Lots of people have reported good things about the toner transfer method of making printed circuit boards. Lots of other people have said it’s a waste of time. I have been trying to use this technique to produce decent quality boards, with quite a few successes so far.
Before I discuss pitfalls and failures from my own experience, here’s a quick summary of what’s worked for me so far: –
- Printing onto sheets from Tesco’s magazine or an Argos catalogue (here in the UK), these gave best results. The Guardian supplement got second place. Any thin, cheap, glossy paper that quickly absorbs enough water to become slightly transparent in less than 30 seconds seems to work well for me.
- Masking tape in tiny (less than 1 sq cm) strips is enough to hold cheap thin paper to a piece of A5. I use three small pads to tape the top of the thin sheet to about an inch or so down the thicker sheet. The remainder of the thin sheet is folded and taped to the back of the thick sheet. This allows my HP Laserjet 5 printer to grab the sheet into the printer without having to fight two paper densities (causing chew).
- I use hatched copper fills and large isolation gaps, or else no copper-fill at all. Copper-fill requires lots of toner, and so less toner goes to the tracks. It’s likely that your laser printer won’t print large areas of black very well – it might look ok, but the toner-density in that area will have suffered. Remember, you want the track’s toner-resist to be as thick as possible. You can always draw copper fill with a waterproof permanent marker. I’ve found that this resists etchant far better than toner anyway.
- If I use cheap paper epoxy boards that are flexible, rather than FR4 which is more rigid, then I make sure the boards are bent slightly, to be convex to the iron rather than concave. In other words, the board lies flat with a slight arch rather than a slight trough. This allows the iron to press the board flat from the centre rather than from the corners.
- When scrubbing the board for cleaning and abrasion, I consciously focus on edges and corners, since these otherwise seem to get the least abrasion. Toner adheres better to scratches than to smooth copper.
- Bear in mind that a laminator can transfer toner in a single pass of less than one minute if it’s running at around 170 degrees centigrade (in other words, if it’s been modified to generate a third more heat). Therefore, I don’t tend to iron for more than a minute or two, I don’t use a lot of pressure with the iron, but I do use a plastic scraper or rubber roller to press the paper on to the board between applications of the iron.
I’m sure my failures have been down to something I’m doing wrong – I often deliberately ‘do things wrong’ just to see what the outcome is; I like to test boundaries! I have had at least one very successful etching from this process, so I know it has the potential to be consistently achievable.
The reason for the mixed reports on using toner transfer to make PCBs is easy to see – there are just so many variables.
- Paper type
- Toner type
- Printer settings
- Layout parameters (track width, clearances, pad-size)
- Board preparation (sanding, scratching, cleaning)
- Cleaning products
- Iron type, temperature, heat distribution.
- Ironing technique (board pre-heat, pressure, movement)
- Soak time
Then throw in possible mistakes such as leaving grease on the board or having dirt on the paper, and it’s obvious at a glance that this is going to require some practice to master. People can only reliably write about their own experiences with toner transfer, probably gained through hours of experimenting, the nuances of which will be difficult to convey.
To be more scientific about it, we would have to go into a lot of detail on the various permutations that work. Alternatively, we could be a bit more empirical and just state what we’ve observed to work.
For me, I use my home 2400 watt Bosch iron. It has a dial with 1, 2, 3, and ‘max’. I’ve found that setting ‘3’ works best to transfer the toner. The maximum setting is capable of burning some types of paper (just small holes). Lower settings may well work.
I use old magazine pages, typically white with small black text (e.g. an article). I avoid areas of heavy ink because I’m suspicious that some can interfere with the toner adherence. In order to get sheets through my old ‘HP LaserJet 5’ (my newer Xerox printer doesn’t seem to give as good results, I’ll probably investigate this later), I fold it over a sheet of A5, so with around an inch or so folded over to the back and taped with a couple of very small tears of masking tape at either end. Therefore, my transfer sheet has to be at least a couple of inches taller than A5 so that it will fold over. I’ve also found that it must be the full A5 width, because otherwise different rollers grab different bits of paper causing it to twist and jam.
It seems that board preparation is an enormously important step. I’ve had good results with a board that’s been sanded with fine sandpaper and also scoured with the rough side of a kitchen sponge. The sanding and scouring introduce fine scratches on the copper that the copper adheres to. Without sanding (i.e. just scouring), I was getting near perfect boards, but ruined by an edge that the toner failed to adhere to.
I use Swarfega to clean the board when I’m scouring, and then I rinse it under hot water and rub to make sure the board is clean. The final stage of cleaning is to rub solvent (rubbing alcohol and acetone are mentioned, I use some stuff bought from Rapid Electronics) over the board with some kitchen roll or similar. I allow the remaining film to evaporate before continuing.
I iron directly onto the back of my transfer paper with light pressure. I don’t pre-heat the board, since I prefer to be able to align my paper flat on the board and simply press the iron flat for ten seconds to make get the toner to initially adhere. Less risky than smearing toner on a hot board as the paper settles.
With the board still hot, I use a rubber-roller to apply even pressure over the surface of the board, and usually repeat this heating and rolling sequence a couple of times. I’ve also had success simply rubbing my finger lightly but firmly over the hot board, wearing a heavy-duty rough suede glove (the kind you’d buy from a builders merchant).
I allow the board to cool before soaking it in water. I’ve not noticed any difference dunking a hot board straight into water, but I usually give it a few minutes until it’s cool enough to hold. After the paper has absorbed its first ‘drink’ of water, I peel it off, leaving a rough paper layer underneath. I allow this to further soak before rubbing it off with my finger.
Toner that has adhered well to the board surface will withstand rough treatment. I can scratch quite vigorously with a fingernail or a scouring pad without removing it. However there is usually some weaker toner on my boards that will come away if treated as roughly as that. Gaps in tracks can often be redrawn with a waterproof permanent marker.
After etching (or to clean off a failed toner-transfer), I clean up the boards with some mild acidic hand-cleaner (Stoko ‘Slig’, in my case); smeared on for 30 seconds and rubs away easily with the rough side of a sponge (it actually rubs away easily even just with your fingers). The only thing you should waste while experimenting is paper, toner, and cleaner.
How to make really really good homemade PCBs – this page contains a wealth of information, clearly the result of much hard work and ingenious experimentation. Although it doesn’t discuss using toner for etch-resist, there’s still lots of useful information.