A New Home For The Raspberry Pi

I received my shipment from Adafruit Industries today. Amongst the things I ordered was their Raspberry Pi enclosure.

The "Pi" in its new home

I must say I really like this enclosure. It was a breeze to put together and keeping all the components visible (and safe from grubby hands) helps with the geek factor as well.

As of writing it's out of stock at Adafruit, but it seems they restock more often than it says on the web page so be sure to check regularly if you want one for yourself.

Now I have no excuse not to start messing around with the "Pi" - should be fun!

SUMPOD Bed Leveling

I now have the final bed leveling mechanism assembled. This version is a lot easier to fine-tune than my first try. Another advantage is that it is pretty heat resistant so I can add a heated bed if and when I decide to.

Spring loaded action...

A big thanks to Mark Hindess for sending the parts needed!

SUMPOD Printing

This weekend I finally got to do a proper test-run on the SUMPOD (total printing time around 7-8 hours). After using most of last week to trouble shoot and figure out usable settings I stumbled upon Airtrippers excelled Skeinforge settings. Since he already has his SUMPOD up and running and having excellent results, from what I can see, at least I knew these were going to work.

After browsing Thingiverse a bit I ended up on the "Mustache Ring" for my first, proper, test. It has some nice detail and doesn't take too long to print so one can experiment a bit.

First proper test object. Mustache Ring.

If you look closely you can see that there are gaps and that the layers don't fill properly.

So far I haven't had any big issues getting the plastic to stick to the printer bed, but changing the filament and using the new Skeinforge settings apparently made a huge difference. All of a sudden it wouldn't stick at all. At first I though I needed to up the temperature in the print head, but that only made it worse. I also played around with the height of the nozzle over the print bed. Supposedly you should have the nozzle touch, or even press a bit into, the blue masking tape, but that just made the PLA bunch up around the print head. Reducing the temperature back to 185 degrees C in addition to increasing the distance a bit  did the trick - although now it almost stick too well and I have to rip the masking tape off of the bed on larger objects. I doubt that is right.

I soon realized I was going to need something to help unwind the filament from the spool since I can't go over and unwind a chunk every 15-20 minutes on multi-hour prints. Yet again Aitripper saved the day with his design for the Pocket Reel Roller. He also lists his settings for printing this, but I needed to change those around quite a bit before I had something passable.

Mainly I had to change "Speed -> Object First Layer" (all to 0.2) and "Fill -> Infill Solidity" (to 0.9). Print time was around 2.5 hours.

Third print of this part. 

With roller bearing installed.

Not exactly pretty, but I think it will do the job. The last part actually came out worse, but I think that's because the belts have gotten a bit of slack. I'll tighten them up a bit and hopefully that'll do the trick.

All installed - turns pretty easily as well.

I still has some way to go to get the quality up to the level I like, but playing around with the parameters in Skeinforge has a lot of potential - I only have to learn what causes what and that's not something that's done over night with the amount of parameters you have access to.

Inventor Fusion

Since I tend to stay in OS X on my iMac the only real reason i boot into bootcamp and Windows 7 is when I need to do something in Alibre Design. Unfortunately Alibre is Windows only and from what I gather (please correct me if I'm wrong) they have no plans to port it to other platforms. Not that I blame them - it's generally a huge task to convert a program from Direct-X to OpenGL. Especially when you have tons of legacy code.

So today I got my fill of jumping back and forth between OS'es so I thought it was time to check out Inventor Fusion for Mac by Autodesk. Luckily they have a free technology preview of this software which I installed it a while ago - just never got around to test.

First impression of the user interface was good, although a bit more simplistic than Alibre, followed by the obligatory; "how do I do this and that in this software". After a couple of quick-run-through videos on YouTube courtesy of Autodesk I had a rudimentary understanding of the software.

Upper part of outrunner motor bracket designed in Inventor Fusion.

Although some operations are a bit tricky I think it's more about figuring out "the correct procedure" to get stuff done so by playing around with it a bit more I think the overall experience is going to get better.

Since this still is a technology preview it crashed on several occasions, which is to be expected I guess, even though the operations I did were on simple objects. It even corrupting my scene file, but yet again my (paranoid?) habit of saving new versions paid off so no big loss.

Over all it is much faster to fire up Inventor than booting to Windows so I think I'll stick with it, at least for simple objects, for now.

Alternative printer bed leveling

While I am waiting for the proper parts for leveling the print bed I came up with an alternative way. It's not something I would recommend if you have a heated print bed, but as I do not it's worth trying.

I found some packing foam and som IKEA cupboard bolts which should do the trick. The foam has the right springiness and by changing the size of the pieces you can get the proper resistance. Covering the whole back side of the bed was too much I found out.

Pink and stylish

Note the black drinking straws

Notice the four, black, drinking straws. These had the exact right diameter to match the predrilled holes and keeps everything nice and centered.

Before tightening of the bolts

First try. I found that since I had to tighten the rear bolts almost all the way the small pads of foam I started with weren't enough in the front.

After tightening of the rear side.

Added some more foam and the result turned out much better! Leveling took some time since tightening or loosening one screw affects the height of the nearby foam blocks so I had to re-measure and adjust the other bolts several times before it all was level. Still; I'm happy to be able to use the printer until I get some better parts.

The Saga of The 3D Printer

It's been a long journey, but now, nearly 10 months after I first ordered my 3D printer it's finally done! Well, as done as a gizmo like this is ever going to be I guess - it will forever be a work in progress with parts being tuned and such.

In my last post I though I were finished painting, but the perfectionist in me didn't like the finish on the sides of the parts. So, more paint!

Primed and ready...

Some of the other parts.

Masked and ready for paint.

Since my significant other loves purple my reasoning was that maybe if I used that as the secondary color she might be persuaded to let me keep the printer in the living room :)

Purple it is!

After painting...

Finally being happy with the paint job it was time to move on with the assembly. The biggest issue was the error in the routing of the feeder part. At first I thought I would see if I could fix it myself since I only needed to drill some holes and remove some MDF to make room for the parts. Well, a seemingly easy job botched up mainly because of me sitting on my living room floor and not having access to a proper workbench.

Result of drilling in MDF without proper support.

After e-mailing Richard about this he promptly replied that he was not only going to send me new feeder parts, but also the newest revision of the hot-end as compensation. That's what I call great customer support - thanks Richard! :) A couple of weeks later I was ready to proceed.

The new hot-end with a single extruder. There are room for
two extruders if I on a later point want to add one more. 

Seen from the top.

The main difference between the old and the new extruder being a connection piece made of peek (the brown plastic in the picture). Peek handles heat very well so we'll have no issues with the plastic in the bowden tube retainer (blue) melting and we can skip the fan and heat sink as well simplifying the design greatly.

EDIT: As mentioned in the comments below the way the print-head is assembled in the pictures is wrong! The correct way is (from the top): push fitting, MDF (upside down from what's seen here), peek and then the nozzle. Unfortunately I forgot to take new pictures before installing the print head in the SUMPOD.

Belt in place for the X-axis.

The Y-axis in place. I took some extra time prettying up the
cables and at the same time making everything more tidy.

In progress with the final connections underneath the
SUMPOD.

Crimping the connectors took a while since I wanted to do
it properly and didn't have the proper tool. Crimp, solder
and then add the plastic casing.

I imagine final assembly took about 3-4 days in total. Much of that time was checking and double-checking the instructions with the SUMPOD forum as well as the other build-logs out there. Instructions were not always straight forward or the same and I wanted to avoid having to pull anything apart - better to get it right in the first place!

Although I did test the electronics before I assembled the SUMPOD it was a bit nerve wrecking the first time I turned on the unit. To my astonishment everything worked perfectly! From what I had read on the forum I knew some people had issues with the feed mechanism not working in ReplicatorG so I weren't surprised my unit had the same issue. A quick swap to Pronterface showed the feeder worked fine albeit reversely. Changing the Configuration.h file in the Sprinter firmware fixed that issue.

// Inverting axis direction
const bool INVERT_X_DIR = true;
const bool INVERT_Y_DIR = true;
const bool INVERT_Z_DIR = true;
const bool INVERT_E_DIR = false;

The malfunctioning feeder in ReplicatorG is caused by a bug which is fixed, but hasn't made it into the main code yet. If you are on windows you can use ReplicatorG 0034 for 5D, or supposedly 0029 should work as well. It doesn't yet for me...

Next up were calibration. Unfortunately I didn't have much to go with on how to to that exactly so to make a long story short I ended up with a nasty collision between the printer head and the print bed - even though I had my finger on the power button. The result of the collision left my printer bed crooked on the Y-axis. I don't have a picture to show this, but it is visible to the eye and I guess it's about 3mm difference between the top and bottom. It was all my fault and being angry with myself I had to leave the project alone a couple of days before I could continue. Not necessarily a bad thing since it gives the subconscious time to work out whatever needs to be done to fix the problem. Which in this case were; where is Z = 0.

Browsing the forums and the Wiki led me to this g-code from Airtripper. Experimenting with the Z value (and my finger on the off-button) I started with the high value of 10.0 and working my way down.

My setting of Z = 2.5

With Z = 2.5 I get the correct distance from the printer head to the bed. It's not ideal yet since the bed is not flat compared to the print head on the Y-axis. Anyways; I'll look into the Sprinter firmware to see if I can set the value there or if I have to pre-process all my g-code with this value.

Pronterface

Since I still have issues with ReplicatorG I continued to use Pronterface. Although it doesn't have a 3D view of the model to be printed, like ReplicatorG has, I love the manual controls. The temperature monitoring also works as expected. That's another area where I have had issues with ReplicatorG with it only sporadically showing any readings. I'm not sure if that has anything to do with me running 0029 or the Sprinter firmware, but I would guess the former since Pronterface works fine.

First print! Red and pretty :D

Closeup of the first print. Seems like the registration is good!
The lens is to blame for the apparent skewing.

I'm impressed by the quality of the print and what little can be seen of artifacts I believe can be linked to the bed and the distance between the bed and printer head being a bit too high. I can clearly see the thread, but playing with the parameters of Skeinforge should allow for higher resolution in that aspect.

I did try to remove the raft, but it's firmly stuck so a knife is needed. After I have made some adjustments I'll try to print without a raft and see how that goes.

What remains now is to get the bed leveled. I'm considering a variation of the mechanism described on the SUMPOD wiki, but it's hard to source springs here in Norway so if my RC-Car spring spare-part idea doesn't pan out I have to think about something else.

I also need to work on the display contrast. Not that I use it, but It's nice to have it working anyway. With the current setup I use a 3K resistor which is a bit much since the contrast is too low.

I have purchased Alibre Design for the purpose of 3D modeling of parts, but since my main computer at home is a Mac I'll keep an eye on the new Autodesk Inventor Fusion as well. It's a bit of a hassle to boot into bootcamp every time I need to fix a part. Now on to the next project - where self-printed plastic parts most likely will play a great part :) After all; I have around 7 Kg of PLA and ABS plastics to play with...

Back To The Roots

After watching "Iron Sky" and reading about the production at CGSociety I got the urge to check out Lightwave again. This is after all the software that got me started with 3D all those years ago. After a brief stint with 3D Studio 4 run under MS-DOS I tried my hands on Lightwave 4 - instant love! I kept using Lightwave for a couple of years, but left it shortly after v6 came out seeing Maya as the future. I believe Maya was at v2.5 back then. Although I kept using Lightwave on and off for a couple of years Maya eventually replaced it completely.

What I remember I liked about Lightwave was the SubD's and the integrated renderer - which was very good. The modelling tools suited me quite well as well, but after learning the Maya tools I really didn't look back.

Head model courtesy Dance Macabre @ TurboSquid

Lightwave 11, as of today, has a free 30 day trial. Registering gets you a serial that unlocks all functionality and removes water mark - which is nice.

The layout is pretty much the same as I remember from Lightwave 6. You still have two different programs, modeller and layout. I must admit I'm a bit rusty not having touched the program in 10-12 years, but I managed to assign some shaders, add fur, light and render a small test.

Although 999 free render licenses are tempting I don't think I'll be jumping back.