PCB photofabrication

[mik3ca]

Until now, I was using transparencies as the media to put my PCB artwork on in order to complete the photofabrication process.

The PCB works excellent with a transparency, BUT, in order for this to happen, the artwork must be excellent. In most cases, it is not. When I pass the transparency (designed for  inkjet printers) into an inkjet printer, it prints fine, but under bright light, I see random holes in my design (non-printable spots). For this reason, I need to pass the same transparency through the printer twice.

The main problem I have with the second pass is that the two copies of the same design do not print one of top of the other. When this is successful (rarely), the image is darkened and no light can pass through it at all.

My printer settings won't help because I already set it to the darkest setting.

Now I am trying tracing paper. The price is excellent for each sheet, but the only problem I have with it is that it takes alot longer to remove the unwanted photoresist, and alot longer to expose the design and the board to light.

With a transparency, I can expose a board and remove unwanted photoresist in an hour.

I haven't changed the ratio of developer to water during both cases.

I am wondering if there is better media to use to print the design on. I don't want to spend alot of money on it, and I want the media to work with an inkjet printer as well.

any suggestions?

 

[MP1]

I have made thousands of boards using transparencies from an inkjet printer. How old is the printer? The old inkjets did not have very good resolution. You can get a new high resolution inkjet for US$50 if you shop around.
I use darkest color and highest resolution. I also select transparency from the menu. I have never had to print twice.
One thing you might try is to go to the adjustment screen for your printer and fool it into thinking you are not seeing the test spots dark enough. This will cause the printer to go darker. It will be useless for printing on paper, but it will be good for your board transparencies.

MP

 

[mik3ca]

I have made thousands of boards using transparencies from an inkjet printer.

I need to learn from you.

How old is the printer?

about 5 years.
My printer uses a parallel port. It doesn't support USB.

The old inkjets did not have very good resolution.

I don't think my printer qualifies as an "old inkjet" if sometimes the transparency works. The major problem I have with using a transparency is that the second copy of the same image does not align properly with the first copy, and the result is a shifted image, making the design useless. I have to pass the transparency through twice in order to even attempt a sharp image.

You can get a new high resolution inkjet for US$50 if you shop around.

I prefer to spend less than $50. I am looking for better media.

I use darkest color and highest resolution.

maybe I need to use the darkest colour.

I also select transparency from the menu. I have never had to print twice.

In that case, my transparencies are junk. What brand are your transparencies?
I used Grand and Toy ones.

One thing you might try is to go to the adjustment screen for your printer and fool it into thinking you are not seeing the test spots dark enough. This will cause the printer to go darker.

I tried a super-dark setting on tracing paper, and the ink almost made the tracing paper wet. LOL

So now I know that my transparencies are garbage.

 

[mugsby]

have you tried press n peel blue? i'd give it a go except i don't like to drill holes.

http://www.cibs.co.uk/etch/x7.html

 

[MP1]

I use 3M transparencies. Hope you get it sorted out. I usually have no problems with traces down to 7 mil.

I hate the press and peel stuff. I ask too much from it and end up with too much touch-up afterwards.

MP

 

[Herman the German]

Here is another suggestion but I am sure it will work with Your system as well as it works with mine.

I am using EAGLE and for best printing results I use the solder and the component side (solder side is blue and component side is red). On the screen they look violet but on the paper they are just black.

That means to make an extra copy of Your layout just for printing purposes. Copy all traces onto the component side and the print will be deep black (as black as a road tunnel at night) printing both layers at one time. There will be no loss of the sharpness conerning traces, pads and vias. There is a good chance to fill unwanted holes on the solder side with colour on the component side.

I tried this also on a colour laser and very much to my surprise there was no significant difference.

For the developer I use "Aetznatron". Sorry I don't know the english word for that but the development is finished within 30 seconds. Any longer development will also erase the photoresist not exposed to light and You'll have a clean copper surface. The solution is 10grams per one liter of lukewarm water. 30 degrees will just be fine. Don't use higher temperatures. The result is unpredictable.

 

[mik3ca]

have you tried press n peel blue? i'd give it a go except i don't like to drill holes.

no, but according to a few articles that I read about it, it requires more effort than using the transparency method mentioned here.

Also, Press-n-peel is not available in any local electronic store.

and for best printing results I use the solder and the component side (solder side is blue and component side is red). On the screen they look violet but on the paper they are just black.

If I understand you correctly, then this method is ridiculous. If the solder side is NOT supposed to touch the component side on the circuit board, then why would I print the solder and the component side at the same time on the same side of the sheet?

That means to make an extra copy of Your layout just for printing purposes.

why?

Copy all traces onto the component side and the print will be deep black (as black as a road tunnel at night) printing both layers at one time. There will be no loss of the sharpness conerning traces, pads and vias. There is a good chance to fill unwanted holes on the solder side with colour on the component side.

I tried this also on a colour laser and very much to my surprise there was no significant difference.

I'm lost.

For the developer I use "Aetznatron". Sorry I don't know the english word for that but the development is finished within 30 seconds. Any longer development will also erase the photoresist not exposed to light and You'll have a clean copper surface. The solution is 10grams per one liter of lukewarm water. 30 degrees will just be fine. Don't use higher temperatures. The result is unpredictable.

The ONLY problem I have is that I'm not able to transfer the image from the computer to the circuit board successfully.
Once I get a super-sharp image on the best paper, everything else is then OK.

I use 3M transparencies. Hope you get it sorted out. I usually have no problems with traces down to 7 mil.

The smallest spacing between any of my two tracks is 100um. Is that less than 7mil?
The smallest track width of any of my tracks is 300um.

 

[MP1]

Your spacing is 10 mil. You should be fine.

MP

 

[mik3ca]

almost fine.

I chose to increase the spacing since I am opting for cheaper medium.

The circuit will print perfectly, but not dark (or solid) enough to make a practical PCB if I use a transparency.

I have a standard blank sheet of printing paper, but I don't know if that will work because it is a notch more opaque than the tissue paper I'm trying to use.

The tissue paper was bought from a dollar store which might be a problem.

When I print on it, it seems that some tracks are connected when they shouldn't be. I bet tissue paper can't isolate ink well.

Can anyone suggest a better medium? I went through tissue paper and transparencies without success.

transparencies happen to be expensive ($1.70 / sheet)
tissue paper on the other hand is cheaper for me ($1.00 / book of 40)

I also want a medium that can absorb ink well.

any suggestions?

 

[mik3ca]

I was wondering if I used the transparency, pass it through once through the printer, and create a longer exposure time, would that make a notable difference?

My exposure times are usually one hour, but if it is worth it, I am willing to up it to at least 3 or 4 hours.

 

[MP1]

wow, 1 hour is a long exposure time. I usually never expose for more than 3 minutes. Perhaps this is why you are getting different results. I use a light table that has 4 of the fluorescent "black lights" in it positioned approximately 120 mm above the glass. You do not need dark black transparencies. On lighter work, you can use less exposure time. This is where the spacing becomes critical. You cannot get real tight artwork with transparencies that are not dark. Try the black light method and make some time tests on the same board using a black piece of cardboard to cover the board. Expose a small area for 1 minute, then expose same area plus another section for 1 minute, then uncover another area and expose again for 1 minute, etc. This will give you 1 minute, 2 minute, 3 minute, etc. exposures. Then when you develop the board, you can find your best exposure time. You might want to switch to a better photo resist and developer.

MP

 

[mik3ca]

wow, 1 hour is a long exposure time. I usually never expose for more than 3 minutes.

according to several documentations, they suggest between 20 and 30 minutes. 3 minutes is very short.

I use a light table that has 4 of the fluorescent "black lights" in it

I don't have those kind of lights. What I am using is a halogen light that is commonly available.

You do not need dark black transparencies.

so what you are claiming is that my current light source is almost useless.

On lighter work, you can use less exposure time.

this isn't light work. I'm doing my own microcontroller circuit, and the spacing is tight (200um minimum).

This is where the spacing becomes critical. You cannot get real tight artwork with transparencies that are not dark.

I already doubled my spacing since I started this thread.

Try the black light method...

Where can I buy a light bulb that shines the "black light" you mention? It would be nice if I could buy it like an ordinary light bulb, rather than buy the whole light box, which is probably not available to me anyway.

and make some time tests on the same board using a black piece of cardboard to cover the board. Expose a small area for 1 minute, then expose same area plus another section for 1 minute, then uncover another area and expose again for 1 minute, etc. This will give you 1 minute, 2 minute, 3 minute, etc. exposures. Then when you develop the board, you can find your best exposure time. You might want to switch to a better photo resist and developer.

What is the absolute maximum exposure time I should use?

 

[MP1]

You would have to experiment around to see what the maximum exposure time would be. The light you are using is not very efficient. Do you have a place in your area that sells party supplies or lights for black light posters? There are many places where these are available. They really cut down the time for exposure. You can also get them in the round bulb type if you do not want to change your setup. When I built my system, I performed the test that I mentioned in the last thread. I killed a couple of boards but found my perfect height and exposure time.
The problem with slow exposure is that the light will find any small imperfections after time. The faster you can expose the artwork, the better.

MP

 

[mik3ca]

I just bought a blacklight from Spencers. It is similar to the model at the following URL:

http://www.spencersonline.com/index.cfm/fuseaction/products.detail/categoryID/4c03fc90-c45d-43ab-9664-11034dcbf79d/productID/30443747-4c20-40e7-8ed2-c81920107086/

I used it to try to expose the board. Fortunately, it was a test board that is 1 cm square.

I attempted to make a ridiculously simple PCB. I just did 5 vertical tracks with random spacing inbetween them. Then I expose the board to the new light for about 20 minutes. When I developed it, the results were not satisfactory. It seemed that 1/2 the lines were missing.

I'm wondering, what is the BEST ratio of liquid developer to Water?

Usually I do 20 parts water to 1 part developer, and I squirt more in if the photoresist doesn't disappear quick enough.

Yes, I store developer in a squirt bottle.  ;D  Each squirt is about a tablespoon.

 

[mik3ca]

I tried again and this time it works. It must be the timing.

 

[faizanbrohi]

Still i must say that Toner Transfer method is the best , i have printed an adaptor which is 7 to 10 mil and with perfect results . Just using simple glossy paper , which just cost's nothing compared to the photographic method and an iron. i have also tried the Photoresist method  but with not sufficient results. i still Stick with The Toner Transfer method .
i will show some pics of the board in the next post.

 

[MP1]

Faizanbrohi: We are discussing how to get good results with photo etch. Not glossy paper. There is another thread about using your method. You should post in that thread. By the way, I totally disagree with your comment. There are many tight designs where glossy paper just pushes out crap. Works well for the hobbyist, but more professional artwork requires a more professional method.

Mik3ca: Sounds like you are on the right track. You have brought the exposure time down from one hour to 20 minutes. Your developer ratio might not be the problem. The black light might have been over exposing your artwork. You might want to try another board with a 10 minute exposure time as well. My exposure fixture has four of the black light tubes like what you have. This is only so that I can expose a 12" x 12" board with no problem. At the height that I mentioned above in a previous post, I never go more than 3 minutes. Just for your own knowledge, you might want to see how short of an exposure time you can get.

MP

 

[mik3ca]

I'm gonna see what happens when I change the developer mix.

Usually what I do is mix developer with water, and then when I am done with it, I store it in a container and reuse the same mix until I no longer can use it reliably. As i use more and more of the same container, the colour of the contents become darker. (lightish brown-green).

As for the light, I have it about 1 foot away from the board and when I expose, I switch all other lights iff so that the rays from the blacklight are applied to the board. It looks nice in the dark anyway