Toner transfer is my favorite method for rapid PCB prototyping down to 10 mil traces. The basic steps are already described thoroughly, you can find some good tutorials on the net :
- Laser print the drawings on an appropriate paper (inkjet printing on a very glossy paper then spraying toner powder on wet traces and dusting off excess toner is a method I’ve tested with success too, very convenient since it works with any home inkjet)
- Iron the paper on PCB’s copper faces
- Soak in water to remove the paper
Step 2 may be very frustrating using an iron. I’ve tried it many times but even with clever tricks like toner transfer with dowel , my success rate is rather low. With a laminator, results are much much more reproducible but you have to laminate the PCB several times to fuse the toner solidly on the copper, « several » meaning « until you’re sick and tired » (15 to 20 passes).
Step 1: Principle
So I decided to hack the laminator by slowing down its rolls’ motor using a triac controller. Instead of laminating the PCB in 10, even 20 passes, the motor is switched on 1/2 second then off 10 seconds (off-times of 3s & 6 s can also be selected). That’s about the right amount of heat for the toner to fuse and stick to the copper. So you can feed once and attend to something else until the transfer is finished, the result will be consistent and perfect.
Step 2: Circuit Board
The laminator I hacked is bought new for 15 euros at the supermarket, it’s supposed to be used on thin films & paper but can be fed no problem with 1.6 mm FR4 PCB. I’ve used a friend’s much more expensive laminator with no better results so if you can buy a dirt cheap machine, I presume you should.
The triac controller is a slightly adapted design from a light dimmer kit I made in another project. The circuit is powered from the mains voltage (220V/50hz in France) through a capacitance. Schematics and microprocessor’s code are provided below. Compiled codes are for Microchip’s 16F84 and 16F628. 16F628s have an embedded 4 Mhz oscillator so the external 4 Mhz resonator can be omitted.
The source code is forked from an existing (tried and thoroughly tested) project , I haven’t tested this circuit on 60 hz mains but it should work.
BTW, I have some PCBs left, if anyone want one, please, let me know.
Step 3: Controller Board Integration
There is free space in the laminator, so I decided to put in the controller board, just below the laminator’s buttons.
I disconnect the laminator’s on/off switch and the red led and reroute them to the controller (see circle in the picture). Since the on/off switch is no more, you have to switch on/off with the mains plug. Otherwise, you’ll have to connect the controller the conventional way : drill a hole and add switch & led to the laminator or put the controller in an external box.
Controller’s switch modes:
1. ON : motor always on, this is equivalent to the original laminator mode (without controller)
2. OFF : motor 1/2 s on, 10s off
3. ON/OFF (switch is used as pushbutton): change the off duration of mode 2. The default is 1/2 s on+10 s off but you can change to more rapid speeds 1/2s on + 3 s off and 1/2s on+ 6 s off. On my laminator, the default speed has been the most suitable for a correct fusing in one pass, on higher temperature laminators, maybe
Now that all is connected and tested, the controller board is inserted into the laminator’s free space.
Step 4: Reassembling and Try a First Transfert
Laminator’s back cover is closed, no hole has been drilled and there is no visual difference after the hack.
Now it can be used to transfer PCBs. I can still use it as a standard laminator with the switch in ON position (motor in continuous mode). It works flawlessly in both modes.
An exemple of what I get from this hack after etching with hcl & H2O2, the double-sided pcb is for an ATMega168 in TQFP.
Step 5: ! Warnings !