Project Logs

Collapse logs

• We're No Longer the $300 Pick and Place / 3D Printer

  Neil Jansen • 03/13/2015 at 02:46 • 6 comments

  There's been a lot of cool stuff going on in our project, even though I haven't updated this Hackaday.io page a much in the last six months or so. Here' s a quick summary:

  DEC. 2014 - COMPANY LAUNCHED
  

  Established in December 2014, I started Tin Whiskers Technology, LLC in order to provide a platform to sell kits and finished units to various people around the world. I realized early on that my business skills were lacking, so I brought on Felix Banuchi as a co-founder. Between his MBA and startup experience, and my technical leadership, we're hoping that we can avoid the common pitfalls that open-source companies have made recently (some bad for the community, like Makerbot going closed source, and some bad for the companies like Arduino LLC vs. Arduino SRL)

  The name Tin Whiskers is sort of a dual-meaning. Most veteran electrical engineers and other folks in the world of electronics manufacturing know about tin whisker growth all too well. It's the metallurgical phenomenon where little metallic tendrils seem to grow out of a flat metallic surface. This problem has worsened since the switch to lead-free solders and other repercussions of RoHS. On the other hand, tin whiskers also means something to the lay person, it brings up images of cats, hipster beards, and robots... Which seems very fitting for such a cool company.
  

  MARCH 2015 - PRE-ORDER KITS GO ON SALE

  This week, we've begun selling pre-orders to our initial group of beta testers. We're doing 50 units for the first run, followed by possibly more after that. The idea is to get hardware in the hands of as many developers as we can, so that we can get immediate feedback on a common hardware platform. This is much better than a DIY approach where everyone ends up with different machines. Obviously, hardware incompatibilities, lighting conditions, feeder types, machine accuracy and precision, differences between DIY machines vary greatly... And that added complexity makes software development a real nightmare.. that is, unless we can get identical kits in the hands of our testers. We set out to do that in late December, and now in March, a few months later, we're to that point, albeit a few months behind where we wanted to be.

  PRICING ISN'T $300, BUT IT'S STILL CHEAP COMPARED TO EVERYTHING ELSE

  In order to make this happen, we've had to do a lot of parts sourcing, cost estimation, BOM analysis, design-for-manufacturing changes, etc. I quickly realized that my original estimate of $300 was off by quite a bit. Those that have been following this project for a while, remember that this project was initially titled "the $300 Pick and Place / 3D printer". I REALLY did want to make a machine this cheap, and that number was based on what I wanted the final price to be.. Our BOM cost has hovered from $300-700 depending on options... I figured that assuming I got some wholesale accounts, and get some sourcing lined up, that I could get the price down to 1/2 of that, sell for a bit of a markup, and still meet the $300 goal. Well, after Felix joined the team, I quickly realized that's not the way things work. Once our company was formed and we actually tried to secure these wholesale and vendor accounts, we've not gotten a single 50% price break. That simply won't happen unless we're selling 5000+ units. That might actually happen someday, if I get my wish. We also have to make enough markup as to not go out of business quickly. We're not getting rich by any means -- We'd get better markups manufacturing women's perfume or costume jewelry... Or social media web-apps, e.g. the next AirBnb or Instagram.. Hardware companies are very capital intensive, ours is no different. We've got a lot of money tied up in this, with not much to show for it at this point, other than a lot of hope, and some excited and eager beta testers.

  In the meantime, FirePick Delta will not cost $300. I take full responsibility for anyone that's disappointed in the price difference, by the way. We're even still hesitant to throw...

  Read more »

• Beta Test Kits, Coming to a Hacker Near You...

  Neil Jansen • 03/01/2015 at 03:01 • 0 comments

• ESD Mitigation Strategy

  Neil Jansen • 09/28/2014 at 22:40 • 0 comments

  Electro-static discharge is a big deal when you're working with modern SMT parts. Doubly so when you're dealing wiht SMT component feeders. Things moving around generate a charge, and these charges can even levitate parts right out of the carrier tape.

  We have a two part approach for mitigating ESD:

  Conductive Coating for all Acrylic and Anodized Surfaces

  In the USA, the best source is TechSpray Licron, available from Mouser.  Other countries have this listed under different names.

  ESD-Safe Conductive ABS 3D Printer Filament

  Part two of our approach is to use ABS plastic 3D printer filament, that is conductive, by impregnating it with carbon nanotubes :D  

  We source our filament from 3DXTech, which can also be bought from PushPlastic, where I buy the rest of my normal PLA filament.

  You can buy the ESD-safe filament here:
  

  3DXTech 3DXNano product page

  3DXTech 3DXNano Filament MSDS Sheet

• SMT Component Feeders

  Neil Jansen • 09/28/2014 at 22:32 • 4 comments

  This is a stub... Stay tuned, hopefully I'll have this done before the deadline!

  Modular Feeder System

  Everything on our system is modular, including the feeders.  Rather than having individual blade-style feeders (which are extremely handy, but complicated to pull of on a small scale), we decided on a 60mm base size, that can hold a certain number of tapes / tubes, depending on the component size(s).

  4x 8mm feeders:

  3x 12mm feeders:

  1x 16 and 1x 24mm feeder:

  1x 32mm feeder:

  1x 44mm feeder:

  Not to mention, any size of weird tube, like for a custom choke or coil, or any other weird specialized stuff, new types of feeders can be created, and added to this system with just a basic amount of CAD work and 3D printing.  

  ESD Mitigation

  If you haven't seen our post about ESD Mitigation, please read it here: 

  http://hackaday.io/project/963/log/10095-esd-mitigation-strategy

  Modular Feeder Holder

  All of our modular feeders go into a 3D printed feeder holder, that attaches to the aluminum extrusion rails with an M5 x 8mm button-head cap screw: 

  The curved bit gently guides tape out the bottom of the machine:

  Tape Feeders

  We assume that for home and prototyping use, that most people will not have dozens and dozens of spools laying around.  They're really expensive and not needed for general prototyping.  So we will assume that most people will be loading in cut tape strip pieces.  And there's no reason to bring in the complexity of a full-auto tape feeder if you're just dealing with cut tape strips.

  HOWEVER, it is exremely likely that many home users may have one or two large reels for those jobs where you're laying down tons of LEDs or 0.1uf capacitors, or whatever.  So right now our system has two full-auto tape feeders, so that the system is less tied to using the drag-pin feed.

  Tape Feeder (drag pin feed)

  Tape Feeder (full auto)

  Tape Feeder (strip lanes)

  (pics will be up, check back soon!)

  Tube Feeders

  We plan on supporting these in the very near future.  We just wanted to concentrate on the tape feeders first. :-)

  Tray Feeders

  We don't support JEDEC trays, because they're so big, but sometimes it's still nice to be able to pick up some random parts that either don't have carrier tape, or are on a small piece of cut tape that would be impractical to load in a normal feeder. So, we support little trays that allow you to load in a few parts at a time.  OpenPnP already has some pretty neat logic that knows how to get to the center of a part, based on the tray dimensions, and the part's dimensions.

• FirePick Delta System Design Document and Youtube Video

  Neil Jansen • 08/19/2014 at 05:09 • 0 comments

  Here's our Introductory Youtube video for the Hackaday Contest, Stage 1 & 2, as per the Official Rules:

  Here's our System-Level diagram (still a work in progress):

  Admittedly the system diagram is very busy, but it does accurately reflect our desired system configuration.  I will try to explain the various parts of the diagram and how they work together. 

  Read more »

• Camera and Computer Vision

  Neil Jansen • 07/13/2014 at 14:17 • 1 comment

  In this post, I'll explain the importance of Computer Vision and how we plan to accomplish this. I'll also cover the state of our computer vision system and where we're going from here. 

  In the picture above, you can see a Raspberry Pi camera, looking at itself in a mirror, in order to see the nozzle, and and parts that might be hanging around there.  We can use computer vision to calculate and adjust the part offset and rotation, to make up for the fact that it's impossible for us to pick a part from a component feeder with enough precision.  We can also use computer vision to calculate the offset of the nozzle to the camera, not only at zero degrees rotation of the nozzle, but also for other rotations, which means we can calibrate out the wobble of the cheap Luer lock syringe tip that we use.  The mirror technique has some challenges though, so we will be offering a traditional upward-looking vision system, via a second camera, in the weeks/months to come.

  Read more »

• 6/29/2014-Present - FirePick Delta Failblog and Wins

  Neil Jansen • 07/07/2014 at 05:23 • 7 comments

  This entry will serve as an omnibus collection of fails, and wins, at the system level.  I'd rather keep them together, instead of having dozens of tiny updates, each with their own entry. 

  As of July 8 2014, we are successfully picking and placing SMT components with the FirePick Delta prototype.  This is a huge accomplishment in itself, but we've really just started.  Although FPD has a 5MP camera mounted to the end effector, and a full suite of custom computer vision functions ready to go, we've still got some integration to do before we can use it at a system-level, for drag-pin feeder advancement, up-looking vision, and fiducial recognition.  It's a minor amount of work, but everything takes time.  For today, we're just trying to make sure that we've got all the various parts of the system online and working together.  You have to crawl before you can walk, as they say.

  Read more »

• 6/15/2014 - Delta mechanism simulation and accuracy determination

  Neil Jansen • 06/14/2014 at 14:42 • 10 comments

  In our previous post, we explained what a delta mechanism is, and why we decided to use it for this project.

  This post will hopefully shed some light on how we came up with the arm lengths, ratios, and other parameters that determine how accurate and fast it can be.  There's quite a few examples of delta robots out there, and even a few pages with the necessary math to calculate position.  But there's not really a lot of "why" out there, or a reasoning behind the chosen geometries and ratios were chosen.  I'm not a math guru, but I will attempt to explain why I chose my design as I did.

  Read more »

• Delta mechanism design and Frame construction

  Neil Jansen • 06/07/2014 at 16:47 • 2 comments

  This post will give the reasonings behind the frame and the delta mechanism design.  We're using a pretty unique design, in fact I'm not aware of any successful pick and place machines that use a delta mechanism.  We'll explain why we went against conventional wisdom and chose the delta mechanism.  We'll also explain what the heck a delta mechanism is for the more casual readers.

  Here's what your 'status-quo' SMT pick and place looks like, with  Hackaday judge LadyAda for size comparison:

  Read more »

• FirePick Delta - Introduction and Overview

  Neil Jansen • 05/05/2014 at 03:57 • 1 comment

  Welcome to the first installment of the $300 Pick and Place Hackday.io project build log.  This is the first of many build log entries that we will be posting over the next few months.

  The official name of this project is FirePick Delta.  We felt that the name "$300 Pick and Place" would be a better name for our Hackaday Projects page, since it better conveys to the world what our project is truly about.  Nobody knows what a FirePick Delta is (yet), but most of the regulars at Hackday know what a $300 Pick and Place machine is.  If not, don't worry.  We've got a section below just for you :)

  
  

  Introduction in Plain English

  We are developing a really cool robotic machine that is capable of assembling electronic circuit boards (it also 3D prints, and does some other stuff!).  It uses a vacuum nozzle to pick really tiny resistors and computer chips up, and place them down very carefully on a printed circuit board.  A lot of the parts are really tiny (sometimes only 0.6 mm x 0.3mm).  Sometimes they are really big, but the adjacent pins are so close together that the risk of two pins shorting out makes assembly difficult.  Our design uses a tiny digital camera board and custom computer vision software to ensure that the parts are being placed correctly.  Computer vision is really hard.  Especially if you're not a math major.

  There are existing machines that do this today, but they cost a LOT of money.  They range from $30,000 on the cheap end, to several hundred thousand dollars.  Large businesses are more than willing to pay these prices, because they produce millions of boards per year.  But what about the small business?  What about the the would-be tech startup?  What about the hobbyists, the college EE undergrads, hardware hackers, and makers?  It's pretty depressing actually.  Most resort to soldering every tiny component by hand (under a microscope if they're lucky enough to have access to one).  A few hobbyists and hackers went off and tried to make their own machine and failed miserably.  It's an incredibly complicated project by standards, so this wasn't surprising.

  
  

  The Holy Grail

  We are chasing the holy grail of rapid prototyping and electronics manufacturing.  Here are a few examples of the benefits that our project would provide:

  Design phase:

  • Electronics prototyping: Those dreadful breadboards, breakout boards, and Arduino shields can finally go away. Instead, work can begin on the first pcb prototype immediately, which reduces turnaround time. This doesn't work in all cases, but is a nice option to have.
  • Crowdsourced Parts Libraries: We will eventually aim to have a very conclusive library of parts for common EDA tools like EAGLE, KiCad, and Altium.  By crowdourcing the library parts out, and using a ranking and verification system to ensure their completeness and accuracy, the burden to the designer will be reduced tremendously.  This is a huge undertaking, but if our project reaches critical mass, it becomes within reach.
  • 3D printing of enclosures, fixtures, and replacement SMT machine parts.  This machine is capable of replicating its own plastic parts, including ESD-safe feeders and SMT nozzles.

  Manufacturing phase:

  • Quality control can be improved if the boards are being made locally, instead of in another continent.  Problems can be caught earlier.  In many cases, this will actually be much cheaper than outsourcing.
  • Intellectual property concerns: We're all about open source, but sometimes you don't want one of your boards outsourced to a sketchy manufacturer.  By making the boards locally, all of these concerns can be alleviated.
  • Small tech startups: Small companies and tech startups can buy a cheap $300 machine instead of taking out loans or giving up equity to buy a machine that costs 100-200 times that.
  • Just-In-Time Production / Kanban: Allows a smaller stock of finished boards to be...

  Read more »

View all 10 project logs