Abouth old-days PCB fabrication method.
OK, now this topic is probably going to frustrate some readers because it's like asking someone how to make a spoon. Whatever. As Iam very interested in building vintage-looking projects, I also want them to be built using the methods of the time period they are supposed to belong to. So, I am looking for information on how printed circuit boards used to be made in the late-sixties to seventies.
More precisely how was a pattenr copied, say, you had an amplifier circuit you wanted to copy on a board four times. HOW was this done?? The only thing I can come up with was that you would take your artwork, photocopy it as many times as you wanted (I AM ASSUMING that photocopiers existed in that period) and then cut out the circuit and essentially past them next to one another onto a master paper and then copy that to a master celluloid sheet, which you would then use to do the UV etch method.
AM I too far off here. Just for history's sake, I would really like some info here, please can somebody help me out here?
Maybe someone who worked at Tektronix, EICO or Heathkit can educate me on this subject.
Thank you very much!
OK, now this topic is probably going to frustrate some readers because it's like asking someone how to make a spoon. Whatever. As Iam very interested in building vintage-looking projects, I also want them to be built using the methods of the time period they are supposed to belong to. So, I am looking for information on how printed circuit boards used to be made in the late-sixties to seventies.
More precisely how was a pattenr copied, say, you had an amplifier circuit you wanted to copy on a board four times. HOW was this done?? The only thing I can come up with was that you would take your artwork, photocopy it as many times as you wanted (I AM ASSUMING that photocopiers existed in that period) and then cut out the circuit and essentially past them next to one another onto a master paper and then copy that to a master celluloid sheet, which you would then use to do the UV etch method.
AM I too far off here. Just for history's sake, I would really like some info here, please can somebody help me out here?
Maybe someone who worked at Tektronix, EICO or Heathkit can educate me on this subject.
Thank you very much!
Originally,
PCB's were laid up on a light table using clear Mylar film. The light table usually had a grid sheet so you arrange things relatively precisely. The work was usually performed at 2 or 4 times actual size.
Small, precut donuts and tapes were used for traces and pads. These tapes were made from black crepe paper so you could bend them into curves etc... very much like masking tape.
Often multiple layers of mylar were used to avoid redundant effort. For example, a 2-sided board was usually made from 3 sheets. All the pads or "donuts would be placed on one sheet... called the pad master. Another sheet would then be layed over the master, this is where the tapes were laid to represent traces for one side of the board. Repeat for the other side of the board. Even for single sided boards a separate pad master was used.
Once all these lovely sheets were done, you took them to a special photoshop. They would hang the sheets on a light table with a vacuum frame. They would then use film the same size as the actual finished board. Using different combinations of sheets, they would create film negatives for the different layers of the board. They would also create a negative of the pad master.
These negatives were then photo'd again to create film positives. These were "the master films". Why the pad master? when making the film positives, the pad master was deliberately over-exposed, this would make the pads larger... this film would be used for the solder mask on the final PCB. Yet another pad master would be created without any exposure tricks.
All these films went to the circuit board shop... this is where the "step and repeat" work was done. The PCB shops used a special type of film that was less sensitive to light. In this way the S & R work could be done in low light (red) conditions. The final working films were usually negatives. And there is bunch of different methods for exposing and developing the PCB material.
Why the pad master?... again? once a full size, S & R'ed working film was created, a program had to made for the drilling machine.
The film was laid on a light table and a special magnifier with a crosshair was positioned over each pad. Once "bomb-sighted", a button was pushed, the computer logged the position and punched the tape that stored the drilling program.
All of this SUCKED! I don't even think they make circuit boards like this in the Steppes of Mongolia anymore. I'll bet you can still find the special tapes and precuts on the Russion black market. The company that owned that market here in the states bellied up in the early nineties after some crazy employee buyout thing to try and save their jobs.
All this stuff was pricey as hell. A grid for light table was $200. Mylar sheets were $20. Little boxes of donuts were $5... as were the rolls of tape.
How to replicate the old look? What makes 'em look old is the gently curving tracks of copper. Do your boards using modern tools. Export the files in DXF format. Replace the traces one by one with splines (curvy-lines). Convert the splines to polylines and assign a width. Bring all this back into your PCB environment and make your gerbs from there.
🙂
PCB's were laid up on a light table using clear Mylar film. The light table usually had a grid sheet so you arrange things relatively precisely. The work was usually performed at 2 or 4 times actual size.
Small, precut donuts and tapes were used for traces and pads. These tapes were made from black crepe paper so you could bend them into curves etc... very much like masking tape.
Often multiple layers of mylar were used to avoid redundant effort. For example, a 2-sided board was usually made from 3 sheets. All the pads or "donuts would be placed on one sheet... called the pad master. Another sheet would then be layed over the master, this is where the tapes were laid to represent traces for one side of the board. Repeat for the other side of the board. Even for single sided boards a separate pad master was used.
Once all these lovely sheets were done, you took them to a special photoshop. They would hang the sheets on a light table with a vacuum frame. They would then use film the same size as the actual finished board. Using different combinations of sheets, they would create film negatives for the different layers of the board. They would also create a negative of the pad master.
These negatives were then photo'd again to create film positives. These were "the master films". Why the pad master? when making the film positives, the pad master was deliberately over-exposed, this would make the pads larger... this film would be used for the solder mask on the final PCB. Yet another pad master would be created without any exposure tricks.
All these films went to the circuit board shop... this is where the "step and repeat" work was done. The PCB shops used a special type of film that was less sensitive to light. In this way the S & R work could be done in low light (red) conditions. The final working films were usually negatives. And there is bunch of different methods for exposing and developing the PCB material.
Why the pad master?... again? once a full size, S & R'ed working film was created, a program had to made for the drilling machine.
The film was laid on a light table and a special magnifier with a crosshair was positioned over each pad. Once "bomb-sighted", a button was pushed, the computer logged the position and punched the tape that stored the drilling program.
All of this SUCKED! I don't even think they make circuit boards like this in the Steppes of Mongolia anymore. I'll bet you can still find the special tapes and precuts on the Russion black market. The company that owned that market here in the states bellied up in the early nineties after some crazy employee buyout thing to try and save their jobs.
All this stuff was pricey as hell. A grid for light table was $200. Mylar sheets were $20. Little boxes of donuts were $5... as were the rolls of tape.
How to replicate the old look? What makes 'em look old is the gently curving tracks of copper. Do your boards using modern tools. Export the files in DXF format. Replace the traces one by one with splines (curvy-lines). Convert the splines to polylines and assign a width. Bring all this back into your PCB environment and make your gerbs from there.
🙂
An even older method
In my day - late 60s - we used to draw the circuits in black Indian ink on a smooth card called Bristol Board, 10 times full size. This was after laying out the circuit in pencil on graph paper!
I once had to draw a multilayer board, 8, if I remember for a flight recorder. Each layer was on a separate piece of card. You had to be accurate - try erasing this ink without leaving a trace.
The cards were then photographed and negatives were made the correct size to expose the board.
Tony
In my day - late 60s - we used to draw the circuits in black Indian ink on a smooth card called Bristol Board, 10 times full size. This was after laying out the circuit in pencil on graph paper!
I once had to draw a multilayer board, 8, if I remember for a flight recorder. Each layer was on a separate piece of card. You had to be accurate - try erasing this ink without leaving a trace.
The cards were then photographed and negatives were made the correct size to expose the board.
Tony
We did a double-sided board once, by using precut tape of the type Poobah mentioned, but this particular PCB manufacturer wanted the layers on the same film, using red and blue tape for the different layers. I assume they then used very strong coloured lights and filters to separate this into two different films photographically. But this was c:a 1980, so it is perhaps to modern? 🙂
For home use and single layer, I also used the black type of tape and precut symbols. Or sometimes, with less success, the special felt tip pens sold to draw the pattern by hand.
For home use and single layer, I also used the black type of tape and precut symbols. Or sometimes, with less success, the special felt tip pens sold to draw the pattern by hand.
That red/blue stuff you couldn't turn corners with... you had to miter all the corners with a knife.
Were you working at 1:1 scale?
🙂
Were you working at 1:1 scale?
🙂
poobah said:
Fortunately, I wasn't doing it myself. We had a very patient guy who did it all the tape work. I don't know about the corners, but you are probably right. It wasn't the crepe type of tape.
Guessing from memory, we probably did it twice the sice of the final PCB.
I think the board maker messed it up anyway in the end, so he had to do a new batch for us, at his own expense.
For home use.
Layout the holes using graph paper (10 or 20 lines per inch). Tape to PCB material and use a center punch to mark all your centers. Drill with appropriate drill size, I use a pin chuck as well to hold these things.
Now you remove the paper, clean the PCB material with steel wool followed by thinners of some kind. Using the pen of your choice, connect the dots and fill copper areas. Let dry and etch. Rinse well and use thinner to get rid of the ink. Now you can tin the board, or use solder coat. If you want more than one, place the board you just did on a couple more and drill using your first as a guide. This is before populating it of course!
I like the "bluing ink" pens I think PC33 by Lenbrook.
Yeah, it stinks doing it that way.
-Chris
Layout the holes using graph paper (10 or 20 lines per inch). Tape to PCB material and use a center punch to mark all your centers. Drill with appropriate drill size, I use a pin chuck as well to hold these things.
Now you remove the paper, clean the PCB material with steel wool followed by thinners of some kind. Using the pen of your choice, connect the dots and fill copper areas. Let dry and etch. Rinse well and use thinner to get rid of the ink. Now you can tin the board, or use solder coat. If you want more than one, place the board you just did on a couple more and drill using your first as a guide. This is before populating it of course!
I like the "bluing ink" pens I think PC33 by Lenbrook.
Yeah, it stinks doing it that way.
-Chris
I never make boards.
I take the CAD/CAM girl at my board shop to lunch at least twice a year. She puts my proto's on the same panel with somebody elses proto's and I get them free... well sort of...
😀
I take the CAD/CAM girl at my board shop to lunch at least twice a year. She puts my proto's on the same panel with somebody elses proto's and I get them free... well sort of...
😀
In the late 80's (not exactly vintage era yet) I bought a PCB starter kit from radio shack, and it included sheets of little adhesive donuts and a couple rolls of flexible tape. The idea was to just lay out the circuit right on the copper and etch it with the adhesive forms being the etch resist. The result was very handmade or orgainc looking. I don't see any reason that a one-off proto would not be done the same way 10 or 20 years earlier.
Today, for small one-off single layer boards I use rub-on transfers for pads and some straight traces, and sharpie pens for curved traces. Because it is done by hand it looks like it was.
Today, for small one-off single layer boards I use rub-on transfers for pads and some straight traces, and sharpie pens for curved traces. Because it is done by hand it looks like it was.
Chris, how well does that pen work? Do the traces come out clean at the edges and no pitting? I made a board a couple weekends ago using a suggestion from Morgan Jones, red enamel paint applied in the same connect-the-dots manner you mentioned. It's tedious and fiddly, a pain to set up and clean up, but it works. The pen sounds easier.
Hi SY,
This pen works great. It is resistant to undercutting and pitting. Board preparation is critical, but after that it's a breeze (and tedious). You can do double sided PCB's this way easily. Registration is guarantied (ha ha
)
I have been considering purchasing some bluing ink from a machinist's shop to paint large copper areas with. I don't know the pen uses bluing ink, but it sure looks like it.
The other way I've made boards is with rub down transfers on clear mylar using a grid underneath. Then exposing positive blanks. If I can get a laser printer to make litho type prints on a clear film, I'll return to doing things that way.
Next time you are in the area, we can look at some examples. I imagine you may have some ideas on how to improve the process.
-Chris
This pen works great. It is resistant to undercutting and pitting. Board preparation is critical, but after that it's a breeze (and tedious). You can do double sided PCB's this way easily. Registration is guarantied (ha ha
)I have been considering purchasing some bluing ink from a machinist's shop to paint large copper areas with. I don't know the pen uses bluing ink, but it sure looks like it.
The other way I've made boards is with rub down transfers on clear mylar using a grid underneath. Then exposing positive blanks. If I can get a laser printer to make litho type prints on a clear film, I'll return to doing things that way.
Next time you are in the area, we can look at some examples. I imagine you may have some ideas on how to improve the process.
-Chris
Well, I've done a lot of work in those photographic transfer processes and have made a few million flex circuits, so I'm familiar with all the right ways to get a resist pattern on a board. The right ways are capital-intensive. My issue is how to do the same thing without access to that sort of equipment- the pen tip was GREATLY appreciated.
Hi SY.
Another hint. Ferric Chloride works fine if you float the board up side down and warm the solution. For trouble sided boards, either hang them in the solution, or sit them on something in the solution. Agitate and flip after some time (like 10 minutes or so). The same process would work fine for Ammonium Persulfate with shorter times.
My boards survived a spray etcher - twice in a row. My instructor at Ryerson lost the bet & I won another FM MPX PCB (the bet).
I am also going to try an iron-on pattern. Must get an ink jet for this. I was considering trying it on my laserjet, but I don't want to muck it up.
-Chris
Another hint. Ferric Chloride works fine if you float the board up side down and warm the solution. For trouble sided boards, either hang them in the solution, or sit them on something in the solution. Agitate and flip after some time (like 10 minutes or so). The same process would work fine for Ammonium Persulfate with shorter times.
My boards survived a spray etcher - twice in a row. My instructor at Ryerson lost the bet & I won another FM MPX PCB (the bet).
I am also going to try an iron-on pattern. Must get an ink jet for this. I was considering trying it on my laserjet, but I don't want to muck it up.
-Chris
macboy said:
I used that technique a lot long ago, but was never satisfied with it. The flex tape had a tendency not to stick tight enough to the pads, so usually some of the traces were broken just at the pad.
SY,
for a cheap and simpler alternative to ordinary photoresist methods, you might try the toner transfer method that has been discussed on the forum before. It requires that you have a laser printer, though. And it may take a bit of experimenting to get it right the first time.
Hi Christer,
I have not been able to find the medium for the printer. This is exactly what I want to try!
-Chris
I have not been able to find the medium for the printer. This is exactly what I want to try!
-Chris
anatech said:
You mean for toner transfer?
Well, you use a laser printer and ordinary photo paper for printers. The problem is to find a photo paper that works well. I had to try a few different ones to find a type that worked well. You want the toner to transfer well to the PCB when you iron the paper, but you don't want the paper to stick to hard to the PCB since you have to remove it. It takes some experimenting, unless somebody can recommend you a working paper type that should be available where you live. The one I found best is unfortunately a swedish brand, although I am sure it is some cheap generic paper they just sell as their own brand. I think somebody reported success with an Epson paper. Maybe you can find something if you seach the forum.
I suppose there could be a theoretical risk that some photo papers could melt and stick in a laser printer, but I have never read about anyone having such problems, and experienced no such problems myself with any of the 3 or 4 brands I tried.
Ah, and don't assume the expensive photo papers are best for this purpose. I have a feeling it is rather the cheap and bad ones that are good this. 🙂
I have been intrigued by that toner transfer thing also. At worst, it could serve as base that you could recoat or touch up with a pen.
There are also the clear films for jet & laser printers that could make nice negatives. I've never had good luck with sensitized boards and the developing and stuff (despite following directions). I think the "home versions" have watered down chemicals etc... so people in California and Europe don't get cancer.
🙂
There are also the clear films for jet & laser printers that could make nice negatives. I've never had good luck with sensitized boards and the developing and stuff (despite following directions). I think the "home versions" have watered down chemicals etc... so people in California and Europe don't get cancer.
🙂
- Status
- Not open for further replies.