Laser printer transparent paper for PCB making

[akis]

I have been making PCBs at home with very good success using specialist "Laserstar film" - it is like a transparency but not quite. Because it is expensive I switched to Gateway Tracing Paper 90gsm.

LaserStar Film:
Pros: good quality, good transparency, toner sticks to it nicely in large black areas
Cons: expensive

Gateway Tracing Paper:
Pros: cheap, does the job down to 0.7mm tracks and 0.5mm (have not attempted better than that).
Cons: not very transparent, ever so slightly stetchy (need to stretch it on the UV box glass), toner does not adhere well on large back areas


I am looking for experiences with other papers else I will buy the Laserstar Film.

[mamalala]

I'm using the film from Avery/Zweckform, #3491:

http://www.amazon.co.uk/Avery-3491-Transparent-Matt-Coated-Processable/dp/B0002S4NVM/ref=sr_1_1?ie=UTF8&qid=1403048951&sr=8-1&keywords=zweckform+3491

On the Germany Amazon it is cheaper, with 28,98 Euro for 100 sheets.

If you want i can send you a bunch of sheets to test them, just drop me an eMail then.

Greetings,

Chris

[nctnico]

I have been using "Glama basic"  62g/m2 tracing paper from Schoellershammer for about 20 years. Although it doesn't look transparent it seems UV has no trouble getting through it.

[sleemanj]

I use 110 gsm acid free tracing paper, I find this heavier weight produces better results. 

[tautech]

I have always cut my TT film/paper slightly larger than PCB and stuck it on top of a previously printed plain paper page, then reprint.
That's how this Scotsman saves on cost. 

[synapsis]

I tried 3M CG3300 on my laser printer and you could still see through it. I've had much better results with an inkjet photo printer and Sparco 01854.

Just in case, I print the mask twice, cut them out and stack them. Touch the four corners with a hot soldering iron (with an old tip!) and they'll bond together. Works great.

[oldway]

I use transparencies Retro laser from MultiCom but the adherence of the toner is low because they are smooth and shiny on both sides.

To let the toner adhere better, I make the printing side mat and rough by rubbing it with emery paper.
I superimpose two copies that I glue together for a better contrast

[jimjam]

when using tracing paper, is it easy enough to do a 2 layer pcb?

[nctnico]

You can. I used an L shaped piece of PCB material and stick the 2 layouts to it while carefully aligning them. However nowadays I just send 2 layer layouts to Seeedstudio. If I need something fast I keep one layer as a solid ground plane and use wires as vias. With mostly surface mount components that is very doable for a quick prototype.

[madires]

I'm using tracing paper too, also for double sided PCBs. To fix the problem with not so solid black areas I spray the printout with Solvent 50. Works great with my toner, but there might be some toners which don't like that treatment that much. That way I've done traces down to 0.3mm and packages like MSOP without any problems. For double sided PCBs the L shaped PCB helps with the alignment quite well.

[mikeselectricstuff]

I've always used tracing paper, usually Gateway 90gsm as it's easy to get, but if you can get heavier stuff, this is better as it crinkles less. Never had a problem with adhesion, and it doesn't really matter if a big solid area isn't always completely solid.
For double sided I staple it to make an envelope.

[mamalala]

It's also a good idea to check the printer settings for the laser printer. Many printers have things like Eco modes, some resolution "enhancement" stuff, etc. Make sure that all this stuff is turned off. Use the highest resolution setting, and select the darkest setting (i.e. most toner applied during printing) if the printer/driver allows for that.

If the layout is not too fine, you can stack two prints on top of each other. However, if you want to do finer layouts, this will cause problems, since the topmost sheet is some distance away from the photosensitive coating, and will cast shadows.

Generally it's not that important if stuff like ground planes or very thick traces look slightly spotty, at least not with semi-transparent paper. Oh, and i guess you know that already, make sure that the layout is mirrored, that is, the printed side of the sheet is in contact with the PCB.

Greetings,

Chris

[nctnico]

When using tracing paper the exposure time is more critical but in my experience it is not necessary to stack multiple sheets even when large copper areas are not deep black.

[synapsis]

If the layout is not too fine, you can stack two prints on top of each other. However, if you want to do finer layouts, this will cause problems, since the topmost sheet is some distance away from the photosensitive coating, and will cast shadows.

I've found that putting picture frame glass over the stacked prints helps with the shadows. I also print two fiducials on opposite corners to line up the prints.

[tautech]

I also print two fiducials on opposite corners to line up the prints.

Yes, good idea. I a similar idea too, even for single sided PCB's.
I lay a right angle 10 thou track on the bottom layer on the extreme corners to help me position the PCB correctly.

[mamalala]

If the layout is not too fine, you can stack two prints on top of each other. However, if you want to do finer layouts, this will cause problems, since the topmost sheet is some distance away from the photosensitive coating, and will cast shadows.

I've found that putting picture frame glass over the stacked prints helps with the shadows. I also print two fiducials on opposite corners to line up the prints.

There should always be a glass there anyways, and some pressure applied (or a vacuum used) so that the film is in contact with the PCB. That's not the point i was getting at. If you use two films stacked on top of each other, this will cast shadows, unless you use a real point source for the UV light that is really far away. However, pretty much everyone at home uses a small (read: thin) exposure box or similar, where the light sources are rather close to the board.

Take a bog standard exposure box that uses four UV tubes, for example. There you will have only four places where the light goes straight from the tube to the PCB: right above the tubes. Everywhere else the light will hit at an angle. Since the paper/film is not infinitely thin, the second sheet will always be at a distance (the thickness of the first sheet) from the photosensitive coating. This in turn will cast shadows at those places where the light hits at an angle.

You can test this easily by exposing the PCB with a simple pattern of thin lines. For example, 10 mil lines spaced 10 mil apart, running in the same direction as the UV tubes. Using a single film, where the printed side is in contact with the photosensitive coating of the PCB, will produce a quite even result. Stack a second film on top of it, and the result will be less even. The lines will get thicker (and the spacing thinner) in the areas between the tubes, and will be the same widths directly above the tubes.

Oh, and about the glas: be aware that most normal glasses tend to filter UV light to some extent, requiring a longer exposure time. Using quartz glass is preferred, if possible.

Greetings,

Chris

[synapsis]

True about the lights. On my box (around 5" by 9"), I spent a lot of time trying to get the light even over the whole area. I have an upper and lower light for two sided boards, and both lights are 98 UV LEDs (cheap 100 count bags from China) with the plastic scuffed to diffuse the light. Then I use some acrylic sprayed with "frosting" to diffuse it even more, mounted about 2" from the lights.

With this setup, exposure time is around 90 seconds. There was a lot of trial and error in getting the lights diffuse enough. According to my Ocean Optics spectrometer, I get good light until around 1/2" from the edge of the box. Of course it's a bit brighter in the center.

[Fred27]

I have to ask why people are using a laser rather than an inkjet printer for photosensitive boards. (Obviously if you're doing toner transfer that's different.) I get excellent result from a fairly average Canon MG5250 inkjet on cheap transparencies. There is a "T-shirt transfer" setting that lays down a really nice thick layer of ink and even mirrors automatically. There's definitely no need for two transparencies which, as mamalala says, inevitably creates slightly fuzzy edges.

I tried glass and it's OK. It may filter some UV, but still works fine. What I found works really well is Ziploc vacuum food bags to press the PCB and transparency together firmly and evenly. They're cheap too - try them.

If you're using a UV light to expose your boards then you should also consider Dynamask 5000 solder mask. You'll need a laminator and some sodium carbonate to develop it but the results make your board look so much more professional.

[mamalala]

True about the lights. On my box (around 5" by 9"), I spent a lot of time trying to get the light even over the whole area. I have an upper and lower light for two sided boards, and both lights are 98 UV LEDs (cheap 100 count bags from China) with the plastic scuffed to diffuse the light. Then I use some acrylic sprayed with "frosting" to diffuse it even more, mounted about 2" from the lights.

With this setup, exposure time is around 90 seconds. There was a lot of trial and error in getting the lights diffuse enough. According to my Ocean Optics spectrometer, I get good light until around 1/2" from the edge of the box. Of course it's a bit brighter in the center.

Probably not really important here, but be aware that using diffused light, as well as as semi-transparent films, will also cast slight shadows.

Just look at, for example, how the printing plates in offset printing are exposed. The films are transparent. Normally there is large, empty carrier film, on which the different typeset or raster-image films are mounted (using clear spray adhesive). The exposure unit has a high-power UV source, about 1 to 1.5 meters above the plate surface. During exposure, most of the time a semi-transparent foil is placed above the plate+film+glas stack, but on the last minute or so this is retracted.

The reason is simple: it diffuses light, so that you don't get an image of the borders of the cut film snippets onto the plate. Retracting the diffusor towards the end of the exposure time will "sharpen" the wanted image a bit. But this diffusing has to be considered when you want exact reproduction: the thickness of the black parts is usually ever so slightly reduced to accomodate for this, otherwise the image would smear out a bit on the exposed plate.

Sure, for home PCB production this is not that important (dot sizes in offset printing are far smaller than what you wouls use for PCB's). But still, it is something to consider in case you want to use tight spacings between tracks or track/copper-planes. Using light diffusing stuff will make those clearance smaller in the end result if you use photopositive, or larger if you use photonegative (which also means that in photonegative, the actual traces will become thinner).

Greetings,

Chris

[mamalala]

I have to ask why people are using a laser rather than an inkjet printer for photosensitive boards.

From my experience, suitable (semi-) transparent films are more expensive for inkjet printers. Also, quite often you have to wait quite some time to let the ink dry, since these films don't "suck up" the ink that good. Finally, inkjet printing is ridicolously expensive compared to laser printing (but that may not be important for hobby use).

Oh, and there are a few "black" inks that are rather transparent to UV, go figure...

Greetings,

Chris

[Fred27]

Thanks for the info, Chris. Maybe I'm just lucky that I got a box of 50 transparencies on eBay for 99p which work well (an old box of Canon CF-102 if anyone cares). I'll have to write it up in more detail and take photos, but the T-shirt mode makes it look like a thick paste on the transparency compared to the more "dried liquid" look of standard mode.

I have thought about the fact that other colours may indeed be darker to UV than black, but there's more black ink in the cartridge and it works well. Why mess around with what works.

For the few boards that I do, the cost of ink, etc. really doesn't make much difference but I take your point that it could.

[mamalala]

Thanks for the info, Chris. Maybe I'm just lucky that I got a box of 50 transparencies on eBay for 99p which work well (an old box of Canon CF-102 if anyone cares). I'll have to write it up in more detail and take photos, but the T-shirt mode makes it look like a thick paste on the transparency compared to the more "dried liquid" look of standard mode.

I have thought about the fact that other colours may indeed be darker to UV than black, but there's more black ink in the cartridge and it works well. Why mess around with what works.

For the few boards that I do, the cost of ink, etc. really doesn't make much difference but I take your point that it could.

Yea, i guess some special printing mode will cause more ink to be deposited. Like, on the other end of the spectrum on laser printers, all these eco- and "ehancement-" modes cause less toner to be used.

About black, yes, usually it is just fine. It's just that i know of some issues with supposedly black ink that wasn't all that balck when it comes to UV.

Side note: Get some transparent blue tape. You can use that to tape down stuff for exposure. You can also use a blue pen to make notes on the film. Normally that blue (if it is "pure", i.e. a single colour/pigment) is transparent to UV, so it won't show up in the exposed image. (Note: blue as printed by inkjets is not pure blue, but a mixture of cyan and magenta, one of which may be opaque to UV).

Oh, try printing a bunch of datasheets in b/w on a laser and on an inkjet. That is where you will pretty quickly see the price difference. Not to mention  the speed...

Greetings,

Chris

[sleemanj]

Probably not really important here, but be aware that using diffused light, as well as as semi-transparent films, will also cast slight shadows.

In practice, for hobby purposes, not an issue imho

I have to use a diffuser in my current setup, I was worried about collimation but found I can still do 8/8

If I needed less than that then that's not what I'd call hobby

Sent from my HTC Dream using Tapatalk 2

[sleemanj]

I have to ask why people are using a laser rather than an inkjet printer for photosensitive boards. (Obviously

Because inkjet printers are the work of the devil, messy, expensive, unreliable, slow...

Inks also have varying transmissivity of uv, so more things to tune in.

A laser can be bought used for a couple bucks, or free, 1200 dpi is sufficient for fine work.
 
Sent from my HTC Dream using Tapatalk 2

[Fred27]

I see more problems with lasers than inkjets  of you look somewhere PCB specific like the HomebrewPCB Google group. And if you've ever spilled toner, you'll know what hell that can be! As with all of us I'm basing my opinions on just what I've observed rather than any scientific testing.

Wax printers like many modern Xerox ones are really nice. In theory they could print straight to PCB, but I'm not sure anyone had done it. As an aside, where I work had the hardest working Xerox laser that they ever made - according to their engineer who we saw regularly. After 5 years and so many replacement parts that it wasn't really the same printer, it's been changed for a wax one. Far less messy and looks reliable so far. It's a very large and expensive machine and I've been told if I even *think* about putting some copper clad on the document feeder I'll be fired. ;-)

I agree that with a single transparency and the ink/toner on the PCB side, collimation isn't really a problem for home use.