support columns horizontally to change the layer alignment and prevent the part from shearing as I threaded it. I used a combination of raft and brim to prevent them from coming loose off of the print bed. They aren't perfectly round, but it doesn't matter much for those.

Cutting internal thread was not nearly as difficult. I was able to just turn the die into the plastic a quarter turn at a time. After each quarter turn I would reverse direction to help clear out some of the excess plastic. Depending on how good the first cut was, sometimes I would put the tap through a second time (being very careful not to cross thread it)

The top of the main shaft with the washer and printed circlip connecting the Curta body and tens bell together.

The bottom of the Curta showing the bearing plate and zero positioning / anti-reversal plate. The nuts at the bottoms of the support columns are also visible.

The Curta looking top-down.

A side view of the Curta with the step drum visible.

The anti-turn plate bolted into the bearing plate -- an observant eye will notice that this picture was taken before I fixed and reprinted the lower part of the main axle.

Ready for some transmission axles next



Tuesday, February 2, 2016

3D Printed Curta Part VI

Time for change

I changed jobs and moved to the home of the Carolina Panthers last year and have been busy getting settled in the new house. There are still a lot of things to do and boxes to unpack, but I got my office (mostly) set up and managed to run a few prints. On a side note, Go Panthers! Play well in the Super Bowl this weekend.

Long prints and support material

The first few prints I did after getting the printer up and running again were little things to help recalibrate the printer after being lugged a few hundred miles. After I was confident in its ability to print consistently I challenged it with the main upper body of the Curta.

The main body of the Curta has a lot of overhangs which cannot be printed without support material. I slice my models with Simplify 3D which has really great support material generation that is easy to break away. It does it by producing narrow zig-zagged walls and a one layer gap (configurable) interfacing with the model. The supports are flexible due to the single pass width and are not as securely bound due to the one layer gap. The zig-zag pattern combined with the flexibility allows the supports to be compressed which helps them to easily break cleanly away from the model.

If you are familiar with FDM / FFF 3D printing, you are familiar with parts curling. As the plastic cools, it shrinks some and pulls on previous layers. If it pulls enough, it can cause the edges of your print to curl / peel off of the print bed. The plastic I use (PLA) is much better at not doing this than other plastics, but it can still happen. Particularly if the printer doesn't have a heat bed such as mine. The first time I attempted to print the main body I ran into troubles with this. The support material being only one extrusion-width wide has a small footprint on the print bed and in the case of the main body, many layers to pull against the first layer. That caused some warping on my first print which I had to cancel.

On my second attempt, I told Simplify 3D to alternate support direction by 45 degrees once every ten layers. That helped a lot. I no longer had any curling / peeling issues. In fact, the print was so strongly attached to the print bed I had to use a heat gun to warm the bed up from underneath before I could pry it up. However those alternating direction layers in the support also worked to strengthen the support material making it more difficult to remove from the main body.

On my next print, I reprinted the tens bell I had misaligned in my previous post (the tens bell handles carry operations in the Curta). I decided to go back to Simplify 3D's default support settings. I chose to do that partially because I had printed that part before with no issues and partially because of the difficulty I had removing the support material from the main body. Removing the supports this time was surprisingly just as difficult (if not harder due to the small spaces). I believe this to be due to the material I printed with this time around. I used a clear PLA from Monoprice. It prints very well and it did come out pretty clean, but I think it binds to itself much more readily making the supports more difficult to remove. Due to the nature of the Curta parts, I will be using a different material from here on out.

After printing the tens bell and main body out and removing the supports from each, I was able to assemble them together using a printed c-clip also designed from the engineering drawings. There is a washer that is supposed to be under the c-clip to provide proper spacing, but I have not yet printed that. As far as I can tell everything seems to line up correctly, but I will know more when I can assemble more. The tens bell rotates well in the main body, but I may smooth the print layers a little bit and add some lubrication before final assembly. I also printed the spring that attaches to the bottom of the tens bell. It does have some spring to it, but I don't think it is enough. I may need to fabricate that out of metal.

I recorded time lapse videos of the prints. Unfortunately my printer does move the print bed back and forth which makes the part appear to jump in the video. I will look into mounting the camera directly on the print bed, but that may not be possible without compromising the print on this printer. I am planning to build a Triple C-bot which only moves the print bed vertically. If I finish that before I finish the Curta, I will begin recording my prints on that.



The main body right off of the print bed.

The underside of the main body showing the support layers. Also note the brim.
Here I have removed the majority of the support material, but bits of it remain making the surface rough.


The roughness from the support material seen from a different angle.
The fully cleaned main body.
The main body from the top.
Another view of the main body.

The corrected and reprinted tens bell

The tens bell, spring, c-clip, and main body

The tens bell connected to the main body with the c-clip