Lost wax^M^M^M PLA casting

After Jeshua tried lost wax aluminum casting but with 3d printed PLA I was sceptical if this was within the possibilities of a DIY project.
The equipment he used looked like it was only feasable if you worked at a steel melting plant or something.

Now David Knaack posted some photos of James Harr and Brandon Norris of the Omaha Maker Group doing the same thing but a way more low-tec way. I'm quite impressed. (Photos below)

 And yet another such casting. This time by David Knaack casting a RepRap filament drive gear from aluminum.

The process is:
  1. make a positive master model (here a 3d print)
  2. carefully cast plaster around it. Avoid trapped air.
  3.  place in a furnace to get all remaining water out of the plaster and burn off the PLA.
  4. slowly pour molten aluminum
  5. wait for it to cool. Expect the aluminum to shrink


Mount a NERF gun to a NERF gun ...with ammo.

I heard you liked NERF guns, so I designed a mount so you can mount a NERF gun to a NERF gun.

Tweaked NERF Dual Gun Connector by dfb
  • added ammo holder for 4 rounds
  • added center ramp for locking in place
  • ammo holes have a flat side and a pin to hold the ammo securely


3d printed 45mm fan grill

I replaced the fan on the Thing-o-Matic extruder because I had my fingers in it one to many fimes (when checking the filament) and there was only 1 fan blad and 10 fingers left.
But that thing is 45mm and no shop here has anything between 40mm and 50mm.

So I made a custom one.
It's parametric, so you can change the parameters fan_size and bolt_size for different sized fans.

3d printed vacuum cleaner attachment for CNC spindle

I designed a 3d printed attachment to the spindle of my CNC mill for a vacuum cleaner to suck in all the chippings created.
It's still a work in progress and has not been tested.
I'm waiting for the (sound-proof!!!) shop-vac to arrive first.


Lost PLA aluminum casting

This is so cool!
Jeshua 3d printed some mechanical parts in PLA and then did the same process others would do with wax in lost-wax-casting to get an aluminum part.
Doesn't looks like anything you can do at home. :/
I guess I rather have to mill blocks of aluminum instead.


Githubiverse - Github template for 3d designs

 In the wake of yesterdays "Occupy Thingiverse" happeningsm I learned about "Githubiverse".
A Gitput page template for 3d printed object designs.
Check out the test page!
This looks very neat!
Add a way for tagging and indexing projects and you have a really nice start.
Especially as you get versioning for free.
Something Thingiverse was always lacking.

Update: Looks like Thorsten Paul made some progress. You can now have and navigate multiple things. Check out the "Designs:" section at the top.


3d printed camera slider manufactured

So yes, I actually got around to making one of my completely over-engeneered 3d printed camera slider edges.
The other one is currently in the printer.
Prints fine. Does the job. :)
Now I'll probably also design a better center piece.
One that grips around the LM8UU. So they can't slide out one end.
With additional 1/4" atachment points.

Occupy Thingiverse?

When I woke up, my RSS reader showed some "occupy thingiverse" things posted on Thingiverse.
I'm still trying to figure out all that happened while I was sleeping.

The timeline seems to look like this:
(As the list becomes longer, I'm starting to highlights the latest entries for the benefit or readers returning.)
  1. July 2010 - RepRap blog posting about the heated conveyor belt
  2. August 2011 - Makerbot Industries receives $10M venture capital funding. Applied for patent on heated conveyor belt.
  3. November 2011 - Makerbot Industries filed for a patent on a heated conveyor belt.
  4. February 2012 - New TOS of Thingiverse
  5. Thingiverse blog post about the TOS. Only one commenter "madscifi" asked about the 3.2 attribution clause but wasn't answered.
  6. July 2012 - Makerbot gets the patent. Their own design is highly flawed and no longer included in the next 2 generations
  7. September 2012 - Replicator II announced as a closed source design. 
  8. September 2012 - Blog posting about the Replicator II being closed source, later added TOS change by Josef Prusa (Twitter)
  9. Included ".thing" file format turns out to be just a .zip of multiple .stl and .obj files to be printed.
  10. "Occupy Thingiverse Test cube" posted on Thingiverse, more attention to TOS
  11. Josef Prusa Google+ posting
  12. Tony Buser (Makerbot Industries) posting on Google+ that the TOS didn't chance in 8 month. 
  13. Thingiverse blog post, clarifying that the TOS change was in february 
  14. First thoughts about developing an alternative to Thingiverse. (0,1,2,3,4,5,brainstorming,cubehero,githubiverse,Thingiverse2Github converter, SKDB)
  15. More "occupy Thingiverse" objects popping up all the time.
  16. The story got slashdottet
  17. Blog posting by Makerbot Industries- They are "working out" just how open source they can make the Replicator II, lots about their intentions, no clear work about the specific TOS sentence.
  18. As a response the blog post by Josef Prusa gets updated. 
  19. Thingiverse replacement taking some shape in the comments of a G+ posting. Who wants to contribute? Alternatives are seldomly a bad thing.
  20. Reaction of Hoeken (RRRF and formerly MBI) 
  21. Hackaday posting about it 
  22. It looks like at least the software is just a thin, closed source UI that calls open source Skeinforge or their new (open source) "Miracle-Grue".
  23. This blog entry is on Hacker News and visitor numbers explode. I'm trying to stay more non-POV here.
  24. Looks like the new closed-source software can be tricked into working with older MBI printer. 
  25. New phrase coined "Makerbotgate". 
  26. Adrien Bowyers (RepRap) views on open and closed source printers
  27. We hit 13'000 visitors on this blog posting 
  28. Make Magazine blogs about the Closed Source issue for the second time
  29. Some interesting posts by Openalia
  30. CNET interviewed Josef Prusa, waiting for response from MBI before publishing.
  31. New Blog post from Makerbot Industries 
  32. Blog post of Josef Prusa gets updated
  33. So does Openalia
  34. MBI goes to explaining the TOS  crossposed in the Thingiverse blog 
  35. CNET interview with Josef Prusa, quoting Zach "Hoeken" Smith now online 
  36. Make magazine inverviews key players about this issue 
  37. Bre Pettis talk "Challenges of Open Source Consumer Products" on the Open Hardware Summit is recorded on ustream (session 3 at 1:20) Josef Prusa is in session 2 at 1:28 . 
  38. Make magazine has a transcrips of the Bre Pettis talk. 
  39. We hit the 15'000 reader mark while I was on vacation.
  40. ...
So it seems that:
Makerbot Industries, owner of Thingiverse updated their TOS.

Previously they had a paragraph
"Thingiverse.com does not claim ownership of the materials you post, upload, input or submit to the Thingiverse.com site."
The current TOS states a lot of granted rights but they are all limited to
"solely for the purposes of including your User Content in the Site and Services".
Seems fine to me.

What is interesting is the next sentence:
" You agree to irrevocably waive (and cause to be  waived) any claims and assertions of moral rights or attribution with respect to your User Content."
I can't find a reason for this one in the february blog posting about the TOS.
The recent blog posting reassures us that attribution is done. In the latest blog post by their lawyer the moral rights part is explained as being able to create conflicts with the chosen Creative Commons license and the license given for the specific purpose of hosting the design on Thingiverse at all.

Of cause being legal language of a foreign jusisdiction and one based on common law (wich is still a very strange concept to me), I may misunderstand this part.
It seems that moral rights in the US are separate from copyright and thus not affected by Creative Commons. However why whould you waive any but the "right to the integrity of the work"? Why the moral right of attribution, of anonymous and pseudonymous publication?
Can you even do that if you are not the only author?
  • Can a person from a country such as German where waiving your moral right or attribution is by law impossible even use Thingiverse? (I think yes but I'm no lawyer.)
  • "Thingiverse ... and services" can include any future service. (But that is nothing to deal with now.) “ - "Company provides a service for users to share digital designs that can be printed on 3D printers to create physical objects”
  • How do the two licenses granted in the TOS work out if you are not the (only) author?
It didn't get much attention back then.  Due to the next thing it did get a lot of attention now:

The second thing that happened is that a "Replicator II" was announced and that it was widely seen as closed source.
I'm fine with it being closed source. (As long as they finally start to behave like it's a product that has support and warrenty and is supposed to work and if it doesn't work reliably it gets repaired, recalled, refunded or reimbursed.)
I'm not fine if they used designs that where published with Creative Commons Share-Alike or Attribution (or plain GPL) because none of these two seem to happen. Inheriting from the "Replicator" and thus from "Thing-o-Matic", "Cupcake" and lots of Reprap related designs this is very likely however no such violation is found as of yet (no one has a Replicator II yet) and there is only a very vague reaction about this part from Makerbot Industries.
The latest MBI blog posting states that they are working out "how open source" they can make it.
After that I got a statement that the software  is just one thin, closed source GUI. Also the new .thing file format is not only just a renamed .zip containing standard STL and OBJ files but also some JSON about the placement of these objects in the build volume. A libthing for using it will probably be published.
No word about design files of the hardware.

We'll have to see what parts and how much of the hardware will be in the open source and what exactly the license will be. The software part has now been answered.

It is very interesting to follow the different reactions and strands of discussions popping up and causing wider and wider ripple on the internet.

This posting is being updated as the events unfold.
If you find errors or inaccuracies in this posting, please contact me.


Things to Visit in SF/Silicone Valley after the GSoC Mentors Summit

...just keeping notes here.
Doing this on Google+ would mean that they scroll off screen too fast.

  1. Arrive
  2. Public Transport (San Francisco Municipal Transportation Agency? BART?)
  3. GSoC Mentors Summit
  1. GSoC Mentors Summit
  • Enough time to visit Silicone Valley?
  • What to see there? (1)
  • Computer History Museum already closed 17:00 (PDP-1 and copy of Difference Engine Nr.2) " only open only on Saturday 11 - 5, and Sunday 12 - 4. It is a 15-20 minute walk from Google"=>wouldn't work anyway
  • NASA Ames?
  1. GSoC Mentors Summit until 17:00
  2. Be in SF to check into the second hotel. (Where to store luggage after checking out?)
  1. Visit Noisebridge for "German corner"
  1. Travel back home 
Unknown day:
  1. Golden Gate Bridge
  2. Al Catraz
  3. Pier 39/Fisherman's Warf
  4. Ride the Cable Car ;)
  5. Yosemite? -> asked other mentors to go together.
  6. Techshop  DIY workshop
  7. USS Hornet



Designed an improved side for the LM8UU (8mm linear bearing) based camera slider Matt Roberts designed.
  1. can be mounted on 1 or 2 tripods
  2. More stability by center 40x10mm L profile
  3. 1/4" UNC camera accessories can be mounted to it
  4. All bolt holes tapered for easy assembly
  5. All human touchable edges rounded to make it nicer to touch
  6. 8mm rods secured by M3 grub screws


RS485 CNC spindle control

My USB to RS485 adapter just arrived from China.
Will be able to test controlling the CNC spindle from the computer by MACH3 later today! 

(remove tape from taped over keys,
 then press PRGM, enter the number of the value to change via UP,DOWN and <<, pres SET.
 You now see the current value.
 enter a new value via UP,DOWN,<< and press SET to change it.
 The device will leave PRGM mode on it's own after a few seconds)

PD163 "communication address"  1
PD164 "baud rate" 1 = 9600bps
PD165 "communication data method" 3 = 8N1 For RTU4
0=8n1 ASCII gives errors in Mach3 (don't know how because it's a one-way only communication)
3=8n1 RTU4 gives no errors but still the spindle doesn't react
PD001 "source of run commands" 2 = communication port (0=operator)
PD002 "source of operating frequency" 2=communication port (0=operator)
now it should work.

PD144 "rated motor revolutions" default=1440

Arduino ADC confusion

I don't get it.
I want to meassure 1g-200g with an  #Arduino .
Alexis Wiasmitinow made me a PCB using the
750g load cell and an op-amp with adjustable gain and offset.
When I place a 150g load on the cell it sometimes increases *or decreases*.
When I remove that load it sometimes doesn't return to the value it had before.
When I adjust the offset to get very small values with no load placed on the load cell, I get no change in the value at all. With higher offsets I get value changes depending on the gain.
But never more then 3-4 bit of resolution .
I'm using pin A1 on a Arduino with the LCD-Keypad shield connected (shield uses A0 and 4-10).
Any idea?

Update: We solved it via Skype chat (G+ hangouts only produced "server error" and "severe server error".)
I simply had my load cell mounted the wrong way. My load was in the wrong place!!!


WIP: camera slider

I'm currently trying to make a 3d printable camera slider design made by mattroberts and then improve on it.
Problem is: I have LM8UU bearings but one of my 2 8mm rods...is not 8mm but more like 8.05mm .
The bearings get stuck.
I'll try to get it replaced on monday and do an extensive posting once I have made it work.


3D printing again

Being so occupied with the new CNC milling machine I haven't 3d printed anything in a while.
So here I am, printing mounting brackets Ryan designed to attach my new limit switches to the CNC mill.
I had lots and lots of issues with filament expanding and blocking on my Thing-o-Matic.
So this time I tried something new. I lowered the PLA temperature from 200°C to 185°C .
I still need to watch the whole print and instantly react when the filament stops to come out or the drive bolt starts to grind on the filament instead of transporting it. But so far I have 30 minutes of productive printing.
Formerly it would hardly do 10 minutes without major issues.
Still a *long* way from actually using the ABP to print a batch of objects while I'm sleeping...


Super cheap logic analyser

Just got my <10eur logic analyser.
No idea yet how many samples it can actually hold.
They didn't tell.  ;)
  • software "USBee AX (Standard, Plus and Pro)" 

New drill arrived too - adding end stops to YOOCNC CNC6040

My new drill arrived too!
(The cheap, old one broke down so I got myself a real good one.)
Finally I can mount the end stops to my CNC6040 milling machine.
If it looses steps and I can stop it in time before the blank gets damaged,
I should then be able to resume a G-code program.
(Without eye-balling or edge-finding  known location to get the current position calibrated again.)

Milling bits arrived

After 22 days from China.
Now I can prepare some blanks and finally test fine detail 4 axis milling on the YOOCNC CNC6040...
Wish me luck!


YOOCNC CNC6040 up and running well

The CNC6040 is now up and running very well.
I just purchased my license of Mach3 and will probably get the hobby version of DeskProto too.
No longer loosing steps and now with a flat and wide machining vice mounted to the table to mount blanks at defined locations.

Haven't tested the 4th axis yet. At least not beyond jogging it manually to see if it moves.

I'm still running the chinese electronics and cables. These will be replaced with shielded cables and a Gecko G540 at some point but not right now. (Paying for the machine and the 2 software licenses is enough for the the time being.)
Now I'm waiting for some cutters and an 0.001mm dial gauge and my new drill to make the holes for mounting the end stops in my pocket.