This thread details how to make a perfect PCB using nothing more than an inkjet printer to produce the artwork. Once you have all the equipment needed its quick and simple and cheap. The light box is the most expensive item but I'm sure its diy-able 🙂
Equipment and materials required.
1) A PC (obviously 🙂) and a program capable of letting you design dedicated artwork. I use Diptrace DipTrace - Schematic and PCB Design Software Its free and relatively easy to get to grips with.
2) A cheap inkjet printer to produce the artwork. To print the artwork you need special transparancy film such as this JETSTAR STANDARD. Although I haven't tried it, Amazon and the like have stuff of similar description much cheaper.
3) An ultra violet light box.
4) Photoresist developer such as this Mega-UK - UV Exposure Units.
5) Some presensitized PCB (Copper clad board that is precoated with the light sensistive film)
So here's how its done.
Using Diptrace or similar you produce your PCB design. I find it easiest to route manually get the layout exactly as I want. (If there is enough interest I could do a quick Diptrace tutorial too.)
IMPORTANT. The view I like to work with is the view AS IF YOU WERE LOOKING DOWN on the complete PCB from the component side. I find that easy to visualise. Also when you print the transparency, the INK side of the transparency is the side that will directly contact the PCB in the UV lightbox. This is important. The transparent film has a finite thickness. If the ink side were not in contact with the copper then you will get "light bleed" through the film stopping you creating fine traces. As you will see in some of the shots, Diptrace allows you to "mirror" the image for those of you that work as if you were looking at the copper side of the PCB rather than the component side.
Printing.
This is always the most difficult part and will be printer specific. In order not to waste the transparency sheets do test prints on paper first (using draft mode to save ink). The secret is to get the PCB located where you want on the A4 sheet such that you can perhaps save the rest of the sheet to use again. You can also cut the sheets into custom sizes (check your printer to what is acceptable) and print on these for small designs. These images show how you can move the image around on the paper. As always. Print a test on paper first.
Remember to UNCHECK the silk layer. A mistake I have made more than once. If you don't then that green outline of the IC (and all other parts) would appear in copper on the final PCB.
Printer Settings.
These are important. You want the darkest image possible. I go into the printer settings and turn the contrast up fully for this. I also select Matte Paper.
Hopefully you end up with a perfect artwork as shown here.
A close up. Notice how fine you can go and still get perfect results. SMD would be a breeze with this. In fact those pads are for SMD resistors.
Getting The Image On the PCB.
Having got a perfect transparency the next step is to use the light box to transfer the image to the presensitized board. My light box has two UV tubes and I run these for five minutes first. Apparently UV output can be 30% higher when they are warm. The presensitized PCB and artwork is now placed in the box. The ink side of the transparency should be in contact with the PCB. Close the light box and expose the PCB. I find three and half to three and three quarter minutes works well. When the exposure is complete remove the PCB taking care not to touch it. Also keep it as dark as possible.
Developing the PCB.
The exposed board is now developed in special solution. The stuff I use says use with a 9 to 1 ratio of water to solution. Thats too strong and can damage the photo resist. I find 18 to 1 works well and takes around 20 to 30 seconds to remove all the exposed resist. Don't leave it developing any longer than needed. You can see the green resist dissolve away like ink in water. Rinse the board thoroughly when developed. During all these steps keep the board out of strong light. At this point you should all being well have a board that is ready to etch in the normal way with ferric chloride. This is the devoper I use Mega-UK - PCB Chemicals (see item 600-010 the 1 ltr bottle). The transparency film is the Jetstar standard 100-070 Mega-UK - Artwork Films
Etching the PCB
For etching the board I must admit I always use Ferric Chloride, the stuff we used decades ago.
It is widely available from many suppliers as both crystals or pellets (also ready mixed) and works well. I use the pellets and mix as directed on the packet. Always add the crystals or pellets slowly to cold water (never the other way around because of the heat generated as they dissolve). Always use plastic containers and gloves are a good idea because it stains your fingers 😀 and keep any splashes off clothes etc. So what I do is this, I normally get a couple of packs of pellets and dissolve both in the correct amount of water and then store the mixed solution safely away. It keeps for years so there are no issues there. Just use a stout all plastic bottle.
For etching the board I always use small plastic boxes. For small boards and old margarine tubs are fine, bigger boards and food boxes are good. Its also surprising how little solution you need, a few mm of etchant in the container is all that is needed. At this stage I "float" the box I'm using in a sink of fairly hot water to heat the etchant. Place the board in the box with the copper side up so that you can see what's going on and so that nothing abrades the resist covering. Don't leave the board just sitting in the solution, constantly swirl it gently around keeping the etchant warm. Depending on the temperature the board should start to show visible signs of the copper coming off (corners and edges usually show first) after only 3 or 4 minutes. Another few minutes and it should all be done. Make sure you look the board over before deciding its had enough to be certain that no copper is left. The copper as it dissolves can leave areas so thin you can see through it so just give it that extra 30 secs or so to be sure. When its done remove and rinse thoroughly. Hold it up to the light and check all the copper is removed. It can always go back in if there is a problem.
The Finished PCB's As you can see, the fine traces are absolutely perfect. Doing a full scale large SMD project should be very successful.
Equipment and materials required.
1) A PC (obviously 🙂) and a program capable of letting you design dedicated artwork. I use Diptrace DipTrace - Schematic and PCB Design Software Its free and relatively easy to get to grips with.
2) A cheap inkjet printer to produce the artwork. To print the artwork you need special transparancy film such as this JETSTAR STANDARD. Although I haven't tried it, Amazon and the like have stuff of similar description much cheaper.
3) An ultra violet light box.
4) Photoresist developer such as this Mega-UK - UV Exposure Units.
5) Some presensitized PCB (Copper clad board that is precoated with the light sensistive film)
So here's how its done.
Using Diptrace or similar you produce your PCB design. I find it easiest to route manually get the layout exactly as I want. (If there is enough interest I could do a quick Diptrace tutorial too.)
IMPORTANT. The view I like to work with is the view AS IF YOU WERE LOOKING DOWN on the complete PCB from the component side. I find that easy to visualise. Also when you print the transparency, the INK side of the transparency is the side that will directly contact the PCB in the UV lightbox. This is important. The transparent film has a finite thickness. If the ink side were not in contact with the copper then you will get "light bleed" through the film stopping you creating fine traces. As you will see in some of the shots, Diptrace allows you to "mirror" the image for those of you that work as if you were looking at the copper side of the PCB rather than the component side.
Printing.
This is always the most difficult part and will be printer specific. In order not to waste the transparency sheets do test prints on paper first (using draft mode to save ink). The secret is to get the PCB located where you want on the A4 sheet such that you can perhaps save the rest of the sheet to use again. You can also cut the sheets into custom sizes (check your printer to what is acceptable) and print on these for small designs. These images show how you can move the image around on the paper. As always. Print a test on paper first.
Remember to UNCHECK the silk layer. A mistake I have made more than once. If you don't then that green outline of the IC (and all other parts) would appear in copper on the final PCB.
Printer Settings.
These are important. You want the darkest image possible. I go into the printer settings and turn the contrast up fully for this. I also select Matte Paper.
Hopefully you end up with a perfect artwork as shown here.
A close up. Notice how fine you can go and still get perfect results. SMD would be a breeze with this. In fact those pads are for SMD resistors.
Getting The Image On the PCB.
Having got a perfect transparency the next step is to use the light box to transfer the image to the presensitized board. My light box has two UV tubes and I run these for five minutes first. Apparently UV output can be 30% higher when they are warm. The presensitized PCB and artwork is now placed in the box. The ink side of the transparency should be in contact with the PCB. Close the light box and expose the PCB. I find three and half to three and three quarter minutes works well. When the exposure is complete remove the PCB taking care not to touch it. Also keep it as dark as possible.
Developing the PCB.
The exposed board is now developed in special solution. The stuff I use says use with a 9 to 1 ratio of water to solution. Thats too strong and can damage the photo resist. I find 18 to 1 works well and takes around 20 to 30 seconds to remove all the exposed resist. Don't leave it developing any longer than needed. You can see the green resist dissolve away like ink in water. Rinse the board thoroughly when developed. During all these steps keep the board out of strong light. At this point you should all being well have a board that is ready to etch in the normal way with ferric chloride. This is the devoper I use Mega-UK - PCB Chemicals (see item 600-010 the 1 ltr bottle). The transparency film is the Jetstar standard 100-070 Mega-UK - Artwork Films
Etching the PCB
For etching the board I must admit I always use Ferric Chloride, the stuff we used decades ago.
It is widely available from many suppliers as both crystals or pellets (also ready mixed) and works well. I use the pellets and mix as directed on the packet. Always add the crystals or pellets slowly to cold water (never the other way around because of the heat generated as they dissolve). Always use plastic containers and gloves are a good idea because it stains your fingers 😀 and keep any splashes off clothes etc. So what I do is this, I normally get a couple of packs of pellets and dissolve both in the correct amount of water and then store the mixed solution safely away. It keeps for years so there are no issues there. Just use a stout all plastic bottle.
For etching the board I always use small plastic boxes. For small boards and old margarine tubs are fine, bigger boards and food boxes are good. Its also surprising how little solution you need, a few mm of etchant in the container is all that is needed. At this stage I "float" the box I'm using in a sink of fairly hot water to heat the etchant. Place the board in the box with the copper side up so that you can see what's going on and so that nothing abrades the resist covering. Don't leave the board just sitting in the solution, constantly swirl it gently around keeping the etchant warm. Depending on the temperature the board should start to show visible signs of the copper coming off (corners and edges usually show first) after only 3 or 4 minutes. Another few minutes and it should all be done. Make sure you look the board over before deciding its had enough to be certain that no copper is left. The copper as it dissolves can leave areas so thin you can see through it so just give it that extra 30 secs or so to be sure. When its done remove and rinse thoroughly. Hold it up to the light and check all the copper is removed. It can always go back in if there is a problem.
The Finished PCB's As you can see, the fine traces are absolutely perfect. Doing a full scale large SMD project should be very successful.
I made my own light box using replacement tubes and starters bought from RS Components. Don't bother with tubes for sunbeds and other UV sources, they put out different wavelengths of UV.
As Mooly found, it is common in the UK for suppliers to sell "Fotoboard 2" pre-coated board. This requires half strength developer. With this board follow his 18:1 recommendation.
As Mooly found, it is common in the UK for suppliers to sell "Fotoboard 2" pre-coated board. This requires half strength developer. With this board follow his 18:1 recommendation.
Last edited:
Useful info there richie00boy, particularly about the tubes.
To begin with for me it was all very much trial and error. The PCB used in this example is sold by CPC as "Economy Board" which is fibre glass and paper laminate. Despite the economy tag I find it very pleasant to work with. It drills and files easily.
To begin with for me it was all very much trial and error. The PCB used in this example is sold by CPC as "Economy Board" which is fibre glass and paper laminate. Despite the economy tag I find it very pleasant to work with. It drills and files easily.
Good info , Mooly.
As additional info, the PCB developer for most positive resist PCBs in the market is nothing but sodium hydroxide. We've been using german Bungard proscts for years at work, and the recommended recipee is 14 g of NaOH pr litre. I've also used eurocard pcbs from other vendors with the same solution with no problems.
For the light box, the tubes used are called actinious blue. Osram type designations is TLA-15, IIRC. Two of those in a 50cm twin fluorescent lamp for e.g. kitchens will do the trick for PCBs up to 15 cm or so. I think Mega-UK alo sells these.
For the printouts, Mega also sells a sort of matte foil for laser printers called Laser Star, which work great with laser printers, A bit costly, though....
Stacked prints on ordinary laser print foil also works for not-too-thin traces.
As additional info, the PCB developer for most positive resist PCBs in the market is nothing but sodium hydroxide. We've been using german Bungard proscts for years at work, and the recommended recipee is 14 g of NaOH pr litre. I've also used eurocard pcbs from other vendors with the same solution with no problems.
For the light box, the tubes used are called actinious blue. Osram type designations is TLA-15, IIRC. Two of those in a 50cm twin fluorescent lamp for e.g. kitchens will do the trick for PCBs up to 15 cm or so. I think Mega-UK alo sells these.
For the printouts, Mega also sells a sort of matte foil for laser printers called Laser Star, which work great with laser printers, A bit costly, though....
Stacked prints on ordinary laser print foil also works for not-too-thin traces.
Thanks 🙂
I'd never heard of actinious blue before... had me looking it up, and the developer is just caustic soda.
I'd never heard of actinious blue before... had me looking it up, and the developer is just caustic soda.
There is Positiv 20 from german Contact Chemie, possibly also others.......
I have used it, but finding that even application is somewhat difficult. I think there are several 'How to's' on the net. Presensitized are relatively cheap, though....
I have used it, but finding that even application is somewhat difficult. I think there are several 'How to's' on the net. Presensitized are relatively cheap, though....
I think I have a can of that very stuff in the garage. It seemed like a good idea at the time but as you say, presensitized board is pretty cheap all things considered. I never actually used it tbh
I guess what is needed is to spin the board with the photo resist as a liquid on the surface and try and get an even coverage that way.
I guess what is needed is to spin the board with the photo resist as a liquid on the surface and try and get an even coverage that way.
Hi friends ,take a look here.http://www.diyaudio.com/forums/solid-state/221741-dx-blame-st-together-dx-super-116.html
Post #1152.May be is useful.
Thimios.
Post #1152.May be is useful.
Thimios.
Thank you Mooly for this!
For the complete newbie could you expand a bit on the following: (a quick synopsis of your procedure would be great)
"At this point you should all being well have a board that is ready to etch in the normal way with ferric chloride."
Steve.
For the complete newbie could you expand a bit on the following: (a quick synopsis of your procedure would be great)
"At this point you should all being well have a board that is ready to etch in the normal way with ferric chloride."
Steve.
Spray on photoresist is utter rubbish. Difficult to get even coverage and you need to apply it out of sun or fluorescent lighting and keep it dark to dry. You will probably get a few specs of dust in which may ruin your layout.
I think it's meant for boards with very big traces and pads.
I think it's meant for boards with very big traces and pads.
I have a lengthy wiki entry on how to make PCBs, I don't think many people know that it exists though.
The light exposure method is capable of giving you amazing detail. I now use the free designspark PCB CAD software and this works very well, although it is a pain to get started with.
I routinely make boards with 0.2mm width traces.
I've also built my own bubble etch tank out of perspex, a fish tank bubble curtain and power resistors encased in large diameter heatshrink tubing to keep things warm.
The light exposure method is capable of giving you amazing detail. I now use the free designspark PCB CAD software and this works very well, although it is a pain to get started with.
I routinely make boards with 0.2mm width traces.
I've also built my own bubble etch tank out of perspex, a fish tank bubble curtain and power resistors encased in large diameter heatshrink tubing to keep things warm.
That's a great write up you have there Thimios. The rice paper idea looks to work well for those with laser printers too.
Thanks Steve, hope you find it useful 🙂
I've added how to etch the board to the first post. Any questions... just ask.
Ta 🙂
Do you know, I've not seen that (you'll have to put a link to it)
I agree using the light method can generate remarkable quality PCB's. I made my first PCB this way at college (around 1980) but then had to revert back to hand drawn PCB's until I got a PC at home and was able to produce artwork.
The problem I always have is getting the image BLACK enough. With either my laser or inkjet printers the image still allows sufficient light through to spoil the final result.
I had lousy results using spray-on photoresist. What did work reasonably well was thinning liquid photoresist 50%, pouring some onto a a board and smoothing it with a disposable foam brush. Then, let it flow out and dry in a dark place until it is hard. Sometimes I used an improvised oven to bake it dry when I was in a hurry. The results were a lot less uniform than the pre-sensitized boards, with dust specs and hairs, and exposure times tended to be long.
I figured out the secret to getting the artwork black enough back in the days of penplotters on film: print two or more copies, and stack them.
I figured out the secret to getting the artwork black enough back in the days of penplotters on film: print two or more copies, and stack them.
That is what I resort to. One copy from the laser and a second from the inkjet. That's OK unless the artwork is complex.
Correct .
That's why I always feed them twice or three times through my El Cheapo Canon IP4500 InkJet and print the same image , that works great ! Set the contrast as high as possible in your advanced printer settings and set your print quality to high and to the proper print media .and let them dry for a couple of minutes before you print again .
It will take some practice to feed the sheet in perfectly for the 2th or 3th time , but the results are amazing ! and are very black !
Cheers ,
Rens
Last edited:
- Status
- Not open for further replies.