I’m a happy, happy guy…

March 11, 2010

It was the worst of times.  It was the best of times.  After struggling for a few days trying to get the thick stock parts file into a format that someone (anyone!) with a laser cutting machine could deal with, I was about at wits end.  The gentleman I originally planned to work with could only use older formats I do not have access to.  I finally called a local larger scale vendor (TAP Plastics) to see if they could handle a small job from IGES or STEP format.  They could, but it would be pricey.  To their great credit, however, they referred me to a new, small company in Seattle’s Capitol Hill neighborhood: Metrix Create Space.  TAP said it would probably save me a lot of money and be a lot faster.  After trying to call a couple of times on my lunch break, I resolved to head there after work since they are open until midnight.  First, though, I output the layout file in as many formats as I could in the hope that one of them would work…

The drive into Seattle was quick, but I then spent the next 20 minutes trying to find a parking place and looking for the shop.  I finally found a space and got on foot, almost walking past the entrance which is down a flight of stairs.  I walk in the door and, lo and behold, the proprietor (Matt) is sitting behind the counter working on (wait for it) a Mendel.  Fully assembled, on the virge of first print.  We talked for a bit and he wound up cutting out a set of parts for me out of plywood, as well as an additional acrylic printing surface with larger screw holes to fit over the bolt heads and protect my (eventual) extruder.  All of this at a very reasonable price (half of what my earlier plan was when materials were figured in), and I didn’t need to have a template file – Matt already had the file from his own Mendel.  Also got some good advice about extruders and such, as well.  MCS is a cool concept: a space for geeks to congregate, share experiences and imaginations, and get access to tools that would otherwise be hard to find or afford.  Check them out!

So, what do the final parts look like?


Note the optional sixth acrylic build table that will sit on the plywood build table.  With these done, I’ve got a clear path forward on the mechanical construction.  Good thing, too, since I’ve been busy:

This shot shows the leadscrews for the z-axis, as well as the y-axis motor and idler, and the y axis bearing sets.  The bearing sets will get mounted to the small rectangular and larger chicken-shaped board in the previous picture to complete the Y axis assemblies.  The leadscrews are combined with the z-axis motor, tensioner, and leadscrew brackets, already mounted on the end frames

to complete the z-axis assemblies.  Things are starting to come together quickly now, but at the same time more care is required:  Dimensions count and I spent a lot of time getting the leg lengths right on the end frames.

I think that brings everything up to date.  I’m looking forward to working on this most of the day Friday (a day off for me) and getting most of the mechanical construction done. Then I will retest the opto-stops and it will be time to start work on the rest of the electronics.  For now, though, ZZZZzzzzz…..


Time for some R&R…

March 9, 2010

It’s been a few days since the last post – we went to the Oregon coast with some friends for a long weekend.  Good company, good food, good weather, and good times.  That doesn’t mean there hasn’t been progress, however!  In fact, I’m pleased to say that the X-axis is finished. I would guess that that means that the mechanical construction is about 10% complete.  As I look forward in the instructions, it continues to be daunting, but I’ll keep plugging away.  Anyway, on to some new pictures:

This is the completed x-axis.  You can see the wires leading to the stepper motors that drive the extruder and x-axis carriage, as well as the z-axis stop flag (hanging down from the right end) and the red x-axis opto-stop board at the top left.

This is the same view, but from the back.

This shows the underside of the end containing the x-axis carriage drive motor.  The four gold washers have bearings underneath them, and you can see the toothed belt that wraps around them.  If you look closely, you can see the stepper drive gear below and in between the two middle gold washers.  Below this, you can also see the hole containing the captive 8 mm nut that rides on the z-axis leadscrew behind the three bearings arranged at 120 degrees to one another.  This three bearing arrangement allows the z-axis guide rail to be fully captured and removes play from the system.

An interesting technical problem arose with the toothed belt and the bearings it rides on.  The BOM calls for a 5 mm wide belt, but the belt from McMaster (suggested by the BOM) is actually about 6 mm.  This means that the suggested bearing arrangement of small washer, bearing, small washer, fender washer did not leave enough room for the belt to move smoothly without rubbing against structure.  Since I had to buy a bunch of extra fender washers anyway, I changed this order to fender washer, two small washers, bearing, small washer, fender washer.  This leaves plenty of room for the belt to ride smoothly on the bearing between the two fender washers, completely isolates the belt from rubbing against structure, and nicely captures the belt to keep it in place.

Next you can see the other end of the x-axis:

This time, there is no motor gear to route the belt around, so there are only the two corner bearings.  You can once again see where the leadscrew nut is, this time between just two bearings.  Since alignment is provided by the three bearing set on the other end, only two opposed bearings are used allowing the z-axis rail to float a little.

Finally, this is an “overhead” view of the x-axis, showing the positions of the motors.  Note here that the toothed belt makes a loop around the entire axis and is clamped off at the carriage (top center of the part I am holding onto).  The short piece of belt sticking up is a little extra that I grab onto with pliers when I am trying to tighten the belt.

There is more to talk about – all of the z-axis assemblies are complete – but I’ll write about them and show pictures next time.  I’m having some trouble getting the thick stock parts laser cut, and will be working on getting that done this week.  I’m hoping I don’t have to do it by hand on the bandsaw.  The first of the laser cut parts is needed at the end of the y-axis assembly, so I need to make a decision on fabricating them pretty soon.

Enough with the Rosie fer cryin’ out loud…

March 3, 2010

By popular demand, I’m letting go of the Rosie theme.  I won’t say who demanded it, but there was a foreign accent and threats of mayhem involved…

I’ve been busy the last couple of days, biding my time until my new fender washers arrive.  Three subassemblies are complete (pictures below), and I’ve assembled all three of the Optostop boards.  Its been a long time since I’ve done any soldering so this was fun and didn’t take very long.  Three boards, each with 21 through-hole connections to make.  Someone handy with a soldering iron could probably do it in 15 minutes.  It took me, well, lets just say it took me longer.  The good news is that I tested all three and they all passed.

Oh, and I learned that you don’t press down on the flux pen – the stuff squirts all over and is incredibly sticky.  Isopropyl alcohol on a toothbrush seems to clean it up pretty well, but I think my shirt may no longer be wearable to work.

So the first set of pictures is a continuing buildup of the extruder carriage in the last post:

This view shows the installed motor on the pinchwheel bracket.  If you were to look down next to the motor shaft, you would see that there is a bearing that is used as a pinch roller with the shaft.  The fiber stock feeds through the small hole in the top, between the shaft and the pinch bearing, and down through the heated extruder head (not built yet).  The thin metal flag is used along with an Optostop board to keep the carriage from going too far as it traverses in the X direction.

This picture shows the biggest problem I’ve run into so far.  The recessed cap screw near the upper left hand corner of the motor is actually about 1 cm too short.  The pinch wheel bracket has been redesigned to form an extension on the side, which made the bolt path about a cm longer.  Unfortunately the designers didn’t catch this and call out a longer screw in the BOM.  To get around this and avoid having to wait a few days for a 50 mm long x 4 mm capscrew, I dug out a Forstner bit and drilled a counterbore just larger than the washer and deep enough that the captured nut can fully engage the screw.

This is just a view of the bottom of the carriage.  You can see that the extruder head is missing.  Once these pictures were taken I removed the motor.  It turns out that the part it attaches to is not quite dry and weeps moisture when compressed by the screws.  It will need to take a trip back into the oven before the motor undergoes final installation.

The next set of pictures show the “vertical-bearing 360”

In this picture you can see the round holes to the left and right that grasp the X-axis rails.  The triple bearing sets at top and bottom surround (thus 360) the vertical axis rails.

The small circuit board is an Optostop with three resistors, an LED, an RJ-45 connector, and a difficult-to-see optical sensor.  Behind the board where you can’t see it is another vertical hole.  Inside this hole is a captured 8 mm nut which will thread onto the vertical leadscrew that drives the completed X-axis up and down.  This picture (respectfully stolen from the Reprap website) shows things a bit better:

Finally, the last assembly I have completed is the leadscrew nut assembly from the other end of the X-axis:

Again, you can see one of the round holes that grab the X-axis rails (on the top surface at bottom left), but this time you can see the leadscrew passage on the right vertical face which also contains a captured 8 mm nut to ride on the second leadscrew.  The Optoswitch flag for the vertical motion (Z-axis) is also clearly visible.

Finally, one last bit of progress:  the three new gears are complete and have been fitted to their motors.  The arbor holes were still a little small because of the STL faceting, but a small rattail file took care of that in a few minutes and a solid press fit was achieved.

The next step is to create a couple more X-axis subassemblies, and then assemble them all into the final X-axis.  Hopefully, I’ll be able to get to work on that tomorrow, assuming the new fender washers arrive 🙂

Rosie Conceived…

March 1, 2010

Well, I was going to wait until I had all the mechanical parts ready (except the extruder parts) before I started the build, but I just couldn’t wait – the laser cut parts won’t be done for a few days yet, and I want to get started.  The last of the printed parts came out of the support dissolving process today – and I wound up with what looks like a big bag of Legos…

Kind of hard to tell in this picture, but there are almost 100 parts in here.  Also kind of hard to tell, but this is really the foundation of the Mendel Reprap.  These parts provide all of the alignment of the rest of the parts, support the motors, and provide a lot of other functions.

Tonight I started on the first subassembly (the extruder carriage) of the first major assembly (the X-axis).  Things went pretty well:

I did learn a couple of things, however.  The most important of these is, of course, the first rule of open source: documentation rarely keeps up with design.  When all was said and done I wound up with two extra washers and two extra nuts, and absolutely no indication in the pictures where they should go.  I’m pretty sure it is just a documentation issue since you would expect the number of nuts to match the number of bolts, but in this case the documentation seems to be in error.  Not a big deal, though.  I have a feeling that the parts were supposed to be used in mounting the extruder bracket, which I’m not doing right now.

I had another run-in with the imperial unit mafia today.  I went all the way down to Mclendon’s Hardware in Renton looking for some metric drill bits.  I mean, really, if anyplace has them Mclendon’s does, right?  Nope – discontinued. Shoulda called first.  The place was never the same after the moved out of the old building.  At least they had a meter stick marked off in mm, though.  More than could be said for Home Depot.  I’m constantly amazed at how big a challenge the metric system seems to be for people – it is much easier to deal with than imperial units.

That’s it for now.  Won’t have time to work on Rosie for a couple of days, but hope to get the X-axis done this week!

Its a very Rosie Xmas…

February 26, 2010

We live in a very small world  these days.  On the 22nd, I ordered a bunch of parts for Rosie from a couple of companies in LA, and 4 stepper motors from a company in England.  They all arrived three days later – even the motors.  It seems like not so long ago sending something overseas was a huge undertaking, and you never quite new how long it would take.  No longer!  Coming home from work today was a little like Xmas morning!

So all of the fasteners, studding, bar stock, bearings, belts, and motors have arrived.  There was a slight problem with one group of washers, so I have to deal with that, but otherwise the orders all look complete.  The last two parts printed successfully, and I have cleaned the support material out of about half the 34 parts that needed cleaning.  The support material is soluble over several hours in a heated detergent bath.  I’ll clean the rest tomorrow and cut all the studding and bar stock to length.  The printed parts will need to soak overnight, then dry, so I am probably still a couple of days away from assembly. That probably works out OK since I need to wait on the new washers and on the laser cut parts.  Soon, though!  The BOM is almost complete for the mechanical construction.

I just realized that the motor shaft does not have the double flats that the drive gears have.  I’m going to have to make a custom gear and print it out.

Right now, though, time for sleep!

Rosie Gets Ready…

February 25, 2010

So this is a pretty big day…  By the time I post this, all of the Reprapped parts will have been printed.  The fifth plate of 11 finished this morning, and the last plate of two small parts was printing when I left work. Tonight I brought the first five plates home and cracked all the parts off, then sorted them by whether or not they would need to be put into the solvent bath for a final cleaning.  Of the 98 printed parts, 63 needed no additional work, 2 are printing, and 33 need further cleaning.  I picked as much of the support material off as I could (messy, messy job – wear safety glasses!) and the rest will get dissolved out.  That will probably take the whole day tomorrow – not much to do, though, just stir the parts in the bath now and then.  Here is a picture of the fifth plate:

The last two parts to print had problems in the STL files that it took awhile to fix.  Otherwise they would probably be on this plate.

Tomorrow my order from McMaster should arrive, and the bearings shouldn’t be far behind.  That is good news.  I’ll be able to cut all of the bar stock on Friday, and hopefully start mechanical assembly this weekend.  It also gives me a couple of days to finish cleaning the printed parts, get them dried out, and get everything organized for assembly.  Can’t wait!

Rosie Picks My Pocket (Again…)

February 21, 2010

I swear, I’m spending more on postage on this project than I am materials. After spending the entire evening  trying to find the right stepper motors from a supplier in the US, I finally settled on a UK distributor.  They had 4 quality motors available that are perfect and relatively low cost.  In fact, the Reprap site listed the B-version of the motors as proven, but they are double shafted.  The A-version, identical except for a single shaft (which is better) is now available from the same distributor.  The cost? Including postage, less than the cheapest equivalent motors I could find in the US.   So the motors are now waiting on airfreight delivery from across the pond.

Wrote an email to a laser cutting house tonight, also.  With a little luck, that will get done this week.  Also fabricated the opto-endstop flags:

They still need to be drilled (I have to find some metric bits), but they came out pretty well.  I’m glad I went with the heavier flashing stock.

Well, once everything on order arrives, I will cut all the bar and studding stock to length, and I should be ready for mechanical assembly.  The only mechanical parts that I haven’t got in the pipeline are for the extruder, which is fine since that will probably be the very last part built – even after the electronics.  That bit is a little complicated and is going to take some time.

OK, time for bed.  Rosie has kept me up long enough for 1 (um…. 2) days.