Text Size

Pick and Place Machine

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 6:37 pm

Tom,

The the soldering could be possible. The machine would need to apply the solder paste in some way (another syringe, or solder mask). I prefer the solder mask option. After the parts have been placed, the board can then move to either a conveyor oven, or a hot plate. The heat profile is not difficult to control using a simple uC. I personally like the idea. Additionally, if non-conveyor, the nozzle (using a larger suction) could pick it up and place it on a hot plate.

Picking and placing with coordinates is not a problem with software that I would write myself, but camera feedback to close the loop would be somewhat challenging, but I'm up for the challenge.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 6:42 pm

I agree with airnocker, a vacuum pump is probably the way to go and is not too difficult to implement using a solenoid controlled valve of some sort. I will be looking into the pneumatic systems. The pump really dos not need to be powerful at all.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 6:47 pm

shawjo,

Stepping motor turning the parts is an absolute. It would be great to get webcam feedback along with this.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 6:57 pm

ranger90 wrote:I could see a few interesting challenges with this.

1. Do you use a gantry configuration? OR a robotic arm? A gantry system would be easier, from an engineering and geometry stand point. But the robotic arm would be sooo much cooler.

2. For a vacuum source. I see three sources. A) A plunger (piston). They wear, but have a long throw. B) A diaphram, won't wear as quickly, but getting a long enough through to create enough suction will be a challenge. C) A constant vacuum, from some pump, but the vacuum (not the pump, just the air pipe(s)) is turned on/off as needed.

3. To activate a plunger or diaphram, I see two alternatives. A) a servo, like an RC airplane servo. It would match well with a plunger, but could be too slow to make the machine useful. B) An electromagnet, matches well with diaphram. Applied current would activate the magnet, pulling the diaphram, creating vacuum.

4. rotation of part.. Since you're not working with exceedingly heavy parts here. A small servo (again, RC airplane servo comes to mind) could be used. It's light, small, and controllable.


The robotic arm is very cool and this provokes a thought: Delta robotic kinematics: http://www.youtube.com/watch?v=KmEsUgvuXvY. We would be getting away from the blueChick, but a very interesting solution and possibly more compact.

I have found diaphragms to be limited in suction. However, I have no tried more rigid diaphragms. I speak of the small pen size.

4. Correct. A hobby servo would be acceptable and small. Stepping motors are a bit more inexpensive for their size and can be found for free in street garbage. All options should be investigated.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 7:02 pm

JimM wrote:This guy was doing pick and place with a Taig mill in 2006.

Lack of a reliable vacuum source was also a problem with his design.

http://www.youtube.com/watch?v=bbBr9cbMqq4&feature=mfu_in_order&list=UL

There's no mention of it on his current website http://www.hightechsystemsllc.com


Yeah, I'm familiar with this video. This was done a few years back. He did this for only a single repetitive project. Gotta give kudos for that!

That video brings the thought of the positioning of the part on the nozzle and accurately placing it on the pads without any offset. This is where feedback is crucial. I would love to hear thoughts on this. Since most of the parts on reels/tape and spread on a waffle, would this really be necessary? The parts will (to a certain extent) auto correct during the reflow of solder.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby MachiningBob » Thu Mar 31, 2011 7:21 pm

I think the xbox kinect would be the right way to go for the feedback in the form of a point cloud. Cheapish and the libraries are getting more and more robust.
Book Machine / BobCAD-CAM v24 / EMC2
3 Axis Keling 425 oz-in on 4042D drivers
CNC4PC C35 BoB
User avatar
MachiningBob
 
Posts: 140
Joined: Sat Feb 26, 2011 6:59 pm
Location: McKinney, TX
CNC Machine: "Book" Machine

Re: Pick and Place Machine

Postby PHoodDaniel » Thu Mar 31, 2011 7:25 pm

Great idea. Could give me an excuse to dive into the Kinect. We should also explore OpenCV, Microsoft Robotic studio (webcam libraries) and how the existing machines apply feedback.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Pick and Place Machine

Postby ranger90 » Fri Apr 01, 2011 2:54 pm

I had a few more thoughts regarding a vacuum source.

An aquarium air pump. The small ones use diaphrams, but could probably be rigged to generate vacuum instead.

Another thought. Instead of turning vacuum on/off. Have vacuum on all the time. But hold the part down while the vacuum hose is pulled away. Once the vacuum seal is broken, the part stays where it is placed. You could probably rig up a simple electomagnet to activate a lever. The lever holds the part in place while the vacuum source is pulled away. Once the vacuum is far enough away, the magnet turns off, and the lever (spring loaded.. maybe) retracts, and the vacuum is ready to pick again.

No matter what you do.. timing will be a challenge. You will need to move the vacuum to the part, activate the vacuum, then move/place the part, then disengage the vacuum.

You could go for some kind of analog deactivation of the placing process. As the z-axis moves down, a switch/sensor (think limit switch, or a plate scenario in your zero-ing axis tutorials) would hit something, and in turn stop the vacuum. You'd want to be able to tune the altitude the sensor. Like a threaded bolt to lift/lower the sensor. Once the item is placed and the z-axis moves up, there would need to be a delay to keep the vacuum disengaged long enough to "get clear"
ranger90
 
Posts: 50
Joined: Mon Mar 07, 2011 3:12 am
CNC Machine: BlueChick

Re: Pick and Place Machine

Postby MachiningBob » Fri Apr 01, 2011 3:57 pm

Ranger90, speaking of aquarium pumps do you think the vacuum hose/pipe will need filtration to keep from gumming up the works while the valve is open? I like the idea of retrofitting an aquarium pump if only because they are quiet. I fear that it'll need to be something along the lines of a shop vac attached to a pvc apparatus since there will be very little surface area and sometimes components in tape reels are stuckish. I also like the aquarium pump idea because you could probably have the release valve actuated with very little force. I'm sure there is a right amount of suction that gives the best part pull with the least amount of mechanical overhead...if only there were some way to use numbers to arrive at a solution for a real world problem.

A magic way.
Book Machine / BobCAD-CAM v24 / EMC2
3 Axis Keling 425 oz-in on 4042D drivers
CNC4PC C35 BoB
User avatar
MachiningBob
 
Posts: 140
Joined: Sat Feb 26, 2011 6:59 pm
Location: McKinney, TX
CNC Machine: "Book" Machine

Re: Pick and Place Machine

Postby airnocker » Fri Apr 01, 2011 6:41 pm

50 micron air filters seem to be typical in P&P vacuum systems.

There are enormous problems to solve the least of which is the vacuum side of it.
The biggest hill to climb dealing with the component placement error on the tape reels. This involves:
1. capturing photo of each component
2. comparing photo to standard component shape library to detect malformed components
3. determining if component should be discarded and another selected
4. comparing component's actual XY location and rotation orientation on the tape compared to tape fiduciary markings
5. calculating relative movement of pick-up head's XY and rotational position to counter components placement error on tape and applying the correction errors to placement movement of head

These represent what to me are the greater challenges.

Does anyone have experience with this?
airnocker

Everything depends on everything else
airnocker
 
Posts: 620
Joined: Thu Mar 25, 2010 9:43 pm
Location: Wellington, CO
CNC Machine: Custom

PreviousNext

Return to Research & Development

Who is online

Users browsing this forum: No registered users and 1 guest

cron

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 158 on Thu Mar 08, 2012 8:54 pm

Users browsing this forum: No registered users and 1 guest