Close

Does this project spark your interest?

Become a member to follow this project and don't miss any updates

$300 Pick and Place / 3D printer

An affordable electronics manufacturing system for hobbyists, students, & small businesses. Inspired by RepRap. Powered by OpenPnP/FirePick.

44 494 376
Enjoy this project?
Share on twitter   Share on Facebook

This project was created on 04/29/2014 and last updated 3 days ago.

Description
FirePick Delta is an open-source electronics manufacturing system, inspired by RepRap and powered by OpenPnP and FirePick's own Computer Vision software. We are taking the beginning steps towards a smart appliance that can manufacture electronic circuit boards in a home or office environment. Our machine is able to assemble open-source hardware boards like Arduino and Raspberry Pi accessories, and also has the capability to 3D print. It features an auto-tool changer that allows multiple plastic extruders, and/or multiple SMT vacuum nozzles. Other tools and applications will be available as our product matures.

We will be launching a Kickstarter campaign once we have a working prototype.
Details

In order for this project to be successful, we must define a set of criteria, and then break the problem down into a set of little problems that can be tackled, one by one. 

CONSTRUCTION

We chose a Delta robot configuration over a more conventional Cartesian design.  This allows us more speed, less BOM cost going bearings and linear slides, and easier build / configuration.  This will be covered in more detail in a future post.

I've taken the RepRap philosophy to heart.  There are no swiss-machined parts and crazy Japanese bearings in this machine.  I've spent weeks pulling my hair out trying to figure out the cheapest way to build  a PnP machine without sacrificing quality.  The original RepRaps used 608 bearings and LM8UU linear bearings because they're the cheapest, most-available type out there.  They built their machines around these parts, rather than designing the machine first, only to be pigeon-holed into a weird bearing size.  The best part is that the popularity of the RepRap movement has brought down the prices of these parts even further. 

I've designed FirePick Delta around the following parts:


ACCURACY AND SPEED

SMT parts are getting smaller and smaller every year.  If we made a crappy machine, it would be outdated by the time we'd finish it.  We've taken the proactive approach, and created a set of simulations that test various configurations for accuracy and speed, that uses monte-carlo trials and evolutionary algorithms to "find" a correct delta geometry that fits our needs.  We are attempting a design that *should* be able to place 0201 components!  Although we'll be happy if it places 0402's and QFN's.  I'll be talking more about this in a future post.


OPEN SOURCE vs. CLOSED SOURCE

We will keep it open and pay it forward. This gives us the ability to leverage the following Open Source projects in our design:

  • OpenPnP - Is a project to create the plans, prototype and software for a completely Open Source SMT pick and place machine that anyone can afford.
  • FireSight - A high-level computer vision framework designed for Pick and Place machines, powered by OpenCV. No programming experience required - A pipeline of image operations is specified with a JSON structure. The results of the operations are returned as a JSON structure.
  • FireFUSE - FireFuse is the FUSE driver for all FirePick machines. FireFuse maps all hardware input/output functions for FirePick to individual files in the /dev/firefuse virtual f Read more »

Components
  • 1 × Raspberry Pi Model B Soon-to-be replaced with the Raspberry Pi Compute Module!
  • 1 × Raspberry Pi camera Used by OpenPnp / FireSight / OpenCV
  • 1 × Arduino-compatible motion controller board Runs modified RepRap Marlin firmware
  • 1 × FirePick Delta frame Misumi HFS5 20mm aluminum extrusion, 300mm x 300mm base
  • 3 × FirePick Delta Motor Drive Assembly NEMA 17 stepper with custom printed GT2 belt reduction and delta arms
  • 6 × Delta rod end links, ~270mm Carbon fiber rods with Traxxas 5347 ball links
  • 1 × Delta Pick and Place Head With custom Auto-Tool Changer mechanism
  • 1 × PCB Tray
  • 11 × Component Tape Feeders Holds up to 11 SMT feeders that each hold up to 4 parts each. 3D printed from ESD-safe conductive ABS plastic.
  • 1 × Power Supply Standard ATX 400W power supply

Project logs
  • FirePick Delta is going to OSCon 2014!

    9 days ago • 2 comments

    OSCON 2014


    It's official, FirePick Delta (aka the $300 Pick and Place / 3D printer) will be at the OSCon Open Hardware Showcase!

    I was approached by the folks at O'Reilly Media (the ones that make the computer books) about being a part of their convention.
    They've invited us to their Hardware Showcase, of which there will only be 12 Open Source Hardware projects displayed. They're primarily an open source (software) convention, which is pretty huge.
    We've been going back and forth for a month trying to get everything arranged, and it's now official, I've got my plane ticket from Florida to Oregon booked. It's times like these that I'm glad FirePick Delta is small and portable. I'm going to attempt to bring it as a carry-on, some assembly required :) I just need to remember my hex wrench. Actually note to self, make sure the TSA isn't going to think I'm a terrorist with a carry-on full of unassembled electronics. That's not good.

    Who:
    Neil Jansen on behalf of the FirePick team. Anyone at all interested in seeing FPD, that live around the Portland area, that can make it is invited to come. I believe the expo hall admission is $25.

    What:
    FirePick Delta @ OSCon Hardware Showcase

    Where:
    Oregon Convention Center, in Portland, Oregon
    http://www.oscon.com/oscon2014/public/cfp/336

    When:
    July 22-23, 2014

    Why:
    Because Open Source! Because hacking!


    Some stuff that I'm feverishly trying to get done before I fly out:
    * Getting the feeder vision working
    * Getting the machine otherwise-usable
    * Getting a custom FirePIck sign made from laser-cut colored acrylic (in place of an expensive banner)
    * Getting some business cards made up



    I'll post some pics here during/after the event, of the booth setup and anything else that's cool.

  • 7/12/2014 - Let's Take a Selfie!

    10 days ago • 0 comments

    Robots need to take selfies too! 

    In this post, I'll explain the importance of up-looking vision, and how we cheat with a mirror 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.

    Read more »

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

    17 days ago • 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 »

View all 7 project logs

Discussions

Boz wrote 9 days ago null point

Awesome! Gets my vote if it works

Especially love the idea of a optional solder paste dispenser to complete the package.

Doing solder paste by hand is messy and time consuming and requires a real steady hand, and having a good template made is so expensive.

Are you sure? [yes] / [no]

Minimum Effective Dose wrote 9 days ago null point

The first time I paid for a steel solder stencil was also the last time. Rough-cut edges, and the "solder paste squeegee" included in the $20 accessory add-on was a scrap piece of steel with tape on the sharp edges. F that.

Now I laser cut them out of Mylar (which is a trade name of DuPont, look for "Dura-lar" at art supply / drafting supply stores.) Much faster, and literally 200 times cheaper.

Are you sure? [yes] / [no]

Pixel Pirate wrote 11 days ago null point

So... can it place BGA packages? *shifts eyes left and right*

Are you sure? [yes] / [no]

Neil Jansen wrote 10 days ago null point

Yes, we are designing it in such a way that it should be able to place BGA packages, and have computer vision to make up for various tolerances in placement and rotation. But we're still verifying how accurate we can actually place. We'll be ordering a few BGA test kits from Topline, that include dummy BGA's and PCB's that are wired in daisy-chain pattern through each pin. That will let us know if we can do it consistently, without breaking the bank buying expensive BGA's and writing verification tests.

Are you sure? [yes] / [no]

D1plo1d wrote 20 days ago null point

Greetings from RetroPopulator, the new team on the open source pick and place block!

Check it out, our RetroPopulator's already done it's first successful pick and place demonstration: http://hackaday.io/project/1605/log/4353-retro-populator-in-action

*And* that was all done from a .brd file automatically parsed by our PopCAM software!

Ok, so it's time for some friendly competition. The ball's in your court now and all that. Can't wait to see your demo! :)

Are you sure? [yes] / [no]

Juha Kuusama wrote 19 days ago null point

Greetings from LitePlacer, the third pick and place project in the competition, too! My project is at http://hackaday.io/project/1755-LitePlacer---a-low-cost-Pick-and-Place-machine

May the best picker-placer win!

Are you sure? [yes] / [no]

Neil Jansen wrote 18 days ago null point

Great, we love competition! Don't get too comfortable, although we've been quiet the last few weeks, we've got some really cool stuff in store. We've got working vision, AND rotation! Not to mention a bunch of other cool stuff.

Btw, the only thing we love more than competition is collaboration! Don'tn be afraid to get in contact with us if you would like to work towards standard feeders or vision software, or anything else.

Are you sure? [yes] / [no]

D1plo1d wrote 14 days ago null point

For sure! We've got a lot of basic testing left to do but I think once we've got reliability and setup sorted out a bit more we should start looking at how the two projects can cross-pollinate.

PS. Check it out, we just pick and placed 2 boards in one run at full speed: http://hackaday.io/project/1605/log/4715

Are you sure? [yes] / [no]

Blecky wrote 10 days ago null point

Cheeky :P
Keep it up guys, a little feather ruffling makes for an interesting competition.

Are you sure? [yes] / [no]

Krinkleneck wrote 23 days ago null point

Call me crazy, but if the designs are mainly 1 sided pcbs then you could make it have a heated bed. That way you can place and solder in one machine maybe? Or, you could have the machine move the boards directly into an oven to reflow. You could connect boards to a conveyor system that starts with blank pcbs, they go under a roller with the pcb pattern, through a slow etching bath, through a preset robotic QC for connectors, into a tinning solution, under a soldermask roller, under a masked uv source, cleaned, More QC to make sure the connectors aren't covered, through a pasting station, in and out of the placement, through an oven, colling station and one final qc to make sure everything is connected.

Are you sure? [yes] / [no]

zakqwy wrote a month ago 1 point

Hi Neil,

I'd love to look through your CAD model in detail. What are your thoughts on putting all project files on a shared repo for folks to freely download and open? It seems like you've put a lot of time in to designing the linkages (amongst other parts) and I think we could all learn a lot (especially those of us that are new to kinematics).

Thanks!
Zach

Are you sure? [yes] / [no]

Neil Jansen wrote a month ago null point

My apologies for not having the 3D CAD files up by now. There are a few things we're working on to get them posted. #1 we're moving some of our repositories around to be a little more self-explanatory and easier to maintain. We've created a github 'org' and we are in the process of forking to the firepick org and deleting the old ones under Karl's firepick1 account. #2 we are trying to finish documentation and part numbering so that people know what they're building, and how many parts to print. #3, I'm putting under a yet-to-be-created Open Source license that I'll explain soon. Hope that's understandable. We have five teammates now, working full-time to get all of this stuff done. They should be posted very soon, I will let everyone know when they go live.

Are you sure? [yes] / [no]

Robert_sl wrote a month ago 1 point

Requesting 3d files for the printable delta section :)

Are you sure? [yes] / [no]

Neil Jansen wrote a month ago null point

Hi, we'll have the files posted soon... We're trying to catch up on documentation and other stuff. I'll make a post when we have files available to download. Thanks for the interest!

Are you sure? [yes] / [no]

Robert_sl wrote a month ago null point

I'm interested in building a reprap with the delta setup you currently have. How suitable do you think this design is in this respect? Have you tried printing with it and how is the accuracy and speed?

Are you sure? [yes] / [no]

Robert_sl wrote a month ago null point

I did a bit more reading and it looks suitable to me.. see my new comment above :)

Are you sure? [yes] / [no]

Kiltjim wrote a month ago null point

I've got something like 10 years in manufacturing on pick and place equipment, electronics, and coding (probably not much help here), but if I can lend a hand, I would certainly love to see this become a reality.

Are you sure? [yes] / [no]

Eric Evenchick wrote a month ago null point

There's devices out there to print PCBs at home, but assembly is the really painful part at this point. Really looking forward to seeing this device progress, and glad that you've set some ambitious goals for the component sizes.

Are you sure? [yes] / [no]

bremenpl wrote a month ago null point

Hi there i am really impressed by the work youve done. I work in a company that creates small factor pick and place machines. I am in RND departament taking care of the low level programming and hardware. I would surelly like to contribute to your work, it has hudge potencial.

Are you sure? [yes] / [no]

josheeg wrote 2 months ago null point

I am interested in trying your computer vision setup what components of this design can be bought and tested for it? Rasbery Pi and Pi camera?

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Yep, just a Raspberry Pi, the camera, and the FirePick1/Firesight github repository linked above (the documentation is in the github wiki).

Are you sure? [yes] / [no]

dumi.gumede1 wrote 2 months ago null point

Printer*

Are you sure? [yes] / [no]

dumi.gumede1 wrote 2 months ago 1 point

Doesn't the project has to be innovative ? Anyone can get a 3d ptriinter

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

You're right. Anyone can get a 3d ptriinter. We must not be innovative.

Are you sure? [yes] / [no]

willyboi111 wrote a month ago null point

Please don't hate!! Also this is not a 3d printer! Very well documented and the Innovation IS THERE!!!!!!

Are you sure? [yes] / [no]

Mike Szczys wrote 2 months ago null point

You have an amazing jump on documenting your work. I'm so glad you entered it in The Hackaday Prize.

Keep doing what you're doing!

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Thanks :) I've just begun the documentation process.. There's so much stuff that I still need to post. Thank you guys for creating a reward system that motivates hackers to document the stuff they build.

Are you sure? [yes] / [no]

egoodchild wrote 2 months ago null point

Hey awesome project. I noticed that you guys are looking for help on this project. I would like to help you guys make it a reality in any way I can, I to would like to see a sub $500 pick and place on the market that is easy to use.

I have a full electronics lab at my disposal at my house, with all of your standard lab equipment scope, sig gens, soldering, reflow. I will also soon have an 80W laser cutter, once I finish building it. I also have access to mills, high dollar 3D printers, epilog laser cutter and CNC mills.

I’m very adept at multilayer high speed and high power PCB layout, FPGAs, MCU, and other embedded hardware/software design.

Keep up the great work!

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Thanks for reaching out, could you send an email to njansen1 at gmail.com? We're always looking for dedicated people to help.

Are you sure? [yes] / [no]

mattf wrote 2 months ago null point

Very cool project. Maybe it is worth considering using a beagle bone black in place of the RPi? The BBB already can run linuxcnc, and has some realtime support so you could maybe remove the need for the arduino in the system. Also, I believe the BBB's processor is more powerful than the RPi, which may come in handy for openCV processing. Just a thought.

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

We like the BBB a lot, and plan to support it (and other Linux SBC's) in the future. The three main reasons we're using the Raspberry Pi right now is 1) price, 2) availability, and 3) the camera is just amazing when unscrewed to macro focus. it's the perfect size and we have lots of manual control of exposure and other settings that other USB cams don't offer. Once we get over these hurdles, there will definitely be support for the BeagleBoneBlack in the future.

Are you sure? [yes] / [no]

rudiahlers wrote 2 months ago null point

I would like to get involved in this project as well, but I am in South Africa. My skills include Arduino, solderings, some coding, and plenty hardware hacking!

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Cool! Send me an email at njansen1 at gmail.com, and we'll talk there.

Are you sure? [yes] / [no]

notexactly wrote 2 months ago null point

A year or two ago I saw a DIY feeder that used just a razorblade to separate the tape, and it seemed to work pretty well.

For motorized tape removal, why do you need a stepper motor? Wouldn't a DC motor with a torque limiter of some kind (slip clutch, or just insufficient power available) work fine?

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Yes, a DC motor with torque limiter, either mechanical or electrical, would work great. We went with one specific approach for our prototype, which is a single stepper motor that drives up to twelve cover tapes, where each tape spool contains a clutch (actually a set screw with a nylon tip). This allows you to simply add another feeder, when you put it in the slot, it will automatically engage the drive shaft from the previous feeder. The stepper motor easily moves the shaft a fixed number of steps, regardless of whether there's one cover tape being used, or twelve (a DC motor's turns would vary with load and voltage). We feel that this is a pragmatic way to make a flexible system on the cheap. Estimated cost of 12 8mm tape slots (3 feeders) and the stepper motor drive is only about $25 to 40. That gives a total of 48 feeders per system for around $130.

Are you sure? [yes] / [no]

KingOfKYA(Travis K. ) wrote 2 months ago null point

I will be watching this project:)
Hope you get it working

Are you sure? [yes] / [no]

Lerche wrote 2 months ago 1 point

Neil, you state you need help. I am rather good with KiCAD and I am an expert in soldering tiny stuff (really, I do 0201 in hand), and I am overly interested in making my own pick and place machine, so why not help out? I do however live in Denmark, long from you, so maybe I can do something from here? Regards, Lerche

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Sure! We've got quite a lot of stuff to work on, and would love to have your help. Send me an email at njansen1 at gmail.com.

Are you sure? [yes] / [no]

Jasmine wrote 2 months ago null point

Great info. Would love to see updates about this, and also see what projects you print.

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Thanks! I'm a bit behind on updates as I'm applying for the HAXLR8R hardware accelerator this week, but I'll hopefully have some stuff posted in the next few days. Maybe even a video of the machine working by the end of the week.

Are you sure? [yes] / [no]

Jasmine wrote 2 months ago null point

Good luck with HAXLR8R. A video would be great.

Are you sure? [yes] / [no]

Neil Jansen wrote 2 months ago null point

Not sure if you saw it already, but I posted the youtube video to the HAXLR8R video in the build log section above... let me know what you think!

Are you sure? [yes] / [no]

zakqwy wrote 3 months ago 1 point

Interested in the "secret sauce". More info on this?

Are you sure? [yes] / [no]

Neil Jansen wrote 3 months ago 2 points

Absolutely. I'm going to start posting bi-weekly build logs starting tomorrow, starting with a project introduction. From there I'll cover the feeder design, nozzle design, custom motion controller, and other pieces in detail.

Are you sure? [yes] / [no]