Pick and Place on a budget - Part 1
This afternoon I did a quick spike test to check out the feasibility of a project I will be starting in a few weeks' time. I'm building a low-cost pick and place device to help me solder Surface Mount Devices (SMDs).
It's becoming essential to use SMDs. The space saving over traditional through-hole components is considerable, and more and more components are only available in SMD format.
Soldering SMDs is fiddly and risky if you're young, and very difficult for someone my age. My eyes are not good enough, and nor is my dexterity. I've soldered the odd SO8 device but a micro controller would be almost impossible for me to manage.
Enter PAP, my low-cost Pick snd Place 'bot for the older developer. Commercial Pick and Place robots cost tens of thousands of pounds; manual Pick and Place workstations are cheaper, but start at a few hundred pounds. I'm hoping that mine will cost me will end up costing me a lot less.
The idea is simple. PAP use a human (me) as the main controller, aided by a USB microscope.
I'll put the PCB on the bed of a compound milling table. The table will set me back about £70, but will normally be used for machining. Above the table will be a mini-drill press with an arduino-controlled geared motor to rotate the part. I hope to pick up the component to be soldered using an improvised vacuum tool - a plastic tube with a large syringe at the far end.
The soldering process will go like this:
It's becoming essential to use SMDs. The space saving over traditional through-hole components is considerable, and more and more components are only available in SMD format.
Soldering SMDs is fiddly and risky if you're young, and very difficult for someone my age. My eyes are not good enough, and nor is my dexterity. I've soldered the odd SO8 device but a micro controller would be almost impossible for me to manage.
Enter PAP, my low-cost Pick snd Place 'bot for the older developer. Commercial Pick and Place robots cost tens of thousands of pounds; manual Pick and Place workstations are cheaper, but start at a few hundred pounds. I'm hoping that mine will cost me will end up costing me a lot less.
The idea is simple. PAP use a human (me) as the main controller, aided by a USB microscope.
I'll put the PCB on the bed of a compound milling table. The table will set me back about £70, but will normally be used for machining. Above the table will be a mini-drill press with an arduino-controlled geared motor to rotate the part. I hope to pick up the component to be soldered using an improvised vacuum tool - a plastic tube with a large syringe at the far end.
The soldering process will go like this:
- Place the PCB on the compound table, and make it secure.
- Use the vacuum pick-up tool to pick up and hold the SMD component.
- Move the PCB under the component by using the compound table
- Use the USB microsope to help me locate and rotate the component
- Use the drill press to place it on the board
- Solder, release the vacuum, and celebrate!
All this will take a while to set up, but today's experiment had a less ambitious goal. I wanted to check that I could use my USB microsope with Linux. I got the microscope for £15 in a special offer from Maplins; it worked well with Windows, but I no longer run Windows at home so I needed to try it out on my workstation, which runs Ubuntu Lucid.
I installed cheese, the standard USB Webcam software for Ubuntu, and the USB microscope worked right away. Here's a photo of a PCB at about 100x magnification. Focus isn't perfect but the detail is amazing.
There's much more to do before I can test PAP for real, but the USB microscope should work really well.
Comments
Post a Comment