This article includes details about the success I have had making Printed Circuit Boards (PCBs) at home. Over probably 25-30 years I have made my own PCBs from projects in EA and Silicon Chip magazines, producing one-offs.

I have tried just about all methods, the Positive resist photo-art, Negative Resist, Direct artwork and etch (decals and pen), but I never did have much success with Toner Transfer method. Many good articles describe it. Here are two good ones:

http://gilmore2.chem.northwestern.ed...garbz2_prj.php
http://www.fullnet.com/~tomg/gooteepc.htm

Anyway, my attempts years ago with Toner Transfer method always were touch and go. I could never rely on fine details to reproduce. Surface Mount Devices a definite no-go. Even DIL ICs sometimes had tracks joined and fine lines with pinholes etc. Well my need to produce a quick, accurate one-off a few weeks back made me re-try the TT method.

The success all comes down to the choice of printer paper as the above links stress several times. I thought I'd be smart and try normal A4 paper. The result was pinholes, tracks lifted, Failure!

I then tried tracer paper A4: The same result.

Then I tried some good quality photo printer paper for bubblejets. I used Kodak Picture Paper Cat # 835 0613 on a HP laserjet (toner) printer.

Outstanding success!

The artwork was reproduced intact and it etched the best I have EVER etched an artwork. That includes some of my best stuff done with the reliable Riston process. I am so impressed with how faithfully it worked that I am confident it would work with dual layer stuff and SMD.

I tried several re-tries so that I could demonstrate it wasn't a fluke. The links above seem to infer that the method is not really for a fine detailed project but I now have every confidence that it is.

Click to Enlarge This is a board I designed and made for my Stepper motor-focuser project

A few notes about the Toner Transfer Method

My aim is not to duplicate the above links but just record the method that worked for me because it was so successful. It may be of some help to people making quick, repeatable PCBs like I do occasionally.

Ok, so here is a detailed description of how this process works. Again, this process makes high detail artwork survive the etch process faithfully, and therefore, is a perfectly viable cheap method.

1. Happy with the artwork? I printed it out on an old laserjet HP-5P model using Kodak Picture Paper (for inkjet printers) A4 75sheets Cat #835 0613.
2. I prepared a bare PCB with a light brush of a kitchen scour pad with gentle strokes. Not steel wool because that can leave harsh scratches that let etchant seep under resist material.
   Then, as per the supplied links, I gave the raw PCB a quick wipe with paint thinner (or acetone) to make it as clean as possible. Then I placed the printout carefully face-down on the PCB ready to iron it on.
3. With the clothes iron on the highest setting, no steam, I carefully put pressure over the entire area without any sideways motion to avoid moving artwork initially.
   Then the artwork 'glues' itself to the PCB allowing sideways motion with the iron. I found I could work the artwork using the tip of the iron (the hottest point) and actually see the artwork appear through the bubblejet paper slightly.
   (Picture on the right below shows the artwork appearing after about 1 minute of ironing)
   

Click to Enlarge The artwork		Click to Enlarge After 1 minute of ironing

4. The PCB is very hot now and I dropped it into the kitchen sink filled with warm water (below).
 After about a minute the backing of the bubblejet paper can be carefully peeled off and it leaves behind the thin printout layer still stuck to the PCB.

5. After very carefully rolling the printout layer with your thumb under water you can remove all of the paper remnants until the artwork (toner) is left intact. In my case it takes on a greyish appearance. Carefully inspect you have 'rolled' all of the paper off, especially between IC pads etc because if you leave ANY paper behind at this stage the etchant will also leave that copper behind.

Click to Enlarge In the kitchen sink		Click to Enlarge The toner survives fairly rough rubbing as long as it isn't scratched by a fingernail. Above shows the board ready to etch, alongside the original artwork and a fully tested board

5. Etching (below left) shows my small etching tank, fed with regulated compressed air to speed up the process. Pre-heating Ferric Chloride to above 30-40°C also speeds it up. Be very careful with FeCl because it is highly corrosive. Etching is complete in about 2-3 mins with good fresh Ferric Chloride.

6. All done! (below right). A completed board, washed down and ready to drill.

Click to Enlarge Etching		Click to Enlarge Complete and ready to drill

Below are some shots of some SMD boards I made.

A rough LM335 test board. The pattern shows an SO8 surface mount spacing. Again, very fine line reproduction. Impressed!

Click to Enlarge A rough LM335 test board		Click to Enlarge A rough LM335 test board

To summarise, this method really works! The best part about it is when I can go from having the design on screen, to a printout, to etch, then sitting in my hands ready to solder in about ten minutes.

The bottom line is that it works reliably with a minimum of variables that might mean  success or failure. You’ve gotta be happy with that! :)

Article by Steve B (kinetic). Discuss this Article on the IceInSpace Forum.