PCB making photoresist

[lm1875]

hello people, I'm new here and this is my first thread on this forum.
so I'm making an lm1875 amp and I used Cadsoft Eagle for design. I printed it out and now I'm ready to expose it. So can I use 20W, 10W or 5W halogen bulb and what distance and time shall I expose it for? please help! :grin:

 

[ehico16]

like my method, i do screen printing like the one they use to do,badge,sticker and design on vest. screen printing on your pcb let it dry and and use etching your chemical to etch.

 

[jpanhalt]

The common phototresists require near UV (i.e., UVA). The range of 366 to 390 nm or so will work. Shorter wavelength doesn't help for a variety of reasons. Halogen bulbs can produce light in that range, usually nearer 390 to 400 nm (Halogen lamp - Wikipedia, the free encyclopedia). The best source, though, is a fluorescent light bulb. I use type BL-B. I think BL will work, but do not use the "germicidal" lamps. They are designed to produce mostly 254 nm. The BL-B and BL may not be standard labels in your area. Simply look for something that emits in the near UV. If you have a white shirt washed in a "whitener," it will tend to look much brighter under the correct wavelength.

As for time and distance of exposure, I use a 15W bulb at 6" . You need to adjust the time by experiment. My PCB's are made by Injectorall and require about 12 to 14 minutes of exposure. From what I have read, the ones from MG Chemical require less time.

John

 

[lm1875]

Okay thanks a lot guys it really hellped me. I'll go and buy board today so I can post some pictures of my amp project!

 

[lm1875]

hey guys i really need help. I bought the board, and did everything like it was supposed but it didn't work. Can someone please tell me how to expose and develop circuit board propery without using any UV and other special bulbs, just regular household bulbs, halogen or energy saving ones, the common ones i can buy anywhere...PLEASE I NEED INSTRUCTIONS QUICKLY!!! DETAIL! PLEASE!!!

 

[jpanhalt]

Where is "here?" Second, ordinary household bulbs are designed to make as much visible light as possible for efficiency. Finally, you have given no details about the product(s) you are using. They do vary. As I mentioned above, I am most familiar (>15 years) with Injectorall sold by DigiKey. I am less familiar with MG Chemicals boards, but I have read about them.

For us to give you detailed instructions, we need to know:
1) What brand of board are you using?
2) What brand of developer and at what concentration are you using it?
3) How did you do the development?
4) What type of transparency film are you using? Brand, as some have UV blockers in them.
5) How are you printing on the transparency, preferably brand of printer too?
6) Are you clamping the transparency to the board with a glass plate? How thick is the plate? Is it a special type of glass? If not, how are you holding the transparency to the PCB during exposure?
7) At what distance did you do the exposure?
9) Details about the lamp you used, wattage, brand, etc. (Skipped 8 to avoid the smilie.)

Believe me, you do not want a board that will quickly be exposed with room lights, unless you want to work in a dark room.

John

 

[lm1875]

Okey so:
1) Board brand is unknown, but it has a blue protective film on it.
2) As well unkown, it looks like salt, a bit bigger. On 10 grams I used 1 liter of water, like it says on webpage)
3) I put the board in developer and waited for image to appear.
4) Board is premade
5) I use laser printer at my fathers office, don't know the brand.
6) I use photo frame to keep it pressed togeather. I think it's regular glass, it's a cheep photoframe you can find everywhere. Glass is 2 mm thick.
7) I did it at 7 inches.
9) I used 15W energy saving light bulb. It dosen't says the brand but It's regular one you can find everywhere.

I work under red light.

 

[jpanhalt]

This is going to be difficult because of all the possibilities. Please bear with me. First and foremost, is your pre-sensitized PCB a positive or negative resist? They are quite different. You need to know that answer with certainty. In the trials that you have done, did the exposed ares or unexposed areas react with the developer? If the exposed areas reacted and seem to be removed, then you have a positive board. With the boards I use, which are positive resist from Injectorall, the exposed areas turn dark purple on immersion in the developer and float off the board.

Okey so:
1) Board brand is unknown, but it has a blue protective film on it.

MG Chemicals now uses a white protective cover on its positive-resist boards. It appears that Ever-Muse (Circuit Specialists) also has a white protective cover. Injectorall has no protective cover. They come in black bags. The dry film, negative resist from MG Chemicals has a transparent-blue protective film on it. I believe the actual product is made by DuPont. Your boards could be pre-laminated with a negative-resist. At this point, I think it is futile to try to make a PCB without knowing whether your resist is positive or negative acting. If you can take your board and cut it into smaller pieces, you can determine that basic fact by experiment. Make a very dense exposure mask with two broad (1 cm), dark lines with a narrow space (3 to 6 mm) between. Lay the image on a board, press it tightly with your exposure frame, and expose to sunlight for 3 to 5 minutes (just guessing). Then develop it. If the broad lines remain with resist on them, you have positive resist. If the narrow line remains with resist on it, you have negative resist. You can test for the presence of resist, if it is not obvious by looking at it, by adding a little of your etchant.

2) As well unkown, it looks like salt, a bit bigger. On 10 grams I used 1 liter of water, like it says on webpage)

Please give a link to that webpage. MG sells mostly liquids that need to be diluted. I use a solid (NaOH or KOH). I doubt what you have is solid NaOH or KOH. They absorb water from the air and look wet if left exposed for any lengthy period. So, I suspect your solid is either sodium silicate, sodium carbonate, or a mixture.

3) I put the board in developer and waited for image to appear.

Did an image appear? If so, was it positive or negative?

5) I use laser printer at my fathers office, don't know the brand.

What did you print onto, clear transparency or paper? If paper, did you treat it in any manner before using it as the mask?

6) I use photo frame to keep it pressed togeather. I think it's regular glass, it's a cheep photoframe you can find everywhere. Glass is 2 mm thick.

That should be fine. The weight of the glass should be enough. I use small spring clips, as that allows me to handle the sandwich and flip it for double-sided boards.

7) I did it at 7 inches.

That should be OK.

9) I used 15W energy saving light bulb. It dosen't says the brand but It's regular one you can find everywhere.

If it is fluorescent (spiral shaped), that is probably OK. They do come in different color temperatures nowadays. You want a high color temperature, if possible. Be sure to let it warm up and reach its full brightness.

I work under red light.

That is OK, but probably not necessary. Don't change now, though. Get the system working first.

John

 

[lm1875]

I did expose the board. Lines and my small signature I printed on a transparency paper. After second trial, when i soked it in, image appeared. It looked just like on transpareny. Lines were green, not to green, olive green maybe. Then I soked it into water. Than i accidently moved my finger over the board and I kinda "messed" it up but the picture was still there. Than I soked the board again in the developer and image became brighter and more clear. After that i soked it into etching solution and it started to react. But when my copper started to dissapear, copper that was supposed to remain started slowly to dissapear. When all unwanted copper was gone, lines were too damaged to make the board usefull,

 

[jpanhalt]

OK, it sounds like you have a positive resist.

Be sure the "ink" or toner side of the transparency is against the board. That involves a mirror inversion, but keeps the light from bending/scattering around the edges of lines and exposing areas that you don't want to expose. Print the image as dark as you can.

As for developer/exposure time, it sounds like both might have been a little short. With my boards, the development goes quickly (about 60 to 90 seconds) and the image is quite durable. In fact, I rub a sponge saturated with developer over the image to sharpen the edges. Wear gloves, of course. I use a stronger developer than you probably have, so you may need more time.

With underexposure, you will still see an image, but the difference between the exposed and unexposed areas will be less. By the time your developer eats through the underexposed areas, it will also be eating into the unexposed areas. Unfortunately, over exposure might give the same effect. However, since it took two trips to the developer to get the exposed areas clean, I am guessing you underexposed.

The way I found my exposure time was to print 4 or 5 copies of the same design on a transparency. Apply that to the blank PCB in the usual manner and cover all but one with opaque plastic (a piece of black trash bag works well). Expose one image, which will be your longest exposure, slide the shield to reveal the next one, so both will be exposed, and so forth. For example, if the exposure you initially used was 6 minutes and you have 4 patterns, I would expose the first pattern for 3 minutes, then each successive pattern for an additional 3 minutes. The longest exposure would then be 12 minutes (3 + 3x3 or twice the failed-run time) and shortest would be 3 minutes or half the failed run time. Of course, any timing sequence can be used. You simply want to bracket the time you used in the first run with exposure times that are sufficiently longer and shorter to tell whether your result was due to under or over exposure. My PCB's typically take 12 to 14 minutes of exposure.

What are you using for etching? The photo-resist is an organic resin. I have had problems with peroxide-based etchants eating at that resin. (I know the etching is just an oxidative process. The peroxide/acid baths just seem more aggressive.) I have best results with ferric chloride or cupric chloride. However, all etchants will begin etching under the resist ("undercutting") if left long enough.

If you can post a picture of your next attempt, that might help solve any remaining problems. I think you are well on your way to getting something usable, though. Whatever you do, be sure to keep notes so you can reproduce what works.

John

 

[rajudp]

for negative photoresist you can use sodium carbonate and
for positive photoresist you can use sodium hydroxide as developer

 

[lm1875]

OK, it sounds like you have a positive resist.

The way I found my exposure time was to print 4 or 5 copies of the same design on a transparency. Apply that to the blank PCB in the usual manner and cover all but one with opaque plastic (a piece of black trash bag works well). Expose one image, which will be your longest exposure, slide the shield to reveal the next one, so both will be exposed, and so forth. For example, if the exposure you initially used was 6 minutes and you have 4 patterns, I would expose the first pattern for 3 minutes, then each successive pattern for an additional 3 minutes. The longest exposure would then be 12 minutes (3 + 3x3 or twice the failed-run time) and shortest would be 3 minutes or half the failed run time. Of course, any timing sequence can be used. You simply want to bracket the time you used in the first run with exposure times that are sufficiently longer and shorter to tell whether your result was due to under or over exposure. My PCB's typically take 12 to 14 minutes of exposure.

Well it all sounds a bit more complicated. Is this way of making the board good? It looks pretty well: YouTube - ‪Circuit Skills: Circuit Board Etching‬‏
I use ferric chloride for etching.

 

[jpanhalt]

I thought that was exactly what we were talking about. The difference is that he is using a known board type (probably the same as MG Chemicals) and knew something about his exposure lamp. You didn't know what brand of board you had and knew nothing about the lamp. So, starting with that lack of information, you need to sort through the variables in a logical sequence. Otherwise, you will just be wandering in the woods and never get to a completed board, except by pure luck.

As for the laser transparencies, double copies works, but I have never tried it. My concern is light bleeding around the edges, accurate registration, and increased exposure time due to the double thickness of the film. I use either an inkjet printer on a high quality transparency (my favorite method), or I darken the laser transparency with dry-erase markers.

The latter is something I had not seen reported on before. The procedure is simple. First, get a dry erase marker. Practice on a blank piece of film. You will notice that once dry, it can be rubbed off very easily with facial tissue or soft toilet paper (American style ). I use black and would recommend that. The other colors may work too. I didn't try them and don't want to get into the reason for that here. It is too off topic.

Take your laser transparency and on the ink side, cover it completely with the dry erase marker. Actually, you don't have to cover it completely. Do it in single, smooth passes. DON'T SCRUB BACK AND FORTH (Fig. 3, attached). Then, once it is dry, gently wipe over the transparency with soft tissue. The dry erase will stick to the laser image and be removed from the clear areas. It requires a little skill, but not much. Your main concern is scratching the image (Fig. 4, attached).

If I get permission for an article, I can publish the whole method here, including photomicrographs and etching results.

I guess the figure titles don't show. Figure 3 is on the left; Figure 4 is the right image. The darker image on top is treated.

John