How to make a printed circuit board.
I occasionally make small simple single sided printed circuit boards and thought it might be helpful if I put down some of the things I have learnt.
Gather your components so you have their dimensions then on some 1/10 inch graph paper draw their positions and the tracks that are needed pretending you are looking down on the component side of the board. You can print out your own graph paper using free software (see my software page software.html ). You will find that most component leads fall on a 1/10 inch grid. If the pins on your components are closer than 1/10 inch you may find that the process of drawing the tracks later on the actual PCB is difficult. Now fix the paper to the component side of the board and mark each component lead hole by using a sharp point. To drill the holes you will need .8mm or 1mm drill bits for most components.
Now you need to draw the tracks and you have the holes to guide you but your paper layout is the reverse of what you will see when you are looking at the copper side of the board. My solution is to sit near a window and stick the paper on it so that I can see through it. Clean the board with a plastic scourer and detergent, a slightly rough surface helps the resist to flow neatly. To make the tracks you can use a resist pen from an electronics hobby store or you can try a mapping pen with some tar dissolved in turpentine. I actually had a lump of tar that I used for 20 years but when it took a hike I substituted it with some tar type roofing paint.
Some other things to try:-
Buy some pre-punched board (1/10 inch perf board) and stick the components in it to to help with judging the spacing and layout. This is what I usually do and it is why I do most layouts looking from the component side.
Print out the graph grid on some stout tracing paper that you can usually buy in pads. You will be able to see through it when you mark the copper side of the board and have a proper picture to follow when making the tracks.
The above board has been laid out from the copper side.
The small surface mount device above required tracks thinner than I could make manually but I got around the problem by hand making a board and using fine wire for the final connections.
The information below is a report of what I have found works with the products I have available. If you deviate from the suppliers recommendations that apply to any product it is your responsibility.
The problem I had with the size of the above surface mount device prompted me to look for a better way and I found it.
I located some PCB pre-coated with positive photo-resist from Dick Smith Electronics. The only clue that I have to the manufacturer is an "L" on the board. I also purchased some KINSTEN branded board from www.computronics.com.au it looks the same and it came with comprehensive instructions.
I also found a software programme that was free for non commercial use up to a 200 pin number limit. Look under downloads on the site.
Layo1 PCB from www.baas.nl is easy to use and can print directly to your windows printer or to commercial board making equipment. I used it for all the boards and layout pictures on this site. If you intend to draw the tracks by hand it is still a good option to use software as it bypasses the graph paper stage. Layo1 PCB can handle many layers but you can turn off the layers you do not need and only print the bottom copper layer. One hint is to make the minimum hole size to about .8mm. You do not need the correct hole size on the print out because you will be drilling the holes your self but if you adjust the size correctly then you will find the etched hole makes starting the drill so much easier, too big and the drill drifts off centre and too small and they may not etch. The software will also make any text you add mirrored as you view it from the parts side so it will etch correctly.
Making the transparency
Watch out for options that change the size and quality of the printout as there are heaps of them in PDF and laser printers, then do some experiments. If you have a laser printer you can print on the special laser printer film for over head projectors. There is a common but minor problem with pin holes as can be seen below.
As you will be using a full page of the film each time you should fit as many copies of your circuit as you can on the one page while you are in the PCB software. Check the print out with a magnifying glass and select the best result.
With an ink-jet you can print on special ink-jet transparency film. You may get a shock when you see the price so it pays to read the internet regarding the best brand to use and to cut the sheets up. I print out on scrap paper first then attach the film over the printout using the sticky part that I cut from post-it type notes.
The only media setting that I find satisfactory on my old Canon BJC-2000SP is "Gray scale" "Fabric printing". the best mode is "Graphics" . The "Fabric" setting really loads the film with ink, note how it builds up. Fabric printing on my latest Canon Pixima always seems to be in colour.
and this results in good opaque tracks as shown below.
Any other setting on my printer results in
the above result. Note that ink-jet transparency film I used had some problems. It was dirty when I got it and it picked up lint etc. as it was a little sticky. It tends to bleed ink away from the tracks over a period of some days and this causes the holes to fill up. This bleeding problem does not occur with photo gloss paper and I can not see why the same technology could not be used with a clear backing for transparency film --perhaps it is on better brands.
You can also use an ink-jet with tracing paper. This is thick and translucent and it can be bought in sheets or pads. The advantage is price and the ink will not bleed over time but the tracks are not quite as clean and the fibres can be seen sometimes as clear in the tracks.
The photo resist PCB is very forgiving and just as an experiment I made a board for the above surface mount chip using a 90 g/m2 ( as thin as I could find at ---www.inkme.com.au---) high resolution single sided coated paper printed using "standard paper" printer settings.
The above shows the difference between a good coated and a standard paper. With the coated paper the result is limited by my old Canon BJC-2000SP and its 720V X 360H resolution. With standard paper (right) the result is limited mostly by the fibres in the paper.
I smeared the printout with cooking oil to make it transparent and blotted it dry. A board exposed for 20 minutes, then given a spray with detergent and a rinse to completely remove residual oil before developing gave excellent results and faithfully reproduced the faults made by my old printer including the occasional almost invisible gap in a track. Without the oil the board required 80 minutes of exposure.
Perhaps if I substituted artists linseed oil or some thinned varnish for the cooking oil it would eventually set and give a permanent transparent result. A problem might be the paper buckling when the oil sets and thus not remaining flat on the board during exposure. This buckling problem would mainly affect larger boards and can also show up with laser printer film if the heat used to fuse the toner in the printer buckles the film.
Basically I think you could use any paper that gives a good pattern and can be made transparent with oil including light weight photo papers with no plastic backing. I tested a sample described as "Mirror finished coated paper for 1440 dpi. 140gms Best seller, gloss photo like finish" and it worked with a 25 minute exposure after oiling.
With coated or photo paper you may find it easier to get good artwork but the residual oil is a problem. Why not give it a try if you already have some on hand.
EXPOSING THE BOARD
It was a revelation when I found out that modern resist coatings could be exposed with a normal fluorescent or low cost compact fluorescent tubes. This means almost anybody can lash up some method of exposing a board. My fish tank light a sheet of glass from a photo frame and the kitchen timer is all I need.
With the board 55mm from the tube exposure takes about 11 minutes with film and 13 minutes with tracing paper. I let the light warm up for a couple of minutes then slide in the assembled stack on a sheet of paper. You should have the printed side against the resist and you should be able to read any lettering correctly through the film.
A compact fluorescent and an old desk lamp will also do the job.
On a simple job I used some board that had been in an unsealed bag for 4 years. I had to give it 3 X the exposure and extra development time but did the job.
DEVELOPMENT AND ETCHING
Development takes about 70-100 seconds in an agitated solution of 1 level teaspoon (5ml) of caustic soda beads dissolved in 750ml water. Do not try to skimp by making up some smaller quantity as measuring less than 5 ml is too inaccurate. It is fortunate that caustic soda works with the board I use as it is available at supermarkets at low cost but be aware that caustic soda is dangerous stuff so read and follow the precautions on the package.
I drill a hole in the board and loop some string through and etch in a ferric chloride solution giggling the string occasionally.
I then wash and dry the board then re-expose it and re-develop it. This exposes and washes away the coating on the tracks.
I used to drill boards with a hand drill and have done some small ones with a pin chuck and fingers. I now use a small rechargeable drill chucked onto the pin chuck. The only special drills that I have needed are .8mm and 1mm that I get from a hobby electronics shop.
Above I am getting the last out of a broken drill.
I hope the above encourages you to have a go at making a simple PCB.
BACK TO THE INDEX OF MY OTHER PAGES