Building a DIY SMT Pick&Place Machine with OpenPnP and Smoothieboard (NXP LPC1769)

Posted on June 26, 2018 by Erich Styger

This article is about a project I have started back in January 2018. As for many of my projects, it took longer than anticipated.But now it is working, and the result is looking very good: a DIY automated pick and place machine to place parts on circuit boards. In the age of cheap PCBs, that machine closes the gap for small series of boards which have to be populated in a time consuming way otherwise.

OpenPnP Pick&Place Machine

Outline

This DIY machine can populate circuit boards with SMD/SMT parts. If you are wondering, what this is: it is about putting Surface Mounted Devices (SMD) on a Printed Circuit Board (PCB). I gave a talk at the Embedded Computing Conference 2018 on this subject (ECC18 Styger – PickAndPlace mit OpenPnP).

The following videos give an overview about the current state of that machine:

The motivation of building such a machine is that at the university we are populating boards by hand:

Solder gets placed on the boards (stencils are rarely used):

Applying solder

Solder paste on PCB

Then parts get placed on the boards

manual placing

We do have as well vision support for the manual placing:

manual placing with vision

Resistors placed on PCB

And finally they run through a reflow oven:

Reflow Oven

That’s all working fine. The problem is that this all takes a lot of time. Doing it for one board is one thing, but doing it for 10-50 boards takes too much time. Clearly, for mor than 50 or 100 boards, outsourcing is the logical choice. But for a few boards as we usually have to do, outsourcing is not the best option as it is expensive and takes a lot of time too.

What I wanted is a machine to automate the manual placing of components. Such a machine is called a ‘Pick and Place’ machine: it picks parts and places them on a PCB.

DIY Pick & Place Maschine: OpenPnP

When searching for such Pick&Place machines, I stumbled over the OpenPnP project at http://openpnp.org/: A open source community and project which builds such machines. So I thought: why not building one myself too? And here we go :-).

OpenPnP offers a framework to run such a machine. They have guides and tutorials how to build such a machine. And it is up to you how you build it and what features get added. I did not want to build the fastest or the cheapest machine: my goal was to keep the hardware costs below $1000, and that the machine is able to place parts down to the 0402 size.

SMD Sizes

All the software and BOM are available on GitHub (see links section at the end of this article).

The machine uses 24V stepper motors for X, Y and Z axis. Two smaller stepper motors on the head (C1, C2) can be rotated. Attached to the head is a down-looking camera.

Integrated in the work area are nozzle changers, a bottom camera and different feeders.

Under the base plate all the other electronics (solenoid, pump, USB Hub, feeder and controller board with LCD and power supply.

CAD (KiCAD, Altium, Eagle) data is loaded on a host PC running OpenPnP.

System Overview

Below are pictures of the machine under construction with some details. I hope this gives you ideas and an inspiration to build your own machine.

Microcontroller

The heart of the machine is the NXP LPC1769 on the Smoothieboard:

NXP LPC1769

It is responsible for all the sensors and drives all the motors. The picture below shows first tests with a stepper motor:

first stepper

Frame

The frame is constructed with 20×20 and 20×40 standard aluminum extrusion profiles:

Aluminium

The profiles allow an easy construction of the frame.

Aluminium Frame Construction

2040 Profiles

Rails

The linear rails arrived very well oiled and in good shape.

Linear Rails arrived

The rails get attached to the top of the frame:

Rail mounted on frame

It has been an iterative process, and many parts first have been built with a laser cutter and plywood. Below is a picture of an early version:

Rail with end bracket

Vacuum Pump

Actuators in the system are 24V. Below is the diaphragm vacuum pump used for the nozzles:

24V Vacuum Pump

Terminals of the pump:

Vacuum Pump Connectors

The pump together with the two high-speed solenoid on a base plate. To reduce noise, the pump has been placed into a box standing on anti-vibration feet:

Pump with enclosure

Pump Top View

Below a first test with three stepper motors:

Bench

Power Supply

For the 24V 13.4A Power supply I added an on-off switch with integrated fuse holder:

Power Supply with On-Off Switch

Flyback Diodes

As flyback diodes for the pump and solenoids I used the Vishay BYV27-200-TAP:

Protection Diodes

Vacuum Valves

I’m using the STNC TM-06 high frequency valve (24VDC) with 1/8″ pipe connectors:

STNC TM-06 Valve

Below how they are connected to the board with the flyback diodes mounted to the connectors:

Protection Diodes for Solenoids

For example the M813 G-Gode turns it ON (2-1 connected), and M812 turns it OFF (2-3 connected). The image below shows the air flow:

STNC Directions

That means the vacuum pump will be on connection 1 and the nozzle on connection 2, with connection 3 used to release the vacuum.

End Stops

The total of 6 end stops are optical ones:

End Stops

The first (not ideal) way to mount the end stops:

First way for endstops

Second version of end stop mounting with 3D printed holder:

3D Printed End Stop Holder

3D Printed End Stop Holder back side

Mounted End Stop

3D Printed End Stop Parts

3D printing end stop blades:

3D Printing End stop Blades

Mounted end stopper blade:

Mounted End Stopper Blade

In a next step lowered the end stop position:

Lowered End Stop Position

Belt and Endstop

Pick and Place Head

Very first version of the head:

First Pick Head

Holder for the pick stepper motor:

Holder for the Pick Stepper Motor

Rails for the pick head steppers:

Pick Head Rails

Frame with first pick head:

Frame with first pick head

Pick head with end stops:

Pick Head with End Stops

Because the X axis was not rigid enough, 3D printed brackets were added:

3D Printed Bracket

X-Y-Rail

Mounting Bracket for X-Y Rail

T-Nuts on X-Y Rail:

T-Nuts on X-Y Rail

Y Steppers

Two stepper motors wired together are driving the Y axis. It works, but I would not recommend that approach: better use a single motor with an extended shaft and coupling.

Wiring of Y motors

Y Stepper Wiring

Laser cutting the motor and pulley brackets:

Laser Cutting Motor Brackets

Of course it was an iterative process, and many design ideas did not end up in the final machine. The good thing with plywood is that it is cheap and is easy to construct with it.

Pile of plywood

Second Pick Head Design

The second version of the pick head was more compact and easier to attach to the linear rail.

Assembling the Pick Head

Below the head attached to the X axis:

2nd Pick Head

Pick Head 2nd Version

Below an early version of the head with no cable chains attached:

Head without cable chains

I used 3D printed parts to attach the heads with the Z axis belt:

Z-Axis Belt with 3D printed parts

The belt is attached to the linear rails using a press-fit connection, and the connector acts as an end stop the same time:

Press-Fit Belt Holder

Here with the rails, belt and endstop mounted:

Head Rails and Endstops mounted

Here the head with the pick heads:

Pick Heads

Downlooking Camera

Below an early design of the downlooking camera integrated into the bottom of the head:

First version of downlooking camera

First down camera and light tests:

First Camera Tests

First Camara and LED ring holder:

First Camera Holder

Head with Camera Holder

A plexiglass with a sheet of whiter paper acts as diffuser:

LED Ring diffuser

Pick heads with downlooking camera:

Pick Heads with downlooking camera

Mounted head with first cable chains:

PnP Machine

Metal Pick Head

The next iteration oft the pick&place had is a version built with aluminium profiles. The first step was to use aluminium profiles for the head mounts:

Aluminium Head mounts

Aluminium Motor Mounts

Then the head backplane has been built with aluminium profiles:

Aluminium Profile Pick Head

That way the head was more sturdy than the one made of plywood.

Aluminium Profile Head

Below with the hollow shaft stepper motors mounted, plus a cable chain for the end stops, LED ring and camera USB cable on the left:

Aluminium Profile Pick and Place Head

Machine with Metal Head

Nozzle Holder

Nozzle Holder, magnet holds the nozzle:

Nozzle in Holder

Nozzle and Nozzle Holder

Improved nozzle holder with 1.5 mm space between the plates and acrylic on the top:

Nozzle Holder with Acrylic

Nozzle Holders mounted on base plate

NozzleHolderLabel

Different Nozzles

Cut Tape Holder

See 3D Printed SMT Cut Tape Holder for more details.

SMD Strip Holder

Cover

Tape Holder with Magnets

Tape Holder 3D Model

SMD Tape Holder

Feeder

The automatic feeder is designed by Simon Huber. The goal was to create a feeder for SMD parts on rolls.

SMD Resistors

The feeder uses 3D printed parts:

3D Printed Feeder

Feeder with SMT Tape

Each feeder uses a NXP K20DX128 (ARM Cortex-M4) as the controller:

AutoFeeder

One DC motor moves the tape and one DC motor peels the cover:

Feeders removing cover

A tinyK20 (NXP K22FN512, ARM Cortex-M4) receives M-Codes from OpenPnP and passes them to all feeders:

tinyK22

The feeders are in a daisy chain, and the machine has space for up to 16 feeders.

Nozzles

The first nozzle stepper motors did not work well. The M5 screw did not match the stepper motor hollow shaft. The designed adapter had too much wobble and was not usable.

First Pick Head with Nozzle Adapter

So I had to put aside the nice and small hollow shaft stepper motors. I ordered new (bigger) ones so I can easier attach the nozzle changes and a vacuum tube connector on the back. The new motors had on both sides an M5 screw so I can attach rotation tube adapters and attach easily the nozzle adapter.

New Nozzle Stepper Motors

Because the motors were bigger, I had to create new motor holders:

New Nozzle Motor on the head

Pick Heads

Cable Chains

The first cable chains were too small to hold all the tubing and cables, so a new one has been installed.Below the machine with the first (too tiny) cable chains:

PnP Machine with small cable chains

Small cable chain attached to the head:

Small cable chain attached to the head

Machine with small cable chains

Machine with second version of head

New cable chains arrived:

New Cable Chain arrived

That larger (40×15 mm) was able to keep all the cables and tubing.

Bigger Chain holds it all

Below with the new chain attached to the head:

Head with new cable chain

Cable Chain

Side View with Cable Chain

Base Plate

Machine with new base plate

The bottom plate with cutouts for the feeders and the bottom camera:

Base Plate ready for paint

It gets painted with three layers of magnetic paint:

Magnetic Paint

Painted Base Plate with Magnetic Paint

Bottom Camera

Laser Cutting the Enclosure for the Bottom Camera

Assembling the bottom camera enclosure:

Assembling bottom camera enclosure

Assembled bottom camera case:

Bottom Camera Enclosure

Testing the camera:

Testing the Bottom Camera

Up camera mounted into base plate:

Uplooking Camera (Bottom Vision) in Base Plate

Bottom Vision

With the bottom camera, OpenPnP can move a part over the camera to correct angle and position with the vision system:

Bottom Vision Capture

It uses a vision pipeline, below with a blur filter applied:

Bottom Vision Gaussian Blur

Masking:

Bottom Vision Circle Mask

getting the part outline:

Bottom Vision Threshold

Getting the contours:

Bottom Vision Draw Contours

Detecting part position and angle:

Bottom Vision Rotation Rectangle

OpenPnP includes a loose part feeder: parts can be put into a bin and the vision system can identify it:

LoosePartsFeeder 1

LoosePartsFeeder 2

LoosePartsFeeder 3

LoosePartsFeeder 4

PCB Holder

As the top camera has rather small focus area, I need to keep the feeder pick (see “3D Printed SMT Cut Tape Holder“) height and the PCB surface on the same height. For this I have created 3D printed magnetic PCB board holders.

3D Printed PCB Holder with Red PCB

3D Printed PCB Holder

Graphic LCD

Similar as for my laser cutter (see “Upgrading a Laser Cutter with Cohesion3D Mini and LCD“) I added a graphics LCD to the board.

Machine with LCD Display

http://smoothieware.org/rrdglcdadapter describes the settings and installation of a graphics LCD on the Smoothieboard.

I decided to order an adapter board from 3DWare:

GLCD Adapter Board Top Side

Below the bottom side of the board:

GLCD Adapter Board Bottom Side

Below how the adapter board is installed on the board:

Location of GLCD Adapter on Smoothie

Below the display attached to the board and working :-):

Smoothie with GLCD

Below the (still messy) electronic parts on a base plate:

Electronics

Added to the display is an emergency stop button. Below the connection to the controller board:

Emergency Stop Connection

LCD with Emergency Stop

OpenPnP Software

The OpenPnP software runs on the host PC.

OpenPnP Software

The host PC does all the image processing and sends commands to the machine.

Running an OpenPnP Job

Drop Box

For bad parts or to drop parts, OpenPnP can use a dedicated ‘drop area’. for this I created a custom (laser cut) drop box: It uses 3mm red acrylic on the top:

PnP Drop Container

4mm plywood on the bottom with magnets so it sticks to the machine surface:

PnP Drop Container with magnets

Plexiglas Brackets

The motor and belt brackets originally were cut out of plywood: here I replaced the 4mm plywood with 5 mm laser-cut plexiglas ones which looks nicer :-):

Plexiglass PnP Bracket

X Axis with Plexiglas/PMMA motor bracket

The machine mounted on a 16mm plywood base plate:

Mounted on BasePlate

Summary

It has been a fun project, and the machine works well, but still needs some tuning. I plan to add a solder paste dispenser: that way, if no stencil is available, the machine can add the solder paste to the pads on the board. Another thing is to use pressure sensors to monitor the vacuum for each nozzle. And for the motor auto-feeder I would like to update the design. So there is always something to improve. Currently the machine can place around 500-600 parts per hour down to 0402 size. With this and little setup, we can run small board series successfully.

PS: a big “thank you!” to the OpenPnP community: without all their work and contributions, such a project would not have been possible.

Happy Picking 🙂

138 thoughts on “Building a DIY SMT Pick&Place Machine with OpenPnP and Smoothieboard (NXP LPC1769)”

Mauricio on June 26, 2018 at 04:11 said:

Hi Erich! That was an impressive project! Congratulations!

LikeLike

Reply ↓
- Erich Styger on June 26, 2018 at 13:25 said:
  
  Hi Mauricio,
  thank you! And it will continue to evolve :-).
  
  LikeLike
  
  Reply ↓
Prakit Junkhum on June 26, 2018 at 05:28 said:

Are you have hardware dimension?

LikeLike

Reply ↓
- Erich Styger on June 26, 2018 at 05:37 said:
  
  It is 50×50 cm.
  
  LikeLike
  
  Reply ↓
sibeliusgriegw on June 26, 2018 at 16:28 said:

Hi Erich. This is such an interesting project. I’m always amazed as to how well you document every process from the beginning. Congratulations!

LikeLike

Reply ↓
- Erich Styger on June 26, 2018 at 17:20 said:
  
  Thank you! In retrospect I should have taken more pictures or even document it better. I hope the files on the GitHub site gets everyone a good start who wants to build such a machine too.
  
  LikeLike
  
  Reply ↓
Darren Moore on June 27, 2018 at 13:43 said:

Hi Eric,

I own a commercial PnP machine and your setup makes me want to sell it and build one like you have 🙂

It’s very impressive indeed.

LikeLike

Reply ↓
- Erich Styger on June 27, 2018 at 16:45 said:
  
  You might be able to run your commercial machine with OpenPnP too.
  
  LikeLike
  
  Reply ↓
Javier on June 27, 2018 at 22:59 said:

Hello,

Thank you for sharing all project. It’s awesome. I followed on twitter.

I see that it is possible to use from 0201, this is ok for me.

But is it possible to use MAP-BGA169?

Kinetis K26 K27 K28

+ one BGA Inspection Tool, as ERSA or BGA-100.

Best regards

LikeLiked by 1 person

Reply ↓
- Erich Styger on June 28, 2018 at 00:43 said:
  
  0402 works well, but 0201 depends on your board and good calibration of the system. I have not used it for any BGA packages, but typical QFN packages work well.
  
  LikeLike
  
  Reply ↓
  - Javier on June 29, 2018 at 15:48 said:
    
    Thanks Erich,
    
    So I’ll use 0402, I only use 0201 resistors to 2 leds but I think that I can to change for 0402. The rest is 0402, 0603, and 0805.
    
    I have seen the ultimaker 3d printer to placed bga chip, I think that with yours design it’s possible.
    
    Regards
    
    LikeLike
    
    Reply ↓
    - Erich Styger on June 29, 2018 at 17:04 said:
      
      Well, “placing BGA” does not really mean much. It is all about how accurately the machine can place components.This is basically defined by the size of the pads.
      
      LikeLiked by 1 person
    - Javier on June 30, 2018 at 20:21 said:
      
      Hello,
      
      About of the noise of vacuum pump, I think that a silent box can serve. For example, but smaller, https://www.youtube.com/watch?v=Bd2nhNf9Tp8
      
      LikeLike
    - Erich Styger on June 30, 2018 at 21:05 said:
      
      Thanks for that link, this is indeed interesting. But it will be challgenging to bring it down in size 🙂
      
      LikeLiked by 1 person
Pingback: A Pick-And-Place That Is A Work Of Art | Hackaday
Javier on July 12, 2018 at 23:15 said:

Hello,

Have you made progress on the solder paste dispenser?

I plan to use T3 and T4 from Chip Quick, SMD291 family.

http://www.chipquik.com/store/index.php?cPath=400

The 3D model is for that syringe?

Is an air pump necessary?

Best regards

LikeLike

Reply ↓
- Javier on July 13, 2018 at 08:14 said:
  
  Hello again,
  
  I have seen that is possible to use air compressor like as http://www.senco.com/tools/details-page/pc1010n or air bottle like as http://www.saxbysdrury.com/images/products/_300x300/air-liquide-aligal.jpg
  
  The important thing is the pressure between 50~80 PSI
  
  http://www.instructables.com/id/How-to-set-up-repair-adjust-and-use-a-solder-paste/
  
  There are also finished economic solutions such as YDL-983A
  
  Best regards
  
  LikeLike
  
  Reply ↓
  - Javier on July 13, 2018 at 08:32 said:
    
    Also exists motorized version
    
    https://www.robotshop.com/letsmakerobots/solder-paste-dispenser
    
    https://hackaday.com/2016/02/03/solder-paste-dispenser-has-no-3d-printed-parts/
    
    LikeLike
    
    Reply ↓
    - Javier on July 13, 2018 at 08:55 said:
      
      Other interesting project is this https://github.com/hzeller/rpt2pnp
      
      and this https://github.com/genaraf/Solder-Paste-Dispenser
      
      Design a pcb in kicad that serves as a support for the syringe and for the stepper motor I think it would be ideal, without using 3D printing.
      
      LikeLike
    - Erich Styger on July 13, 2018 at 16:20 said:
      
      Thanks for the links!
      
      LikeLike
  - Erich Styger on July 13, 2018 at 16:19 said:
    
    that first air compressor looks like really big and noisy (not sure). We do have compressed air available in the lab, so most likely I don’t need any extra pump. Or using the vaccuum pump in reverted mode I’m using for the PnP machine.
    
    LikeLike
    
    Reply ↓
- Erich Styger on July 13, 2018 at 16:17 said:
  
  Not much progress, will continue on that in a few weeks I hope. What I have done is creating holder for the Auger pump and an adapter for the stepper motor.
  That way I can apply a highly defined amount of solder. I got the idea from https://www.youtube.com/watch?v=PeYb3snGrU8
  And yes, it requires some amount of air pressure too.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 17, 2018 at 16:15 said:
    
    Thank you,
    
    I have seen one that dispenses 0.25mm, I need something like that for MAPBGA and the size of the pitch 0.65mm
    
    LikeLike
    
    Reply ↓
Javier on July 13, 2018 at 15:28 said:

This is other interesting open source syringe pump project

http://www.appropedia.org/Open-source_syringe_pump#Source

I have found the syringe of the motorized project, from Nordson website, In case someone is interested, I was looking for them but in the pharmacy the ones they sell are different.

http://www.nordson.com/en/divisions/efd/products/syringe-barrels-and-cartridges

LikeLike

Reply ↓
- Erich Styger on July 13, 2018 at 16:23 said:
  
  I think the problem with all these approaches is that the put pressure on the whole solder in the reservoir, instead pumping it out in a reliable way. To me, still the Auger pump way should work much better?
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 17, 2018 at 16:37 said:
    
    Finally I bought an stepper motor and 30ml syringe. I’m going to try the stepper motor version.
    https://www.aliexpress.com/snapshot/0.html?spm=a2g0s.9042647.0.0.58674c4dQumSeB&orderId=93144902223017&productId=32820896968
    
    https://www.aliexpress.com/snapshot/0.html?spm=a2g0s.9042647.0.0.61764c4dp4DIL6&orderId=93144902213017&productId=32843100869
    
    I am going to try this piece that has 22mm of exterior and it is possible that it enters the 30cc syringe. If so, with two pieces of these and some screws it is possible to build the piece of the video without using a lathe.
    
    https://www.aliexpress.com/item/3D-printer-M8-TR8-8mm-4mm-2mmTrapezoidal-Screw-for-Supporting-All-the-Copper-Nut-Stepper-Motor/32547081654.html?spm=a2g0s.9042311.0.0.3da24c4dF6c1Pv
    
    Also I have paint enclosure box of older pc to the power suppliers and selenoid
    
    https://drive.google.com/file/d/1InrpHGymFbhTM4wvRWe9O0WZ6MwUwYlT/view?usp=sharing
    
    https://drive.google.com/file/d/1zmGLJZf7ts0NAK2GHYTYpXKdEOQBjwV4/view?usp=sharing
    
    https://drive.google.com/file/d/1Esf0M6WUEg97bxcVcHmCjJbeaxac2mvg/view?usp=sharing
    
    I have an IP67 hermetic aluminum box and I am going to fill it with armaflex, I think it could reduce the noise a lot.
    
    https://www.aquatuning.es/refrigeracion-liquida/radiadores/accesorios-chiller/10340/armaflex-sheet-19mm-330x330mm?sPartner=googleshoppinges&gclid=CjwKCAjwp7baBRBIEiwAPtjwxGGL9buO5jf7GDWGwyBCoX2eYTPE3lLuiwo1ry8zypbahJJGNm86WhoC8sYQAvD_BwE
    
    https://www.digikey.es/product-detail/es/bud-industries/ANS-3807/377-1592-ND/1868428
    
    LikeLike
    
    Reply ↓
Javier on July 17, 2018 at 16:11 said:

Hello,

You can update the next in the hardware doc?
https://github.com/ErichStyger/McuOpenPnP_Machine/blob/master/Hardware/McuOpenPnP%20Bill%20of%20Material.xlsx

Line 31 – 2040 Cross support 42.5cm –> https://www.aliexpress.com/item/CNC-3D-Printer-Parts-European-Standard-Anodized-V-Slot-Linear-Rail-Aluminum-Profile-Extrusion-2040-for/32813313600.html

I do not see any of 42.5cm

Line 32 – 2020 Ground Plate Mount 38 cm –> https://www.aliexpress.com/item/HOT-Sale-CNC-3D-Printer-Parts-European-Standard-Anodized-V-Slot-Linear-Rail-Aluminum-Profile-Extrusion/32816279219.html

I do not understand what this is, the url is a rail but there are no measures of 38cm.

Line 28 – USB Cameras –> Which is the size of the focus?
3.6mm
6mm
8mm
12mm
16mm
2.8-12mm(manual)
2.1mm
No Lens

Thanks and regards

LikeLike

Reply ↓
- Erich Styger on July 17, 2018 at 16:35 said:
  
  I have updated the BOM file on GitHub. As for the aluminium profiles: if you don’t get the exact size: order a longer one and cut it.
  For the camera I ordered them with a 6 mm lens.
  The Cross Support is optional. It can be used to strenghten the base plate. If your base plate is very solid, you don’t need it.
  The ground plate mount are optional too. They are used to mount the feets to the second ground/bottom plate I have used. You see the feets mounted to it in the following picture:
  https://mcuoneclipse.files.wordpress.com/2018/05/mounted-on-baseplate.png?w=584&h=411. Corner mounts are used to tie the feets to the 2020 profiles, and the 2020 profiles are attached to the wooden base plate.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 17, 2018 at 16:54 said:
    
    Thank you!
    
    Only one thing, in the line 31 “Optional, order 45cm if 42.5 is not available”
    
    In the url only is available 100mm 150mm 200mm
    
    The url it’s ok?
    
    https://www.aliexpress.com/item/CNC-3D-Printer-Parts-European-Standard-Anodized-V-Slot-Linear-Rail-Aluminum-Profile-Extrusion-2040-for/32813313600.html
    
    LikeLike
    
    Reply ↓
    - Erich Styger on July 18, 2018 at 19:31 said:
      
      Yes, that’s the URL from which I had ordered it. Maybe simply use the shop link instead:
      https://www.aliexpress.com/store/top-rated-products/1799120.html?spm=2114.10010108.100005.1.13935de9RpiNTr&shop_ab=itm2home-4
      As for all these ‘commodity’ items: there are many other sellers of these too.
      
      LikeLiked by 1 person
Javier on July 23, 2018 at 08:45 said:

Hello,

I have bought all the pieces, they are on the way.

I am doing the pieces with a drill and a jigsaw and with wood and aluminum. I do not have a 3D printer or CNC.

Can you share a image of the new pick head made in aluminium?

Also how you placed the endstop and the top camera.

Regards

LikeLike

Reply ↓
- Javier on July 23, 2018 at 08:46 said:
  
  Also can you share the sizes of the aluminium profiles that you used?
  
  I want to go now to the polygon to look for some scraps.
  
  LikeLike
  
  Reply ↓
- Erich Styger on July 23, 2018 at 08:52 said:
  
  Not sure what ‘new pick head’ you mean? https://mcuoneclipse.files.wordpress.com/2018/04/alu-profile-head.png shows it, there is no magic behind it. Just two L shaped profile on the front and two other L shaped ones on the back to mount it on the linear rail.
  The two endstops are mounted on two 3D printed L shaped holders: https://mcuoneclipse.files.wordpress.com/2018/06/pick-heads.png?w=584&h=411
  Top camera is mounted the same way on the back of the head as shown in https://mcuoneclipse.files.wordpress.com/2018/03/first-camera-holder.png?w=584&h=619, just with another L shaped profile attached to the head.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 23, 2018 at 09:04 said:
    
    Sorry, the Metal Head
    
    Yes, it’s L profile, but what thicness did you use?
    
    LikeLike
    
    Reply ↓
    - Erich Styger on July 23, 2018 at 09:06 said:
      
      Thickness does not really matter, as long as it is stable enough. Use 1 mm or 1.5 mm or even 2 mm, whatever you have available.
      
      LikeLiked by 1 person
    - Javier on July 23, 2018 at 09:07 said:
      
      I do not know what thickness to use. In the aluminum warehouse there are “L” profiles with thicknesses of 1.2mm – 2mm 3mm and 5mm
      
      LikeLike
    - Javier on July 23, 2018 at 09:07 said:
      
      Ok,
      
      Thank you.
      
      LikeLike
Javier on July 24, 2018 at 03:42 said:

Update this please,

20T 5mm_Toothed are 6 units instance of 4.

LikeLike

Reply ↓
- Erich Styger on July 24, 2018 at 06:21 said:
  
  I have added a note and updated the file. You need at least 3 (one for each axis) depending on how you do the belt.
  
  LikeLike
  
  Reply ↓
  - Javier on July 24, 2018 at 16:42 said:
    
    In total 7 units, with this:
    
    LikeLike
    
    Reply ↓
    - Javier on July 24, 2018 at 16:44 said:
      
      Today they called me from TNT and tomorrow they delivered the package with 4 units, yesterday I ordered for 2 more but today I will have to order 1 more.
      
      LikeLike
Javier on July 25, 2018 at 04:02 said:

I haven’t BYV27-100 and the electronic local store neither. https://www.digikey.com/product-detail/en/vishay-semiconductor-diodes-division/BYV27-100-TR/BYV27-100CT-ND/3104384

I have SB350 https://www.digikey.es/product-detail/es/on-semiconductor/SB350/SB350CT-ND/4744689

If is 3A but it’s 50V Io-3A schottky instance of 100V Io-2A avalanche.

Can I use SB350?

LikeLike

Reply ↓
- Erich Styger on July 25, 2018 at 07:10 said:
  
  You need good flyback diodes, so I recommend the BYV27-100. Some use the 1N4001 (see Flyback Diodes in https://github.com/openpnp/openpnp-openbuilds/wiki/Build-Instructions although they are not as good as the BYV27-100).
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 25, 2018 at 15:40 said:
    
    Thank you,
    
    I was looking for the BYV27-100 at the local store but they do not have, they do not even know avalanche type diodes.
    
    On the island there are no more electronics stores, I would have to buy it from DigiKey. But there are only 4 diodes and it would take at least a week.
    
    Now, I have IN4007 Reverse voltage up to 1.000V Io-1A and SB350 Reverse voltage 50V Io-3A
    
    Also I have 80SQ045N 45V Io-8A
    
    View post on imgur.com
    
    Drop voltage of IN4007 is 1V and SB350 0.7V
    
    Speed recuperation of SB350 it’s the same of the BYV27-100, = 200 mA (Io)
    
    I think the two could work, but between the two I do not know which could be the best option.
    
    At least for now, for a few weeks I can place an order on DigiKey
    
    Regards
    
    LikeLike
    
    Reply ↓
    - Javier on July 25, 2018 at 15:50 said:
      
      I think that SB350 because IN4007 it’s only 1A and the vacuum pump draw 900-1500mA@ 24v
      
      And my selenoids, they are different from yours, they are not chinese, they are spanish:
      
      https://www.aignep.com/esl/Electrovalvulas-para-Fluidos-FLUIDITY/Serie-01F/ELECTROVALVULAS-DE-ACCIONAMIENTO-DIRECTO-2-2-NC
      
      I use the code –> SOL10 024 C 4 000
      https://www.aignep.com/esl/Electrovalvulas-para-Fluidos-FLUIDITY/Serie-01F/BOBINAS-22-MM
      
      https://www.aignep.com/esl/Electrovalvulas-para-Fluidos-FLUIDITY/Serie-01F/CONECTORES-22-MM
      
      Draw 270mA @ 24V
      
      For the selenoids I could use IN4007
      
      LikeLike
    - Erich Styger on July 25, 2018 at 17:05 said:
      
      Then you have to choose fast diodes matching your machine. Make sure they are very fast.
      
      LikeLiked by 1 person
Javier on July 29, 2018 at 15:52 said:

Hello,

When I insert a nozzle with my hand, I have to do a lot of pressure. Did you eliminate any of the rubber grommets? because in the video it seems to come and go more easily.

LikeLike

Reply ↓
- Erich Styger on July 29, 2018 at 15:53 said:
  
  I had to clean the nozzles and the nozzle holders up, plus used silicon oil to make it more easily to insert/remove.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on July 29, 2018 at 21:07 said:
    
    Than you,
    
    I cleaned it with Soudal Cleaner & Degreaser and I applied PTFE grease, I haven’t silicon oil, but now it’s softer. I suppose that as it is used, the rubber rings will be worn out and it will be even softer.
    
    Cleaner
    
    LikeLike
    
    Reply ↓
Javier on July 31, 2018 at 17:37 said:

Hello,

I have received the 2040x50cm and 2040x15cm profiles.

How big is your base plate?

46cm x 50 cm?

Regards

LikeLike

Reply ↓
- Javier on August 1, 2018 at 00:07 said:
  
  I’m going to use 46cm x 46cm
  
  LikeLike
  
  Reply ↓
- Erich Styger on August 1, 2018 at 07:26 said:
  
  that depends if and where you want to place auto feeders. I recommend that you first build the frame and then measure it.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on August 1, 2018 at 09:58 said:
    
    Not now, but in the future.
    
    LikeLike
    
    Reply ↓
Javier on August 10, 2018 at 09:50 said:

Hello,

What configuration did you use for the DIP switch of the DM542?

And how did you connect the motors, in series or in parallel?

I connected the dm542 to the smoothieboard like this:
PUL- DIR- ENA- to GND
ENA+ to EN5
DIR+ to DIR5
PUL+ to ST5

Best regards

LikeLike

Reply ↓
- Javier on August 10, 2018 at 10:02 said:
  
  In Dynamic Current Settings I have SW1, SW2 and SW3 in OFF in order to use RMS Current 3A, it’s ok so?
  
  SW4 Autotunning in OFF
  
  SW5 SW6 SW7 SW8
  OFF OFF ON ON
  
  LikeLike
  
  Reply ↓
  - Javier on August 10, 2018 at 10:42 said:
    
    Macpick driver connection
    
    LikeLike
    
    Reply ↓
    - Javier on August 10, 2018 at 15:09 said:
      
      Now I understand, if I have SW4 in OFF I get half current so 3A/2=1.5A would not be enough for both engines.
      
      Also, I changed the resolution to 1/16th microstepping:
      
      SW5 SW6 SW7 SW8
      ON ON OFF ON
      
      LikeLike
Javier on August 10, 2018 at 12:23 said:

In Smoothieboard you have placed jumpers in JP28, JP27 and JP11 ?

As it say here

LikeLike

Reply ↓
- Javier on August 10, 2018 at 13:27 said:
  
  Sorry, I have seen in a image that if they are installed.
  
  LikeLike
  
  Reply ↓
Javier on August 10, 2018 at 22:17 said:

Other question , please

You have used 3.3V or 5V with the DM542?

LikeLike

Reply ↓
- Javier on August 11, 2018 at 01:52 said:
  
  I have read in a forum that DM542 can work with 3V3 but to take no risks I have changed it to 5V (common anode)
  
  View post on imgur.com
  
  LikeLike
  
  Reply ↓
- Erich Styger on August 11, 2018 at 09:50 said:
  
  I used 5V
  
  LikeLiked by 1 person
  
  Reply ↓
Javier on September 2, 2018 at 14:07 said:

Works Erich. Works!!!!! XD

I’m having a problem with serial communication in linux, but I managed to connect once to 115200 /dev/ttyACM1 and I could operate the vacuum pump, the lights, the selenoids and move the Z and the X. I could not prove the Y because I closed the application and then I could not connect anymore.

I do not know yet what can be if on the side of linux or on the side of the smoothieboard.

I also made a prototype for the dispenser, it is functional and simple, I would like to try it when I clarify with this.

Best regards

LikeLike

Reply ↓
- Javier on September 2, 2018 at 15:48 said:
  
  Solved, I have to change the baud rate every time I start my pc.
  
  I have done this but OpenPnP does not recognize it
  
  http://smoothieware.org/linux-drivers
  
  But if the rules are created.
  
  ls /dev/smoothie*
  /dev/smoothie0 /dev/smoothie1
  
  I have tested Y and C but they do not work, everything else works fine.
  
  LikeLike
  
  Reply ↓
  - Javier on September 2, 2018 at 16:20 said:
    
    Sorry, C1 and C2 works fine.
    
    LikeLike
    
    Reply ↓
    - Erich Styger on September 2, 2018 at 17:56 said:
      
      great!
      
      LikeLiked by 1 person
  - Erich Styger on September 2, 2018 at 17:57 said:
    
    I did not had any issues with this on Windows, I have not used the Smoothie on Linux yet. But I now that I had issues with serial ports under Linux for other boards too.
    
    LikeLiked by 1 person
    
    Reply ↓
    - Javier on September 2, 2018 at 20:18 said:
      
      Don’t worry, now it’s working well.
      
      LikeLike
  - Javier on September 2, 2018 at 22:17 said:
    
    The solution is to add this in machine.xml
    
    /dev/smoothie1
    
    So if the port name change to /dev/ttyACM1…9] always forward to /dev/smoothie1.
    
    LikeLike
    
    Reply ↓
Javier on September 2, 2018 at 20:27 said:

Can you share your DM542 configuration?

I have changed from common anode to common cathode but not works to me.

I have changed it back to common anode. I read that 3V3 is not enough.

I have +5V connect to PUL+ DIR+ ENA+

EN5 –>> ENA-
DIR5 –>> DIR-
ST5 –>> PUL-

Nema 17 Motors:

Phase A
Red A+ –>> A+
Gray A- –>> A-

Phase B
Yello B+ –>> B+
Green B- –>> B-

LikeLike

Reply ↓
- Javier on September 2, 2018 at 20:29 said:
  
  Green led it’s ON
  Red led it’s OFF
  
  LikeLike
  
  Reply ↓
  - Javier on September 2, 2018 at 20:46 said:
    
    I guess the error it’s in the smoothieboard config file
    
    # Using Y (beta) with external stepper driver in open drain mode (‘o’ added to the pin, see http://smoothieware.org/general-appendixes)
    # NOTE: using external stepper motor driver!
    beta_step_pin 2.1o # Pin for beta stepper step signal
    beta_dir_pin 0.11o # Pin for beta stepper direction, add ‘!’ to reverse direction
    beta_en_pin 0.10o! # Pin for beta enable
    beta_current 2.0 # Y stepper motor current
    beta_max_rate 5000.0 # Maxmimum rate in mm/min
    
    I’m using M5 (or epsilon) and that port (2.1) it’s of M2, I think.
    
    M5 it’s used by extruder:
    
    ## Extruder module configuration
    # See http://smoothieware.org/extruder
    extruder.hotend.enable true # Whether to activate the extruder module at all. All configuration is ignored if false
    extruder.hotend.steps_per_mm 18 # Steps per mm for extruder stepper ==> 1mm == 1 degree: (200*32)/360==> 17.7777
    extruder.hotend.default_feed_rate 600 # Default rate ( mm/minute ) for moves where only the extruder moves
    extruder.hotend.acceleration 500 # Acceleration for the stepper motor mm/sec²
    extruder.hotend.max_speed 100 # Maximum speed in mm/s
    
    extruder.hotend.step_pin 2.3 # Pin for extruder step signal
    extruder.hotend.dir_pin 0.22! # Pin for extruder dir signal ( add ‘!’ to reverse direction )
    extruder.hotend.en_pin 0.21 # Pin for extruder enable signal
    
    # Extruder offset
    #extruder.hotend.x_offset 0 # X offset from origin in mm
    #extruder.hotend.y_offset 0 # Y offset from origin in mm
    #extruder.hotend.z_offset 0 # Z offset from origin in mm
    
    # Firmware retract settings when using G10/G11, these are the defaults if not defined, must be defined for each extruder if not using the defaults
    #extruder.hotend.retract_length 3 # Retract length in mm
    #extruder.hotend.retract_feedrate 45 # Retract feedrate in mm/sec
    #extruder.hotend.retract_recover_length 0 # Additional length for recover
    #extruder.hotend.retract_recover_feedrate 8 # Recover feedrate in mm/sec (should be less than retract feedrate)
    #extruder.hotend.retract_zlift_length 0 # Z-lift on retract in mm, 0 disables
    #extruder.hotend.retract_zlift_feedrate 6000 # Z-lift feedrate in mm/min (Note mm/min NOT mm/sec)
    
    delta_current 0.7 # First extruder stepper motor current
    
    # Second extruder module configuration
    extruder.hotend2.enable true # Whether to activate the extruder module at all. All configuration is ignored if false
    extruder.hotend2.steps_per_mm 18 # Steps per mm for extruder stepper
    extruder.hotend2.default_feed_rate 600 # Default rate ( mm/minute ) for moves where only the extruder moves
    extruder.hotend2.acceleration 500 # Acceleration for the stepper motor, as of 0.6, arbitrary ratio
    extruder.hotend2.max_speed 100 # mm/s
    
    extruder.hotend2.step_pin 2.8 # Pin for extruder step signal
    extruder.hotend2.dir_pin 2.13! # Pin for extruder dir signal ( add ‘!’ to reverse direction )
    extruder.hotend2.en_pin 4.29 # Pin for extruder enable signal
    
    extruder.hotend2.x_offset 0 # x offset from origin in mm
    extruder.hotend2.y_offset 0.0 # y offset from origin in mm
    extruder.hotend2.z_offset 0 # z offset from origin in mm
    
    epsilon_current 0.7 # Second extruder stepper motor current
    
    2.8, 2.13 and 4.29 they are the ones that I need to use, I think.
    
    LikeLike
    
    Reply ↓
    - Javier on September 2, 2018 at 21:15 said:
      
      Ok, sorry, I’m rolling it.
      
      I have to change the wires, not the ports, extruder is C1 and C2 (M4 and M5)
      
      I have to connect the DM542 to M2, I think that’s the error, I’ll try.
      
      LikeLiked by 1 person
    - Javier on September 2, 2018 at 21:50 said:
      
      Now if the motors move, but it does not move well on the beta axis. As if they were uncoordinated, it throws more than one side of the other.
      
      LikeLike
Javier on September 2, 2018 at 22:36 said:

Solved!

I crossed the phases of the Y engines, and it is already working well. Thank you very much, when I can share a video.

LikeLike

Reply ↓
Javier on September 7, 2018 at 13:00 said:

Hello,

This is the dispenser prototype that I have done with components that I have found it in internet. I have tested and works. With the Nema17 48mm, Nema 17 2mm not works.

Macpick dispenser

LikeLike

Reply ↓
- Erich Styger on September 7, 2018 at 13:05 said:
  
  Do you always get the same amount of solder paste for a given number of stepper motor steps? What I have found is that depending on the compressability and fluidness of the solder paste I get different results, and it depends how much solder is present in the tank. That’s why my thining is that an auger pump would get better results?
  
  LikeLike
  
  Reply ↓
  - Javier on September 7, 2018 at 13:07 said:
    
    Not, sorry. I have not tested yet with solder paste, I think I will do it next week.
    
    LikeLike
    
    Reply ↓
    - Javier on September 7, 2018 at 13:14 said:
      
      You use T3?
      
      I have that use T3 and T4 for BGA.
      
      For the moment I’m going to try with XG-Z40, I think that it’s T3. 25-45um 183º
      
      LikeLike
  - Javier on September 19, 2018 at 18:42 said:
    
    Hello,
    
    I think that this system is more precise.
    
    LikeLike
    
    Reply ↓
    - Javier on September 19, 2018 at 18:48 said:
      
      This is the part.
      https://www.dosieren.de/en/products/consumables/disposable-feed-screw/dfs16-a/disposable-feed-screw
      
      LikeLike
    - Erich Styger on September 19, 2018 at 19:53 said:
      
      Yes, this is exactly the Auger pump I’m using.
      
      LikeLike
- Javier on September 7, 2018 at 13:06 said:
  
  *22mm
  
  LikeLike
  
  Reply ↓
brynn rogers on October 24, 2018 at 23:06 said:

Hi Erich,
I am making progress on my PnP machine. https://youtu.be/U4nb6RTz-xY
There is one design feature of yours I am stealing 🙂 It is the base and magnetic mounting system. Are the 3d files for your various magnetic holders available? How well are they working for you with the magnetic paint?
Another question I have is why dump the laser cut head components and move to aluminum? You never mentioned exactly why that was.

My machine is similar to a MyData Machine, X is on a fixed gantry, and Y moves an entire tray with boards and fixed strip feeders attached. My X and Y are linear motor slides, they will accelerate at 5G and run 3000mm/s.
You used wood with magnetic paint for your base. Mine, Ideally it will be light and rigid.
I’m thinking a sheet of magnetic stainless steel, with ribs and features on the bottom to add rigidity.

I notice you have a photo of a solder paste stencil with your magnetic holders on it. Is that made from magnetic stainless? A stencil alone would probably be to floppy for me, but maybe sandwiched to some other flat rigid and light material will work for me.

LikeLike

Reply ↓
- Erich Styger on October 25, 2018 at 21:01 said:
  
  Hi Brynn,
  your machine looks really great! Yes, the 3D files are available on https://github.com/ErichStyger/McuOpenPnP_Machine/tree/master/3D.
  They work well with the magnetic paint (You have to paint it in multiple layers (4-5)), and because the surface is rather rough and sensitive to moisture because of the iron parts, I added another normal color layer on the top.
  The plywood head was cheap and easy to build, but not as rigid as it should be. Plus there was the concern that water/moisture in the air could affect precision.
  A sheet of stainless steel is something others use too: but that would have been expensive for me, it was not really available and I did not had the tools to cut it in shape with the holes for the cameras/etc. So wood was much easier for me. The solder stencil is magnetic (I think, but I don’t have one with me to test right now).
  
  LikeLike
  
  Reply ↓
  - Brynn on October 26, 2018 at 02:13 said:
    
    Hi Erich,
    the McuOpenPnP_Machine/3D/PCBHolder.stl that I see looks like the long strip feeder with cover. I was looking for the pieces you have that are holding the PCB edges up. I’m probably going to have to redesign them anyway, because I want more height – My Z travel currently will be 75mm max. that gives me room
    do 18mm parts on top and bottom of the board, and held on the nozzle.
    
    Z=0 max height above board
    Z=18 bottom of tallest part I hope to handle when picked
    Z=19 1mm extra clearance
    Z=37 top of board – also top of fixed magnetic strip feeders and tray feeders
    Z=39 bottom of board (rounded off)
    Z=57 magnetic tray
    Z=70 allowance for tray and stiffener thickness
    Z=72 position of top of components in autofeeders
    
    I think that will give me enough room to work with. (I’m just thinking out loud…)
    
    LikeLike
    
    Reply ↓
    - Erich Styger on October 26, 2018 at 07:24 said:
      
      Hi Brynn,
      I’m sorry, I have fixed the .stl file on GitHub (there was a mixup from the .123dx to the .stl). Anyway, the holders are very simply to build or construct.
      I wanted to keep the height of all components (boards, feeders, components, nozzle holder) at the same z-height as my camera has a fixed focus: having parts on different height requires a camera with variable focus).
      
      LikeLike
    - Javier on October 26, 2018 at 10:07 said:
      
      Hello,
      
      I was looking for it too. In Thingiverse there are some alternatives.
      
      https://www.thingiverse.com/thing:1763485
      
      https://www.thingiverse.com/thing:2969116
      
      LikeLike
    - Erich Styger on October 26, 2018 at 17:33 said:
      
      Just one thing to mention: make sure the magnets are not too close to the PCB: my initial version had issues that the magnets were pulling parts from the PCB because they were magnetic.
      
      LikeLike
    - Javier on October 26, 2018 at 10:20 said:
      
      I made some with a greenhouse pin, an 8mm threaded sleeve and a 8x5mm magnet but they need a little more perimeter because if I move the pcb falls easily.
      
      https://drive.google.com/file/d/1QhFxusFV7Cl2qHAm-gsVuSIwevv3LT2x/view?usp=sharing
      
      LikeLike
Brynn Rogers on October 30, 2018 at 07:44 said:

hi Erich,
I have a couple commercial solder stencils in aluminium frames. I checked with a magnet, and they are apparently non-magnetic stainless steel. though the frames are sturdy, I think with some ribs and a sheet of magnetic steel they could by my Y tray.
Are you using you pnp a lot?
Brynn

LikeLike

Reply ↓
- Erich Styger on October 30, 2018 at 16:57 said:
  
  The machine is in use, but was stalled for a week because a lasercut part was not stable enough. Too busy with things this week, but should run again next week.
  
  LikeLike
  
  Reply ↓
Javier on October 30, 2018 at 15:31 said:

Hello,

Thanks for the pcb holder updates, finally I has designed someones carriers and pcb holder.

To EIA-481-D by the moment I has done to 12, 16 and 24mm

https://github.com/hexagon-robotics/MacCarriers

And the pcb holder

https://github.com/hexagon-robotics/MacPick-Accesories

https://drive.google.com/file/d/1z2qrolEdu99UX6ogDROkaScNr0RugyOY/view?usp=sharing

LikeLike

Reply ↓
Javier on December 31, 2018 at 00:11 said:

Hello,

Where is the file of the end stop holder?

Best regards

LikeLike

Reply ↓
- Erich Styger on December 31, 2018 at 07:34 said:
  
  They are in https://github.com/ErichStyger/McuOpenPnP_Machine/tree/master/3D.
  They are in Endstop.123d. I see that the .stl did not had all parts in it, so I generated the .stl again with all parts.
  
  LikeLiked by 1 person
  
  Reply ↓
  - Javier on January 1, 2019 at 16:09 said:
    
    Thank you.
    
    Happy year.
    
    LikeLike
    
    Reply ↓
lab on January 3, 2019 at 00:50 said:

I ordered your bom today except some items that i had already.
Impressive project.

LikeLike

Reply ↓
- Erich Styger on January 3, 2019 at 07:42 said:
  
  Great, I wish you lot of fun and success with that build!
  
  LikeLike
  
  Reply ↓
lab on January 3, 2019 at 10:51 said:

I have done similar projects with laser machines
and from my experience nema17 are prone to loose steps at higher speeds & accelerations
also gt2 6mm at high accelerations will have backlash & oscillation (is there room for 15-20mm gt2 ?)
There is a dc brushed servo open source project for small dc servos (tarocco)
and another for brushless dc servos (odrive).
Without too much cost increase you can replace steppers with dc servos.

LikeLike

Reply ↓
- Erich Styger on January 3, 2019 at 13:56 said:
  
  Yes, loosing steps can be an issue. But I have not experienced this as I did not want a fast machine: I wanted an accurate machine. And driving the NEMA17 with the correct and enough current did not loose me any steps.
  Of course if I would build a machine with my own controller, I would go for DC motors with encoders. But because I built the machine around the Smoothieboard, I was going to use stepper motors.
  Other machine designs are using stepper motors in combination with encoders to make sure they do not loose steps, but these machine are running at higher speed than the one I have built.
  Servo motors usually are not able to generate enough torque imho. Again, it depends on what controller you are going to use.
  
  LikeLike
  
  Reply ↓
lab on January 3, 2019 at 10:53 said:

Also how much and what length/type m3 & m5 screws needed ?

LikeLike

Reply ↓
- Erich Styger on January 3, 2019 at 13:53 said:
  
  I did not count them, as I had them available in different length and sizes. I belive most had 15 mm length. But I’m away from the machine right now, so I cannot tell for sure.
  
  LikeLike
  
  Reply ↓
lab on January 3, 2019 at 15:25 said:

For example odrive take step dir as input & quadrature incremental encoders
also in ebay you can find bldc nema17(+encoder) so you can replace steppers
without any change to controller.
Controller will output step-dir odrive will get it as input to drive bldc.
odrive will handle the servo loop not controller.
If i order 50/50/50 15mm/20mm/25mm M3/M5 hex screws will be ok ?

LikeLike

Reply ↓
- Erich Styger on January 3, 2019 at 16:26 said:
  
  I would go for 100 each: you can easily use them for other projects too 🙂
  
  LikeLike
  
  Reply ↓
- Javier on January 15, 2019 at 23:56 said:
  
  Hello,
  
  This version is better, because you can place them without having to disassemble the profile.
  
  https://www.amazon.es/T-Slot-Europeo-Standard-Perfiles-Aluminio/dp/B07CN25H49/ref=asc_df_B07CN25H49/?tag=googshopes-21&linkCode=df0&hvadid=298037784885&hvpos=1o1&hvnetw=g&hvrand=5945049428666243423&hvpone=&hvptwo=&hvqmt=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9049235&hvtargid=pla-562811639913&psc=1
  
  LikeLike
  
  Reply ↓
  - Erich Styger on January 16, 2019 at 07:48 said:
    
    Yes, I wish I would had these. Disassembling the profiles to add another nut is really a pain.
    
    LikeLiked by 1 person
    
    Reply ↓
    - Javier on January 16, 2019 at 17:35 said:
      
      Yes, I haven’t, but for the next time I’ll use these.
      
      LikeLike
lab on January 15, 2019 at 16:29 said:

I have 2x850mm & 2x550mm mgn12 linear rails
So i have the option to use the long rails (850mm) on x or y axis
y uses 2 rails and x 1 rail.
On which axis to use the long rails ane why?

LikeLike

Reply ↓
- Erich Styger on January 15, 2019 at 16:58 said:
  
  I would use it to make my machine wider (x axis) to make the area for the feeders as wide as possible.
  
  LikeLike
  
  Reply ↓
lab on January 16, 2019 at 04:20 said:

ok i will use one 850mm rail for x axis & two rails 550mm for y axis.
What do you think size bigger than 500×500 really needed ?

LikeLike

Reply ↓
- Erich Styger on January 16, 2019 at 07:47 said:
  
  A bigger size is not *really* needed. I mean: 500×500 is fine. A bigger size will give you more space for feeders and parts. The disadvantages of a bigger is size are that you need more space for the machine, and if you are using longer belts, the backslash compensation will be bigger (potenitally less accurate, so you need to compensate more. The mass of the moving parts might be bigger too (depending on which axis you make longer). I think 500×500 is somewhat of a sweet spot: not too big, not too small, packages <=500mm costs less to be shipped, rails are cheaper. It is all about compromises.
  
  LikeLike
  
  Reply ↓
lab on January 17, 2019 at 16:24 said:

If i use 850mm rails i will use 10-15mm wide gt2 belt/pulleys.
10mm-15mm gt2 fits inside 2040 profile?

LikeLike

Reply ↓
- Erich Styger on January 17, 2019 at 16:55 said:
  
  yes, that should work. I quickly measured it with the ones I have and there is up to 16 mm space (at the widest position if the shape).
  
  LikeLike
  
  Reply ↓
lab on January 18, 2019 at 22:12 said:

There is room for small improvements
look these videos
https://www.youtube.com/watch?v=6agTZWNQBGc (you can not use gear but timing belt in the opposite direction something like servobelt https://www.youtube.com/watch?v=OdJoVh6DRPA )

https://www.youtube.com/watch?v=bFcTj3hsVrI (simple feeders in both sides left & right with 2 actuators that bring next part close to machine )

odrive dc brushless servos for exceptional speeds ( https://www.youtube.com/watch?v=FUh36RUHzdU )

LikeLike

Reply ↓
- Erich Styger on January 19, 2019 at 09:38 said:
  
  Yes, there is definitely room for improvements. The videos of that machine require a CNC milling machine which I do not have. Definitely using more milled parts would be better.
  
  LikeLike
  
  Reply ↓
lab on January 19, 2019 at 16:50 said:

For reducing cost have you tried V-slot aluminum profiles with wheels for V-slot ?
Have this system enough stability like mgn12 linear rails?
Plus is reduced cost / more simple / less weight/inertia
Minus is less stability-accuracy

LikeLike

Reply ↓
- Erich Styger on January 19, 2019 at 17:35 said:
  
  No, I have not used V-Slots. But I saw that others have built machines based on V-Slots, so it is definitely possible. I was aiming at the linear rails because expecting higher accuracy. I did not consider higher mass as a problem, as I was not aiming at speed but more at accuracy.
  
  LikeLike
  
  Reply ↓
Snigg on February 4, 2019 at 10:31 said:

Hi, I have seen the slides of a presentation you had and noticed that you are aware of the “lite place”. While I love to build thing, one must acknoledge that life time is limited. So I wonder what the key advantages of this solution over the lite placer are? Is your machine faster, more precise or is the main benefit the price.
I absolutely like this project and I want to congratulate you to a job well done. With all the public documentation you clearly gone above and beyond.

LikeLike

Reply ↓
- Erich Styger on February 4, 2019 at 10:54 said:
  
  Thank you for your kind words. I have not used the LitePlacer, so I cannot really comment on speed or accuracy.
  If you are looking for a mostly complete kit, the LitePlacer is definitely the way to go.
  I would say that both machines have similar capabilities, but the LitePlacer has only one nozzle. To me a dual head has advantages with more speed.
  But if your main concern is about limited time, then I recommend you buy a commercial machine instead?
  
  LikeLike
  
  Reply ↓
Hoanghoang on February 22, 2019 at 01:30 said:

Please give me a mica laser file

LikeLike

Reply ↓
- Erich Styger on February 22, 2019 at 07:55 said:
  
  Please read this article and see the links to the design files.
  
  LikeLike
  
  Reply ↓
Артем Станчак on March 24, 2019 at 17:01 said:

Hi Erich Styger. I am very impressed with your skill. Having reviewed your article, I decided to make such a machine for myself. I wanted to ask you if the vector files for cutting plywood are relevant? I plan to make all of the plywood as you and looking at the files, I noticed that in some details there are uneven contours. Namely in the OpenPnP_Head file. Maybe if you have the latest drawings, please send me. Thanks! Good luck to you in such projects.

LikeLike

Reply ↓
- Erich Styger on March 24, 2019 at 17:21 said:
  
  Thank you! The plywood files are the latest on GitHub. But be warned: using plywood is good for a prototype to check if things are working to together.
  But having the final machine with plywood is not a good thing, unless you are not aiming for a precision machine.
  You only might be able to place 0805 passive parts that way, but not 0603, because the wood is expanding and changing depending on temperature and humidity. And it is not very rigid too.
  Good luck with your project!
  
  LikeLike
  
  Reply ↓
  - Артем Станчак on March 24, 2019 at 19:44 said:
    
    I will change it over time, but first I want to run on plywood details. But I noticed outlines incomprehensible to me there. There is a screenshot in the link.
    https://drive.google.com/open?id=1jr7SAviyMTgHNGC4mFHHL6GAp-TMpF4vdHUKLlTDvV4
    
    LikeLike
    
    Reply ↓
    - Erich Styger on March 24, 2019 at 19:54 said:
      
      I cannot access that screenshot. The files are .svg files you can open and use with Inkscape.
      
      LikeLiked by 1 person
    - tyomanik7 on March 24, 2019 at 20:13 said:
      
      the error came out, here is the link https://drive.google.com/open?id=1X0wTLXFTetmHOVXgEd7i38hFeCZYqQrW
      
      LikeLike
    - Erich Styger on March 24, 2019 at 20:25 said:
      
      You have to open the files in Inkscape: Only the blue and red color lines are the ones to cut. Other lines in other colors have to be ignored in the Laser cutter.
      You have to adapt these in Inkscape. Some lines might not be 100% perpendicular, but this is not an issue (or you can fix it in Inkscape).
      
      LikeLike
tyomanik7 on March 24, 2019 at 20:11 said:

You have mail or Skype, but I can not send screenshots

LikeLike

Reply ↓
tyomanik7 on March 28, 2019 at 00:21 said:

Thank you for your responses. Tell me please. What is it and for what. OpenPnP_PickHeadEnclosure.svg

LikeLike

Reply ↓
- Erich Styger on March 28, 2019 at 08:13 said:
  
  it is an enclosure for the head.
  
  LikeLike
  
  Reply ↓
Michael on April 21, 2019 at 03:05 said:

Cool project and perfectly implemented 🙂
You write you wanted to stay <$1000. What would you estimate you've spend?

Michael

LikeLike

Reply ↓
- Erich Styger on April 21, 2019 at 09:19 said:
  
  thank you 🙂 I have a BOM on GitHub: https://github.com/ErichStyger/McuOpenPnP_Machine/tree/master/Hardware
  Without counting for some smaller parts, the costs are $974
  
  LikeLike
  
  Reply ↓
  - Michael on May 1, 2019 at 02:18 said:
    
    Thanks Erich.
    
    After a look at the BOM I noticed you used the N40 nozzles, compared to the BOM referenced by https://github.com/openpnp/openpnp/wiki/Build-FAQ.
    
    – Could you please let me know what your experience is using the N40s and what led to this choice?
    – Do you still use the wooden head or did you switch to alu? There is a nice looking head available at robotdigg, but $270 is increasing the build cost quite a lot.
    
    Thanks Michael
    
    LikeLike
    
    Reply ↓
    - Michael on May 1, 2019 at 02:22 said:
      
      I just saw further down your new alu head. 🙂
      
      LikeLike
    - Erich Styger on May 1, 2019 at 09:27 said:
      
      Yes, because the plywood head was not stable enough, I constructed the ALU head 🙂
      
      LikeLike
    - Erich Styger on May 1, 2019 at 09:29 said:
      
      Some use the Yuki nozzles, I used the N40 ones because this was what I easily was able to get from RoboDigg. It was easier for me to use the N40 nozzeles and they are usually less expensive.
      About the head: probably for a new machine I would consider the RobotDigg head. Yes, it is expensive, but building my own head took a lot of time.
      
      LikeLike
Kane49 on April 23, 2019 at 20:23 said:

Hey Erich!

Thanks a ton for this, its the only comprehensively documented build of an openpnp i could find.

Your belted head assembly is different to the ones most people use with cam and springs, i like this one alot better tho.

LikeLike

Reply ↓
- Erich Styger on April 23, 2019 at 20:25 said:
  
  Thanks, appreciated :-).
  The machine is currently under rework and a student is working on adding a solder paste dispenser head.
  
  LikeLike
  
  Reply ↓