Tuesday, January 5, 2016

Aluminum 3DPrintMi Released

After spending two years designing, documenting, and refining the 3DPrintMi, I am now releasing the next iteration, Aluminum 3DPrintMi, to the general public. As an open source project, I hope my printer will give those interested the opportunity to build and use a 3D printer with ease. I firmly believe that you learn by doing since that's exactly how I got into 3D printing. No amount of books about 3D printing will prepare you. Just go and get your hands dirty. This design may be like every other reprap 3D printer available except most of my time was invested in the documentation process than the design itself. My main goal was to provide an easy to follow assembly guide along with other pertinent documentations to improve the user experience in building a cool machine and get started as quickly as possible. Please share my design to the world.

Happy Printing, Making, and Innovating!

http://www.thingiverse.com/thing:825277

http://www.instructables.com/id/Aluminum-3DPrintMi-3DPrinter-Assembly-Guide/

Sunday, February 22, 2015

Aluminum 3DPrintMi - Automatic Bed Leveling/Tramming

Auto-Leveling Sweetness

Hey all,

2 week ago, I was able to install and test the new auto-leveling/tramming (whatever you want to call it) feature in the Marlin firmware. I always wanted to add this feature to the 3DPrintMi. I was always on the fence because I never felt confident in the servo/mechanical switch setup. There's too many moving parts involved for something that needed to read data consistently. Other setups came along like the auto touch Z end-stop and the FSR (Force Sensor Resistor) but impletenting them meant overhauling a few components. It wasn't until Printrbot introduced the induction probe which was a real game changer for me. After many hours of preparing, printing, and testing, I finally have a reliable setup.

Induction probe setup on the Aluminum 3DPrintMi
Full setup - Induction Probe with ducted fan cooling
After using the auto leveling feature, I'm actually blown away by the results. I'm getting better quality prints and consistent dimensional accuracy. For a long time, I was designing multi-part components with a +0.3mm allowance for clearance. Now I'm able to go as low as +0.1mm and still achieve a loose fit.

What sets my setup apart from the others is that I'm still able to print on a glass plate. Read on and I'll show you how to pair an induction probe with a glass plate.

Induction Probe

SN04-N probe 
Inductive Proximity Sensor Switch SN04-N (DC 4.8-30V)

For this auto leveling setup, I'm using the SN04-N induction probe instead of the NPN model probe Printrbot uses. There's a few reason why I went for this version:

  • Smaller profile, doesn't take much space on the tool head (extruder mount)
  • Relatively cheap, about $7.
  • No need to reduce the output voltage, the rated DCV is within the electronic board spec (5V)


Connecting to the Printrboard
Got really confused with the wiring. The schematics on the induction probe doesn't match up with the Printrboard. Here's how the wiring is done. (Blue = ground, Brown = input, Black = output)

Finding a spot for the SN04-N probe
Induction probe mount

The low profile probe gave me more flexibility in locating a mounting spot . If I have gone with the NPN model, I would have to redesign my extruder mount, thus increasing the tool head profile. I had one mount hole to work with which was good enough for me. It sits really close to the heater block so I have to keep an eye on it in case it starts melting. *knock on wood*  

Induction probe and glass plate setup

Now, the typical induction probe setup requires a metal plate in order for it to work. Most printers use an aluminum build plate as a build platform. I never had much success printing on an aluminum plate so I wasn't ready to give up the glass. Apparently, sticking an aluminum plate under the plate doesn't work. Bummer.

Since that didn't work, I tried:
  1. Sticking aluminum foil under the glass plate 
  2. 3mm thick aluminum sheet under the glass plate
  3. Stuck aluminum tape on four corners of the glass plate (detects but lose build volume)
Just when I was about admit defeat, I noticed my sensor would detect my steel spatula from a much greater distance, about an extra 2-3 mm.  Doesn't seem a lot but it was worth testing. Luckily, I had a 28 gauge steel sheet lying around.....

Using a steel sheet for the induction probe to detect the build plate.
Cut the 28 gauge steel sheet to the same size as the glass plate

Clamp the glass plate and steel sheet with binder clips
Steel sheet under glass plate, clamped with binder clips. 
Cut it down to size, clamped it under the glass plate, crossed my fingers, and homed the Z. Hell yeah! It works! Now I get to keep my glass and barely adds any mass to the build plate. A win win.

-Note -

Stainless steel sheets DO NOT WORK.  Bought a pre-cut, 26 gauge stainless steel plate from onlinemetals.com and the probe just couldn't detect it. My poor printer crashed itself into the bed. The only difference between the two is that one is stainless while the other is galvanized. Galvanized steel just means it has a protective layer of zinc which is probably what the induction probe picks up instead. I have heard copper pcp works but they can be pretty expensive. 

Marlin Firmware Setup

If it wasn't for Thomas Sanladerer's youtube video and Zenmaster's blog, I wouldn't be able to get this going. Thank you guys!  I made a pictorial guide for myself just in case I forget. 










Results


auto-leveling sequence

thin wall calibration cube, measured 0.50mm on all 4 sides. I used to get 3 consistent readings and 1 way off.

Custom hexagon fan shroud, top infill came out perfect.

Tall prints remain consistent, no artifacts visible

Walkaway cube test






Sunday, January 18, 2015

Build Log #? - Introducing the Aluminum 3DPrintMi

Back again and reviving this good ol' build blog with some new content. I mentioned on my last post I was working on revamping the 3DPrintMi by switching to a more robust frame. After that, I got tied up with work and never found time to post my progress. Since I didn't post anything for a whole year, this post will show off all the cool new frame and features from the last post up to today.

3dprintmi
Alpha Aluminum 3DPrintMi

This was the last picture I took of the new 3DPrintMi. I replaced the threaded rod frame with aluminum extrusion 2020s and boy, they are surprisingly cheap. I purchased them through Misumi and I can get one 370mm extrusion for $2.10 pre-cut. You can even select the length too. Even though threaded rods are cheap as dirt, you save on labor time. No more measuring, cutting and grinding.  

IIRC, I was trying to retain the original design into the new frame to see if it was feasible. It wasn't. The Y gantry (build plate direction) had some major wobble when you apply pressure to the top frame. 

3dprintmi
The Aluminum 3DPrintMi
Here it is! Looks very different from the alpha version. I completely toss all the threaded rods out It's my workhorse right now, hence all the tools lying around. This little guy already put in 300+ hours of service and still going.

Trying to remember the steps I've taken to get to this stage. Below is a series of re-design stages from the past year.


Removing the Y Wobble


3dpintmi
(a) Parallel bars
3dprintmi
(b) Z Motor mount
3dprintmi
(c) Z Motor mount
The solution I came up to stabilize the frame was by simply laying another 2020 parallel to the bottom frame (a). I redesigned the Z motor mount (b,c) to hold the 2020 and it came out really sturdy. Adding more function to a part reduces the part count needed.


New Y Gantry 

3dprintmi
(a) Y End
 With the new framing in place, I had to toss the Y Gantry setup which were the same parts used on my original 3DPrintMi. I didn't want to keep wasting my limited stock of 2020s so I challenged myself into designing a Y gantry using only one. What I came up with didn't look very robust at all (a) but after bolting it on and applied some tension to the belt, it remained seated. 


Top Z Stabilizer

3dprintmi
(a) Top Z Stabilizer
Replaced the top threaded rod with a 2020. The threaded rod at the top was meant more to hold my spool while providing some rigidity to the frame. It was a bit too bulky and ugly to look at so I designed a new Top Z Stabilizer (a) that turned out better and requires less plastic.


Tangle free spool mount

3dprintmi
(a) spool mount system
3dprintmi
(b) System with spool

(c) Spool mount feed tube
The spool mount is a re-imagination of my previous spool design and mounted on the 2020 (a). It was originally placed in the middle with the filament going straight down to the extruder but later moved to the far left (b). I have this issue where my spool would dismount during mid-print. I thought it was from the vibration shaking it off but it realized that when the spool nears be empty, it becomes lightweight. Coupled with the filament moving left and right during printing, it will fly off. 

I came up with this filament feed tube guide (c) similar to the MakerGear M2 setup. The filament enters the feed tube that's fixed in place but free to move around at the other end. This setup removes outside forces to the spool and firmly seated.


Re-designed Vert-Xtruder 

3dprintmi extruder
(a) Vert-Xtruder with ducted fan
The new Vert-Xtruder. I modified a MBE (Makerbot Extruder) lever to pair with the Vert-Xtruder motor mount. On the right is ducted fan for cooling PLA prints. Using a blower fan that delivers a good amount of air to the build plate. 

3dprintmi
(b) Modified MBE paired with Vert-Xtruder
Redesigned the Vert- Xtruder motor mount to be completely flushed with the MBE lever. This is necessary for printing with flexible filament where it tends to buckle during mid-print. I'll be uploading the new design on the thingiverse page.
3dprintmi
(c) Ducted blower fan
Closer look at the ducted fan setup. I plan to bring the duct closer to the nozzle. It works really well printing with PLA but I the fan is so strong that I have to increase the nozzle temperature to 240C to prevent the extruder from jamming due to the nozzle cooling down too much


That's it so far and I still have a lot more to do. Here's a few things I'll be working on in the next few days:


- Finalize the BOM
- Redesign the ducted fan
- Start on preliminary designs for auto leveling probe  setup




























Monday, May 19, 2014

Evolving the 3DPrintMi

I have a new 3DPrintMi in the works with the intention for easier construction, reduced printed parts and cheaper BOM (hopefully) but still retains the ability to adapt to the current RepRap Tech. As for the current design, I will now stop updating new parts for it and upload the final BOM here on my build blog. I have two BOMs available, one with vendors from the USA and the other will have vendors for international users.

Since my last post, I've been busy getting this 3D printing start-up up and running so I haven't had much time to update anything here. With new developments underway, I thought it was about time to finally load up my notes before I forget again.

The hackerspace I had my 3d printers at decided to shutdown recently and now they are all gathered up at my start-up. It's the first time I had all 4 sit together. I do have a 5th 3DprintMi but he's sitting at home right now.

All four 3DPrintMI sitting side by side. 5th one sitting at my place.

The Mk. I.  Same printer I used for my video assembly clip. This one is most quiet out of all my 3DPrintMis. It still sports the QU-BD MBE clone which works great with ABS, not so great with PLA.

The test build using the cheapest components available from different vendors. Most of the electronics came straight from the East, to be more specific, robotdigg.com and geeetech.com. Still haven't made the maiden voyage to print yet since I had issues with the anUBIS hotend (Ubis clone) but may switch it out for the Hexagon Hotend from RepRapDiscount.

A made this for someone but I later re-acquired it. Took me 3 days to print all the parts and was up and running soon after. It made a few awesome prints but I parted all the components for the new build.. 

The new prototype in the works! After getting into the Misumi's "Reddit150" deal, I was able to get my hands on some aluminum 2020. I really like it and came pre-cut. This will also be the flagship printer for my "Build a RepRap" class. I was able to import the X gantry and the Y plate to the new prototype. I am working on designing a fan mount for PLA prints and a new Z adjuster. No plans for auto-leveling since that's new craze right now but still experimental.


For those wondering what the start-up is all about, it's a 3D Printing studio in Waltham MA on historic Moody St.  It will be a go to area for people wanting to learn everything about 3D printing and I'm really excited about it. For more info, come check out the site

www.ThePrintingBay.org


Before I forget, Here are the new BOMs




Tuesday, January 7, 2014

3DPrintMi turns one year old


Happy Bday to 3DPrintMi. 

Haven't touched this blog in a while and it's long overdue for a post. New developments for the 3DPrintMi has been put on hold for now. I've been pretty busy lately since I'm currently involved in a 3D printing startup. Launching soon! When I have time, I usually work on the 3DPMi but I really need to prioritize what I want to get done since I keep expanding my laundry list of new ideas. Actually, the only thing I should work on is implementing a duct fan mount for PLA printing. Another thing I've been doing is printing parts for customers through MakeXYZ.com. I didn't think I would get any customers through that site but after adding a few pictures of my work, people started asking me to print a few jobs. To date, I have made about $400!

The only project I have going on right now is completing my 4th 3DPMi. The difference here is that I'm going for the absolute cheapness build. My main goal is to get the 3DPMi price down to below $300 and release the BOM with all the vendors listed. There were a few sacrifices that had to be made for this build.  It would have to be PLA only printer with no hot plate installed. The build plate was also limited to a 6" x 6" aluminum plate with window glass.

After much browsing and scouting, I realized that the motors, board and the Hot-end makes up most of the cost of a RepRap printer. The hardware is not a big deal to me, cost around $90 and it's a bit inflated too since some of the purchases come in a pack of 100 which is more than enough. The two vendors that really made a dent in the pricing was Robotdigg.com and GEEETech.com. For the hotend, I went for the QU-BD's anUBIS hotend which I will write about down below. I will now show you what I purchased from each site

Motors + additional hardware
Robotdigg sells Nema17s for $6.80 a pop which really knocks down the price. They also sell other hardware such as LM8UUs, GT2 pulleys & belts, 608s skate bearings, etc for a cheap price. They are now offering chrome plated smooth rods and 200mm x 200mm hot plates at a bargain price. The only con with them is that their paying system is a bit outdated. You make an order, they email you back with the shipping price (by weight) and then you pay through Paypal. You never get some sort of shipping confirmation so you basically sit and wait for 2 weeks, hoping you didn't get scammed. Not to worry since they deliver as promised and the quality of the parts is quite good. Shipping prices depends on the weight but it's as bad as I thought. When I ordered 10 motors and enough GT2s, LM8UU, & bearings to build 4 reprap printers, the shipping came out to be around $60.

Electronic Board
GEEETech has a huge list of boards to purchase from. They have other 3D printer products but it's a bit pricey. The only thing I get from them is their cloned Printrboard since it's so easy to use. The initial price is $89.99 but sometimes they have weekends/holiday specials, knocking 50% off the initial price. You can't go wrong with a $45 board! Although, when I purchased 4 of their boards at the time, they were dinky 2 layer pcb boards that can't even handle more than 10amps. Ugh! The trace wire burnt out on all four but easily fixed with a jumper wire. I recently heard they resolved this issue and updated their current stock with beefier ones.

Hotend
The cheapest Hotend on the market was the QU-BD's MBE v9 for $34 w/o motor. Having dealt with two of them, it totally sucks with PLA. I looked at the Jhead but it was a bit pricey for a 1.75mm version. I had my eye on the Makerfarm's Jhead for $55 but then came along the anUBIS hotend from QUBD. It's a complete copy cat of the Ubis Hotend from Printrbot but for $29.99, i just couldn't pass this deal so I bought two to play with. When recieved, I was expecting it to be the ceramic version (no pictures provided) which I'm more familiar but instead got the aluminum block version.  It works OK but not as good as the ceramic UBIS hotend. I can't compare with the Printrbot's aluminum block version but I'm not too fond of it. It takes a good amount of force to extrude by hand and the filament will randomly slip from the MK7 filament drive gear when extruding at a constant rate. I'm already pushing 220C for PLA and it's not looking so great. Honestly, I have no idea how it can't perform the same as the UBIS, it's almost identical. Well, I will have to print real slow with this guy and give it a go. I still have a E3D on backup.

If you're all wondering how much this machine cost me, here is the price
$231.73

BOM RIGHT HERE

If I add the hot plate from Robotdigg ($9.80), it only adds another 5 bucks to the total price. I didn't account the shipping price into this but if you wanted to keep it under $300,you have $70 to spare for shipping. Just don't mess up your orders!


The Cheap 3DPMi. The red parts were printed during the weekend of Makerfaire NYC. It survived the wind!
Robotdigg Nema 17 motor

Drill rods from McMaster

LM8UUs from robotdigg. The new X-Ends using zipties as a fastener.

Skate bearings and different angle of the zip tie mount

Super cheap Printrboard clone. GEEETECH is the same vendor where QUBD uses for their OneUp & TwoUp printers.

152mm x 152mm aluminum plate. I will have to add a cheap window glass since it's slightly warped.

My direct drive extruder I designed, the Vert-Xtruder paired with the modified anUBIS hotend. Replaced the aluminum core with a MBE heater block.
Vert Xtruder closeup

Modified anUBIS. This version works OK but still tough to extrude

4th 3DPMi sits idle for now







Monday, September 30, 2013

Build Log #38 - New BOM and X Ends

Just uploaded version 4 of the Bill of Materials. Some of the prices have changed but added a few more options as well. I replaced the Wade Extruder in favor of my Direct Drive Vert-Xtruder. It requires lesser parts and was designed to be compact. It is also compatible with multiple hotends and mounts on Wade Extruder mounting holes making it a versatile extruder. 

Get the new BOM here!

The X Ends have been redesigned from a press fit to a zip tie fastened operation. The design was actually inspired from the Printrbot Simple. I was able to run into Brook Drumm, the creator of Printrbot, at the Open Source Hardware Summit and had a nice long chat with him. We started talking about the Printrbot Simple and asked about his design decision in using zip ties to fasten the linear rods. Brook stated that even though he cut a lot of corners with Simple, zip ties works surprisingly well in fastening the linear rods. I figured I could do the same with the 3DPrintMi's X Ends since it takes a bit of muscle to press fitting the linear rods. The new X Ends is a big improvement, cutting 25% of it's original volume and knocks 30 minutes out of the print time. It requires 8 zip ties to fasten the rods but it's definitely a lot easier to assemble. Thanks for the advice Brook! 
X End Motor ver2

Last week, I exhibited the 3DPrintMi at Makerfaire NYC and got a lot of positive feedback.  I had a lot of fun besides being covered in dust by the end of the weekend. 

Nicholas Seward with his RepRap Simpson & Reprap Wally

I finally got to meet Nicholas Seward and his two repraps, the Simpson and Wally. He's a great guy and I really got to give him props for developing two completely different repraps in matter of months. The Simpson is really fun to watch and it's something out of a sci fi movie, sorta reminds me of the tripod monsters from Half Life. Wally looks very elegant yet sturdy. It's not as crazy as the Simpson but it's still unique in it's own rights. It was great to witness it come alive and print it's first piece. Nicholas was having a bit of trouble getting it to work at first since he told me the TSA handled it like a ragdoll during check in. Watching both run at the same time was definitely mind boggling. The reprap Morgan did make an appearance but not the creator, Quentin Harley. A few fellows from Minnasota? came all the way down to NYC to exhibit the RepRap Morgan. Another great machine and well deserved first place Gada Prize winner. Before I left, I was able to get a picture of all three Gada Prize winners together. 
Gada Prize winners!

More pictures here

Also want to say thank you to the people that sent me photos of their 3DPrintMis to showcase. People were really impressed that it can be repeatable. 




If you guys made it this far, if you can be so kind of to fill out a quick survey for me, that would be awesome. 

Wednesday, September 11, 2013

Build Log #37

Time for a quick build log. Logged about 10 hours of print using the 3DPrintMi Plus with the E3D Hotend and it prints great! I made a few changes to the new Direct Drive by adding a filament guide since the filament buckled as predicted. The only thing I dislike about the E3D is the fan mount. It doesn't have a tight grip on the heatsinks so it just vibrates, creating a god awful noise. I'm going to have to ziptie down because I can't have that type of noise during Makerfaire NYC.

Speaking of Makerfaire, I will be exhibiting the 3DPrintMi this year in NYC. I'm pretty excited and I hope 3DPrintMi gets at least some publicity. I really don't think the RepRap community knows about the 3DPrintMi, even after it placed third in the Gada Prize. When Josef Prusa did his presentation, "State of the RepRap," at the Open Hardware Summit 2013,  I was pretty disappointed that he only displayed the first and second place Gada prize winners, the RepRap Morgan and RepRap Simpson. For whatever reason, I just hope 3DPrintMi will make an impression at Makerfaire NYC. Please come see me at the 3D Printer Village!
Exhibit #11402

Also, I revived my first 3DPrintMi and gave it a new home. It now resides at the Hackerspace, Ideas to Things. 
3DPrintMi Mini

For those wondering about the Direct Drive extruder, it's now up on Thingiverse. So far, it only has the E3D mount but working on a Ubis and JHead. The Ubis is currently being tested and been hearing good things so far.
Vert Extruder



Little sneak peak on the re-designed X Ends I'm working on. 
Re-designed X Ends