New Life For Old Blue

I’ve decided to give a new life to my old CNC machine, Old Blue. There is one piece of equipment that I know would get lots of love and attention that we currently don’t have, and that is a good PCB mill, that can also do solder paste dispensing as well as pick and place. The frame and motors are perfect for that kind of application and with some new electronics, a bit of 3d printed parts, I think we can make a pretty good machine.

I’d like to base it off the open sources machines… but with some cool additions.

Thoughts?

Bill

That‘s a very good idea! We are supposed to receive today at work exactly the machine you have described:

http://cirqoid.com/

The dispenser is fun but I think it’s not very relevant for PCB since you can have very fast stencils for a few dollars (https://www.oshstencils.com/) and you even can make your own from Pop can of Kapton/PET sheet with the laser cutter.

The pick and place would be nice. I am very interested about the accuracy you can reach. We have also ordered (still for work) this machine with vision system:

http://www.liteplacer.com/

We are using 0402 components so the accuracy matter…

Julien

And by the way, the old blue was on my radar for the next items to clean the CNC area… so you took the right decision to recycle it!

Julien

I’m +1 on this.

I’ve been putzing around with getting parts for a camera/vision based pick-and-place machine (probably my spring/summer project), and would love to do something similar at the hive.

Ian

I’m in… Pick n Place is on my radar as well. Julien, you’ll have to let me know what you think about liteplacer. I considered getting one for the house.
-D

Fort Both the ciquoid and lite placer, I will bring this machine at he hive during a Tuesday meeting. If some members will be interested, I will run a quick demo. I am currently discussing with both the cirquoid team and the liteplacer creator to learn how to use their machine in an optimal way, I can absolutely share my knowledge with you.

For pick and place, there is also the FirePick delta project but it seems to be in a less advanced state that the liteplacer.

Julien

liteplacer seemed to be closed source for a lot of it, or I didn’t look hard enough.

There are a few open source PnP machine projects but they’ve not made it to the point of being very useful yet.

-D

I think the priority functionality should be the pick and place components, since it would seem that the Shakeopo be used to mill… and solder paste can be handled by stencils and the laser cutter.

I think we can break this up to some discrete components… being:

  1. New electronics to drive it - Printrboard/Raspberry Pi
  2. Vacuum source - Fish Tank… in reverse?.. MOSFET or airvalve controlled from Pi/PRB
  3. Rotating Head - Servo Motor, laser cut, 3D printed parts, hooked directly to PRB
  4. Camera System - Raspberry Pi / USB (mounts in effector)
  5. Tape Holder Slots/Bins - 3D printed and Laser Cut
  6. Effector Head - Metal Tip Glue Applicator with laser cut and 3d printed parts

What else am I missing?

Bill

Few questions. What do you mean pick and place? I am not sure how this will be different from our current cnc tools, can you explain? i am not familar with alot of this jargon.

Daniel,

Have a look on the link, I have posted for the lite placer. there is a video on the website that shows what iq pick and place.

You also have this :

https://www.google.com/url?sa=t&source=web&rct=j&ei=Si_nVNapJ4q_ggSrj4CwBA&url=http://en.m.wikipedia.org/wiki/SMT_placement_equipment&ved=0CC4QFjAA&usg=AFQjCNGtlw2RDd1XgfHe_FMPDZuy-Od_nA&sig2=3Ukio1aEHFiDdOu11XtYug

Our three CNC can’t handle pick and place right now. you need a vacuum system to lift the components and usually one or two camera to get a sufficient accuracy for tiny modern components.

Julien

Aquarium air pumps used on the suction side are really weak. I’ve done this before for other things with limited success.
You can use shop air to create vacuum, but our air compressor situation sucks. (figuratively)
There are plenty of industrial/scientific vacuum pumps available for couple hundred dollars that would be a much better choice IMHO. The other thing to do would be to use an air tank as a “vacuum tank” so that the pump wouldn’t have to run continuously.
I will look into sourcing the vacuum parts end of things.

The MAC valves that I use for controlling turbocharger wastegate actuators would probably work fine for this. They’re rated to 90-100psi ish positive pressure. I don’t really see any reason why there would be an issue using them with vaccum as it’s just a pressure differential in the other direction. I am 95% sure I have one of these laying around I can measure the electrical characteristics of. If not, they’re cheap, i.e. <$30. I’m going to say 12V @1-2A off the top of my head. I’m not sure how fast they switch on/off but it is fast enough to have a relatively fine control over flow while running the valve at 30hz PWM with 8 bit duty cycle control.

-Dave B.

I agree with Dave aquarium air pumps will be too weak. Some people have tried to use microblower but if you want to lift big component like wifi or Bluetooth module (bluegiga or Nordik) it is apparently not strong enough. see this like:

http://www.tricorderproject.org/blog/vacuum-head/

I was more thinking of using shop air to create vacuum, We can fix our current compressor or buy a tiny one just for the machine. I am more concern about the vision system and the accuracy that you can get. What software do plan to use for the vision? OpenPnP?

Julien

I’m also +1 on using shop air – I’ve mentioned to Jon that I’d love to have (and am willing to work on/invest in) plumbed air in some of the major areas of the shop (Wood area, CNC area, etc).

My biggest issues with OpenPnP is that the math is…not amazing for what we’re trying to do. For example, when I was messing around with openPnP, it had a difficult time identifying the degree of rotation on the items as they were pulled off of the plate or pan. OpenPnP’s solution at the time was just to rotate it clockwise until it could identify that it was in the right spot for placement – and it didn’t matter if it was out of alignment by -5 or +5, so it was spending a lot of time tweaking the part a lot to get to where it needed to be.

I don’t know if the solution is to fork/tweak it so that it works for us, or, if we roll our own – but, I’m open to suggestions/ideas.

Ian

Ian, the issue is the shit compressors and power available to run them. The lack of distribution is the least of the issues.
-D

Easy enough, let’s fix it :slight_smile:

We can run new power for the air compressor – and I think of an air compressor as a core utility of the space – just like power and water.

Let’s work together to find something that is appropriate for the space, our needs, and get it in here.

Screw compressor :slight_smile:
-D

On a more serious note, I did a little digging. The best idea I’ve been able to come up with so far has been to use an electric vacuum pump from a car. They’re plentiful and relatively cheap (~$50 new). They’re designed to pull enough vacuum to operate HVAC stuff and sometimes vacuum-activated hub locks on some trucks. Combined with a small manifold, should be sufficient to pick up parts. Might even be able to find one light enough that it could be mounted on the “head” responsible for moving parts around as the routing of vacuum lines through the CNC machine was something that was noted to be problematic by others.
-D

Although this would be a really nice compressor:
http://www.ebay.com/itm/Brand-New-Eaton-10-HP-Single-Phase-Rotary-Screw-Air-Compressor-/171680922655?pt=LH_DefaultDomain_0&hash=item27f8fafc1f

Total overkill for the hive, but still would be really nice. :slight_smile: 45SCFM @ 150psi continuous. Mmmmm.

More realistically, this 5hp 2 stage compressor would be a very good choice in terms of both storage capacity and flow.

http://www.ebay.com/itm/INGERSOLL-RAND-2340L5-Electric-Air-Compressor-2-Stage-5-HP-230V-1-Phase-28-Amps-/381020648792?pt=LH_DefaultDomain_0&hash=item58b699b558

Only 15SCFM@140psi but that’s still quite a lot. Should last a LONG time. Beefcake of a compression unit. We could also daisy chain old shitty compressor tanks on there to increase storage capacity. Probably sell off some of the working compressors that don’t quite suit the space too… It’s money, but not ridiculous money. I think the only tools that this compressor MIGHT not be able to keep up with at the space would be some of the air grinders, especially if more than one was in use simultaneously.

It might not be a bad idea to look into a decent air dryer. And possibly a drop-down regulator, i.e. store air ait 140-180psi (as compressor allows) and drop to 90psi to feed hard lines to space. With the compressor set to kick on at 110-120psi, a system like this could pretty much guarantee regulated pressure air at all times.

Shop air at the hive has always been a joke. A decent compressor combined with some hard lines and points of access to compressed air would be awesome. While I agree with you Ian that it’s important, I never really felt like many other people shared my opinions so I’ve remained quiet on the issue.

-D

Good shop air is something you never knew you needed so much and will NEVER be able to live without again. I agree with Dave. I'd second a dryer and regulator. "Build once, build right" on something like this, and experiment more with the creative tech that is everyone's specialty. Without dryer and regulator, we would increase potential to damage the pneumatic tools supplied. I wouldn't want to put a brilliant, precision device on an air line and see it's functionality threatened by a lack of dry, constant air.

Regarding vacuum, I might suggest a high quality "diaphragm" pump (usually medical application). I got a Gast a few years back for $40 and it was very well suited to this.

Lorin

Maybe someone mentioned this, but a simple venturi vacuum pump using our compressed air (once that's reliable) is probably one of the easiest and cheapest solutions.

Bit a pick and place machine would be an awesome addition to the hive.