Anyone tried this?

[Tod Landis]

Hi Guys,

Well, I've had my laser (Epilog Legend 24TT, if that helps anyone reading) for about one year. I got it to initially to laser mark my newly patented "spot-check" welding gages. (Of course, I've since used it for many different "crafts"...)

Well, along with the gages, I developed a device that I want to upgrade to using PCB boards. That is, I want to create my own PC boards using the laser. This will make the device a bit more difficult to "reverse engineer", and protect the patent further.

So, my question for the group is: Has Anyone ever managed this trick? If so, what settings did you come up with? (I intend to purchase the copper clad boards from Mouser.com, but I'm guessing that the thinner substrate is best.)

I hope that using the Laser in this "practical vs. artistic" manner doesn't reflect too badly on me, but what can you expect from an Engineer?

Thanks in advance for any help that you can send my way!


Tod

[Joe Pelonio]

The closest I've come to electronics on the laser is the ends of enclosures, vector cutting the shape and the holes for USB connections etc. I really know nothing about making PC boards, but I thought they were laminated with a sheet of copper that typically is chemically etched away, leaving the designed circuits only. If that's copper sheet it won't cut with a CO2 laser like yours (I have the same one) You need a yag laser to cut metal.

[Brent Brod]

Seems like you could use the laser to raster off a mask/resist prior to etching. I've never made PC boards, but seems like the process would be very straightforward.

[Lee DeRaud]

Seems like you could use the laser to raster off a mask/resist prior to etching. I've never made PC boards, but seems like the process would be very straightforward.Very straighforward...and very slow, even for very low-volume work. Lots easier to make a reuseable mask in a laser printer on clear material (think "vue-graph slide"), then use that with photo-resist to lay down the real mask for etching.

Might be able to use the laser for vector-cutting the substrate after the copper is gone, but I have no idea how well that would work. I suspect that registration would be an issue, given the size of modern components.

Secondary question: what's the point? Reverse-engineering a 1-2 layer circuit board is trivial no matter how you produce it. If you want to protect the design, you're going to have to put in some ASICs (or at least FPGA) components that aren't off-the-shelf.

[Matt Warfield]

Secondary question: what's the point? Reverse-engineering a 1-2 layer circuit board is trivial no matter how you produce it. If you want to protect the design, you're going to have to put in some ASICs (or at least FPGA) components that aren't off-the-shelf.

I agree. I work for a computer manufacturer and it's fairly routine to x-ray a simple board with up to 11 layers to reverse engineer it. When you start delving into the realm of high density traces then it becomes difficult to trace everything. Best way to slow down those that reverse engineer boards is to place components with false value and component labels.

MW

[Tod Landis]

Being a trival point could be true, but I do want it to be easier to purchase than it would be to "knock-off". Yes, the patent is in place, but having the prototype completed, and proven would same me costs going to an outside firm that makes production runs. (Besides, we will be encasing the components in epoxy as well...)

Basically, I want to prove out the design by making the board that I want before mass producing. I think that the file that I should have attached goes through how to CNC prototype boards. (I just figured that I could utilize the laser to vector off the thin copper layer.)

Of course, I intend to make the attempt in any case, but I figured that I'd ask the group for pointers. I am already making the custom enclosures with the laser.

If it doesn't work on the CO2 laser, does anyone here have a YAG service to offer up?


Tod

[Lee DeRaud]

I agree. I work for a computer manufacturer and it's fairly routine to x-ray a simple board with up to 11 layers to reverse engineer it.Yup. Of course, multi-layer boards aren't exactly prime candidates for "home-brew" or even small shop production.:eek:

[Tod Landis]

Wow, false components on the board. (I wouldn't have thought to go there.) I'm a Manufacturing Engineer by trade, and reducing components was my goal.

Basically, I think that major manufacturer's would for after the patent for the obvious reasons. I just wanted the end users (metal stamper that weld) to not undermine me.

I will definitely consider throwing false components on. If this works, I can spend more time making the fun projects!

Thanks,


Tod
;)

[Lee DeRaud]

Basically, I want to prove out the design by making the board that I want before mass producing.Are you going to have the whole assembled product produced or just bare boards that you will populate with components by hand? I'm under the impression that most production shops prefer to do their own layout from the wirelist so they can optimize component placement for their particular machinery.

[Joe Pelonio]

There may not be many here with a yag since they don't do wood well. I doubt that you could vector cut the copper even with yag, since the beam would hit the softer plastic under it and you'd have a mess. It would have to be rastored to the depth of the copper but stop before hitting the plastic.