Transferring Images to PCBs with Laser Prints: A Successful Test

[camerart]

Hi,
I am trying out ideas for using Laser prints to transfer image to PCBs which is resist and allows etching the circuit.

Here is a small circuit, attached, showing a successful test. This is approximately 20x40mm. You can see the top 2mm aluminium plate has a sheet of silicon, which spreads the load, then yellow transfer paper, then the printed circuit, then the bottom plate.

This sandwich goes into an old toaster for 1min.20Secs. More time anneals the Ali, less time is not enough, to soften the print.

This is about the largest PCB possible when enough pressure can be applied, before bending the Ali.
I want to scale up to take a 4"x6" PCB, so obviously the plates must be stiffer.
so far, my idea is to get 5-6mm and end-mill holes making a honeycomb pattern.
does this sound feasible?
Cheers, Camerart.

 

[Baluncore]

As I understand it, you are printing an image in a laser printer, with has a fuser that bakes the powder toner onto the transfer paper. You then transfer that toner from the paper to the bare copper PCB by reheating while clamped in close contact. The toner is then the etch resist.

I think you will have problems using thick plates in a toaster. One problem is keeping the thermal mass of the clamp plates small while still evenly heating the sandwich. I would avoid the toaster by using a heat press that had solid pre-heated plates to momentarily clamp the sandwich. Another solution could be to use a laminating roller, or the fuser from a laser printer to do the thermal transfer to the copper. That would pinch the sandwich between rollers and not need aluminium plates. One problem with rollers is maintaining registration on double sided PCBs.

If the clamp plates were made from a transparent material, a bright lamp could re-heat the toner. If the material blocked IR then the greenhouse effect would increase the efficiency. The clamp would then not have thermal mass, only the paper and toner would be heated.

 

[camerart]

Hi S,
You understand the process ok.

I have seen others mention they have a registration problem using rollers, this is why I am concentrating on the 'plate sandwich' idea.

I calculated, 2x thicker plates+2x silicon sheet+PCB, and it has started to add up to 'Thicker than toast', so perhaps too thick as one sandwich..

Judging the time it takes my 2mm 20x40mm plates to cool down, I think one idea could be heat each 'say' 100x150mm 'say' 5mm plates in the toaster and assemble a sandwich and press it.

I'm also looking at sandwich presses, to see if they could be modified. (image attached).

C.

 

[jim hardy]

I'm also looking at sandwich presses, to see if they could be modified. (image attached).

How Practical !

Check thrift shops for an old waffle iron with reversible plates - one side waffles flip them for griddle.. Should cost under five bucks.

I knew a guy whose office copier had a straight through paper feed path. He Scotch-taped his boards to a piece of paper and fed them right through.

 

[camerart]

Hi J,
I try to be!

In a 'charity' shop here it would be 5 Quid :)

To clarify, your printing guy had a laser printer that printed circuit boards directly? Sounds good, but perhaps not for double sided, due to registration problems.
C.

 

[camerart]

Hi,
Just did a test. Sandwich: plastic-silicon-Transfer paper-PCB-transfer paper-silicon-plastic, wrapped in rubber bands. Placed in Microwave for 20 mins. The plastic was starting to melt, but the transfer almost worked, so this needs investigating.
I tried the silicon in the microwave on it's own, and it got hot, so this appears to be how the heat could be injected.

Questions: If anyone knows, the dangers of materials like the ones above?
+ The temperatures needed to transfer the image?
+ The max temperatures of known silicon etc?
Please reply.

C.

 

[jim hardy]

To clarify, your printing guy had a laser printer that printed circuit boards directly?

It was the office copier where he worked. Wasn't made for that, just a plain photocopy machine.
He noticed the straight through paper path and said "Hmmmm i wonder if.."
Yes, it printed onto the copper clad board.

 

[camerart]

Hi S,
Ok. thanks.
C

 

[Tom.G]

+ The temperatures needed to transfer the image?

The fuser in a Laser printer operates just below the ignition point of paper, 451F or 233C. As I recall, My HP Laser printer is about 430F - 440F, 221C - 227C.

 

[camerart]

Hi T,
Ok, thanks and noted.
When I did the first test, I stopped reading the temperature of the sandwich, when my thermometer got to it's max of 200C, so this figures.
C.