Low-power laser needed for polymer bonding

I am building an X/Y CNC machine primarily for drawing acid resistant patterns onto copperclad boards.

Initially I planned on using a simple etch-resist pen for this but that makes the machine more complicated to build (extra actuators etc). So I figured I'd try to use a laser instead to melt printer toner onto the surface of the copper.

Now I don't really know much about laser technology so I'm looking for advice really.

What kind of laser should I use for such an application? Here are my requirements...

1. Visible light (for safety more than aesthetics)
2. Intensity must be variable - either by altering input current or by pulsing the laser via PWM or something like that.
3. Laser will be operating at a very short distance from the target object - between 10mm and 50mm.
4. Needs to be able to melt printer toner (melting point approx 180 to 220 celcius) but not burn it. The toner is required to melt and bond with the copper surface.
5. The laser's beam must be circular and as narrow as possible - ideally less than 0.3mm, although if this is not possible then a lens could be used to create a point of focus. What kind of lens (material, type) should be used?
6. Power consumption is effectively unrestricted (running off a wall-wart, via a voltage regulator (LM317).
7. Should be able to melt (but not incinerate) 1 microlitre (1x1x1mm cube) of loose black toner powder in 0.2 seconds (by my rather sketchy calculations).

I've got all the hardware built already apart from this idea for the laser attachment.

I'd like to keep the cost down on this laser as much as possible, this is just a toy project I'm doing in my garage .

So I'd like to hear what you guys would recommend

I suggest a 405nm laser, they are the most powerful diodes with a manageable beam pattern so they will be easy to focus to that small spot you need.

As far as melting toner, you'll probably need to get a laser that's too powerful and dial it back till you get the results you want. You can control output power by limiting input current.

A good 405 diode can be had for ~$50. I suggest a cheap glass lens for focusing, that way you can just swap lenses if you accidentally burn something and trash the lens. They should be pretty cheap >$10.

Good luck!

Could you estimate the wattage I'd need for this kind of thing?

I would say 700mW's would be fine but that might still be too strong. Im not too familiar with etching.

I see some extra possible problems ..... sorry, for mention them, but is probably better that you consider also them, before to start to build .

First, you need a way for distribute an uniform layer of toner powder on the copper clad, and this is not exactly easy (and, speaking for direct experience, toner powder is a damn bad bitch going everywhere and dirting anything) ..... then, assuming you can do that, consider that copper is a very good heat conductor ..... once the toner start to melt and stick on the copper clad, the copper draw heat quickly and dissipate it, or spread it around ..... with a low speed unit, like one that can be built at hobbyst level, this can cause two different problems.

First case, the heat is enough for continue to melt the toner, so the extra heat spread around and melt also the surrounding toner not in the laser track ..... second case, the heat is too low, then the melted toner don't melt uniformely, nor stick in the right way on the copper clad.

A different thing is using CO2 lasers at high speeds, but this is not exactly an easy realization (i mean, you cannot probably use a standard "etching machine" setup, with the moving carriage, cause the high speed of the carriage will cause air flows that spread toner powder everywhere ..... this means using a high speed "galvos head" setup, and these ones are usually high cost units, for CO2 systems).

A simple way to bond the toner to the copper would be using a powder coating setup, it would give the copper plate a positive charge and the powder a negative charge. It won't come off at that point unless you rub it off. Toner is about the same consistency as powder coating powders, plus it's designed to work a similar way in the printer anyway so I'm sure it would work. You can get a self contained powder coating gun dirt cheap, I have this gun here and it would work perfect for what you want to do.

Amazon.com: Craftsman Complete Portable Powder Coating System: Home Improvement

I will agree with HIMNL9, you'll need to go as fast as possible to keep the heat from bleeding to surrounding powder. This is why I suggest the strongest laser you can get so you can experiment and tune the laser to your machine. If you are running with simple allthread leadscrews and steppers you won't be able to go as fast as servos and ballscrews, for example.

I can't really estimate how much power you will need, but if you can find the specific heat capacity of toner you could calculate a very very rough estimate.

I would suggest going around this problem the other way around:

Us photosensitive PCB material, and light everything you want to remove with a 405 laser. Afaik this wavelength is short enough to break the bonds in the photosensitive layer - some people use masks lit with NUV leds.

This is if you insist on using the CNC to apply the etch resistant material, it would probably be easier to prepare the boards for etching using a either a transparency mask on photoboard, or toner transfer on plain copper board (from special paper onto the copper using a clothing iron or such).

Have you already built the CNC set up? I'd be interested to see some pics.

As to getting toner on pcb's, it might be easier to just modify a printer to accept a board. Eg. cut a slot in the back lol.

There is another method that still uses a Laser and Toner...
Mirror image print your Artwork on special heat transfer material
on a Laser printer.

Then you heat transfer the image onto the copper clad PCB using
a clothes iron and etch in the same manner.

1) it is quick
2) you don't need a CNC machine
3 It uses a standard Laser Printer
3) it is quite inexpensive


Jerry

While the toner transfer method works pretty good its had to get fine traces. I'm sure the op knows the toner transfer method, he is just looking for something better.

Hi guys, thanks for the great feedback. Much to think about for sure.

@Benmwv: Here's some pics of the plotter... Plotter Project - a set on Flickr

I've done a lot more work on it since these shots were taken, and I'm currently working on designing the electronics that will go between my PC and my plotter to handle the "gcode" math.

It's design is aimed at a compromise between speed and resolution. It's stepper/belt drive for both axes which means it's fast but not hugely accurate - but it does beat manually drawing traces with an etch resist pen as I was doing before (which I quite enjoyed actually).

There are some other reasons hy I'm using a CNC machine approach rather than traditional "clothes iron" methods for toner transfer...

• I can use the same machine with a different attachment to drill the holes for PTH components.
• I can use the same machine to bond toner to other materials (such as white plastic) in order to print graphics onto solid plastic -think project boxes
• I can use the same machine to draw schematics onto paper (with a pen) if I want, or perhaps to do a test run without risking a piece of copper.
• I could modify the machine to cut paper or thin card.

So this machine could be quite versatile if I manage to achieve any of this stuff. Plus of course it'll be fully documented on my website when it's all finished

I'd love to hear any other suggestions you guys might have regarding this build, whether or not it's laser related.

As for the laser question - I'll have a look into 405nm diodes, no doubt I'll be back to ask about how one sets up such a device safely and without blowing it up as soon as I give it some electrons.



Thanks very much for the help everyone, much appreciated. :bowdown:

Pontiacg5 said:

While the toner transfer method works pretty good its had to get fine traces. I'm sure the op knows the toner transfer method, he is just looking for something better.

Click to expand...

With his last explanation I now see what the OP is looking to
achieve...

I agree that fine traces are difficult to get... but the toner transfer
method is still more accurate than....

manually drawing traces with an etch resist pen as I was doing before

Click to expand...

I just have my PCBs made....

@ wosser....
Keep us informed of your progress with your interesting project....


Jerry

Just as idea, as possible alternative to toner transfer paper .....

10pcs Dry Film Photoresist Sheets for DIY PCB 6x8" q | eBay

Photoresist Film Sheets Kit for Photo Etched PE PCB | eBay

Somewhere is also possible to find the "paint" form of the photoresist, but not in all countries, cause one of the components (at least of the old yellow/brown types) is extremely dangerous for nervous system, so some countries banned it from civil market ..... the blue type, apparently, is not so dangerous, if they still sell it worldwide.

The main complaint that I have with normal toner transfer methods is consistency. It appears that it would be difficult to, for instance, make five boards to the same level of quality/reliability. Also there's inherent difficulty if you're trying to do double sided copper clad boards - registering the back in relation to the front is hit-and-miss due to the amount of manual handling that is required.

I'm not really looking for something "better" than toner transfer, but rather something that is more reliable at the expense of reduced resolution. If it can be more reliable than hand-drawing the traces and the quality is consistent (as it should be with a CNC setup) then for me that's a big win, even if the resolution is as bad as my hand-drawn efforts.




Does anyone know a reputable place to get a 405nm 700mw (as suggested earlier in this thread) laser module from? I'd rather avoid ebay for a purchase like this. Farnell doesn't seem to carry anything below 450nm.

Also should I go for a discrete laser diode on it's own or should I get a complete pointer module?

Don't know about the diode, but i have a suggestion for you about double-sided PCBs alignment, that work pretty well, usually ..... print the papers larger than the needed, and use centering crossmarks on the corners of the sheets, so you can align them in the right way.

For photoresist systems, just align the crossmarks, stick the sheets corners together, then slide the board between them and center it ..... for the toner transfer system, transfer one side on the board, then flip it without take away the sheet, align the second sheet with the first one, and transfer the second one.

Here is what you need:
BDR-S06J 12x Blu-Ray burner lens(500+mW) [M0037] - $51.99 : Modwerx Shop

That is a sled and you will have to extract the diode yourself and then press it into a module. (find Aixis modules on the same site) As far as I know this is the only way they come. You will also need a heatsink to put the module in and they have a nice one in that site as well.