DIY problems with an enormous 508mm board

[ampdoctor]

I've got a question similar to the one asked several days ago regarding large copper boards and maybe any discussion here will help out that poster as well. 

I've got a design I need to prototype that comes out to 508x108 mm.  Clearly this is enormous by DIY standards. Due to mechanical constraints I can't bust it up into smaller boards so I'm stuck with this size.  I looked around for places that would do one off's and the price is sky high until you get into larger quantities. The price point becomes problematic for a design that I'm still not sure won't need tweaking or adjustments in either the circuit itself or the board layout. So unless I want to hand over a lung or my first born I'm somewhat forced into figuring out a way to prototype this thing myself. 

So here's the question. Has anybody ever tried to home brew a double sided board this large and if so how did you do it?

Obviously it's larger than a single sheet of A4 or any other size of commonly available paper. So I'm thinking I'd have to come up with some sort of method that won't blow the registration between layers out the window. I've got a few thoughts on how I might approach this using either toner transfer or uv exposure methods.

Using toner transfers I might be able to carefully cut the sheets and butt them up against each other. This seems a bit impractical given that should I get the indexing pins wrong along with the typical paper shrinkage it's a disaster waiting to happen.  There would most likely also be a lot of rework with a sharpie after the fact as well. 

Another option I've thought about is printing on clear vinyl with several overlays that fractionally overlap so I wind up with a single sheet that's large enough, then expose using sunlight instead of a light box. I'm also batting around the idea of using a UV emulsion coating that gets used with silkscreens where the unexposed areas can be washed off with tap water.

I'm in some uncharted waters here, so any comments, ideas, or suggestions would be greatly appreciated.

[free_electron]

Iteadstudio will do it.
I've done really largo boards with them. They have a category for it

[sleemanj]

Find some A2 size tracing paper of a good grade (110gsm is what I use), art shops do sell it. A2 is 590 something mm long.
Go to a printery with a large format printer that can handle A2, shouldn't be too hard (although try and find a laser rather than an inkjet).
Get artwork printed onto the paper.
Dry film laminated on to your PCB, couple panes of glass, couple clamps, sunlight exposure.  You might have to do a couple of test runs first to get the timing right.
From then on, standard developing and etching.

[ampdoctor]

Iteadstudio will do it.
I've done really largo boards with them. They have a category for it

I checked out iteastudio, and unless I'm missing something somewhere the only thing I was able to find was a category for boards larger than 30cm. For a board this size their schedule gives me a cost of about $150/board with a minimum order of 5 units...unless I'm misreading something. On a board that may need a respin 750 bucks is pretty pricey. If you're a corporation yea sure no big deal. Just write it off as R&D costs, but for the average Joe....OUCH!

Find some A2 size tracing paper of a good grade (110gsm is what I use), art shops do sell it. A2 is 590 something mm long.
Go to a printery with a large format printer that can handle A2, shouldn't be too hard (although try and find a laser rather than an inkjet).
Get artwork printed onto the paper.
Dry film laminated on to your PCB, couple panes of glass, couple clamps, sunlight exposure.  You might have to do a couple of test runs first to get the timing right.
From then on, standard developing and etching.

I've never really thought about tracing paper as a viable solution.  At first blush I would think that you would have a tough time getting a sharp image due to the nature of the medium and the way it diffuses light somewhat.  Since you offered it as a suggestion is it safe to assume you've gone this route before. And if so how well does it work in general?

[sleemanj]

I've never really thought about tracing paper as a viable solution.  At first blush I would think that you would have a tough time getting a sharp image due to the nature of the medium and the way it diffuses light somewhat.  Since you offered it as a suggestion is it safe to assume you've gone this route before. And if so how well does it work in general?

Tracing paper works just fine as a medium for photo artwork, 12/12 is my standard parameters but I etch down to 10/10 in a pinch, and could do 8/8 if I really really really really wanted to.

Of course it's not as transparent as, well, transparent stuff, but it's pretty good.  My standard exposure (dry film) is currently 53 seconds under UV leds, and if under sun, well about the same really, you just have to experiment a bit.

I use a single print, some people double it up and do a longer exposure, but I find it's better to just have one print and a shorter exposure myself. 

After developing the dry film, I re-expose for about 5 minutes to fully harden off the traces.

[BravoV]

After developing the dry film, I re-expose for about 5 minutes to fully harden off the traces.

With the mask on ? or without ?

[sleemanj]

With the mask on ? or without ?

Without.  Note that I'm using dry-film here which is negative acting, naturally if you're using a positive acting resist, that wouldn't work so well :-)

My initial exposure with mask is quite short, then I develop it, and then re-expose the developed board to UV which causes the traces to darken considerably.  If you use a longer exposure first, the developing can be more difficult and bridging easier, a short exposure makes developing easier but the traces may be less secure, so re-exposing hardens them up.  This is a common recommendation from Dry Film manufacturers, also I've seen recommendation (in a riston document from memory) to oven bake the developed board as an alternative to the second exposure.

[ampdoctor]

Nice! I may have to give that a shot just on general principles.

And the more i look at silk screen emulsion the more interesting it gets. The cost is a fraction of dry film and you don't need to mess with developer, just wash away with water.  The resolution from what I've read will allow for fine-line artwork and text on the order of 8pt or possibly smaller. Don't know if you can get 8/8 out of it but it should be acceptable for some pretty fine trace widths and spacings. Might even be suitable as a makeshift solder mask as well.  the real wild card is how well it adheres to bare copper. Guess there's only one way to find out.

[Psi]

I seem to remember Mike from mikeselectricstuff just made some very long boards in a recent video. Maybe he knows where to get them at a good price.

[BravoV]

With the mask on ? or without ?

Without.  Note that I'm using dry-film here which is negative acting, naturally if you're using a positive acting resist, that wouldn't work so well :-)

My initial exposure with mask is quite short, then I develop it, and then re-expose the developed board to UV which causes the traces to darken considerably.  If you use a longer exposure first, the developing can be more difficult and bridging easier, a short exposure makes developing easier but the traces may be less secure, so re-exposing hardens them up.  This is a common recommendation from Dry Film manufacturers, also I've seen recommendation (in a riston document from memory) to oven bake the developed board as an alternative to the second exposure.

Thanks, I just knew this technique that after its developed, the re-exposure will not screw the unmasked region. 

Btw, I'm using the negative acting blue colored film type and no idea what brand it is, whats yours ?

[Codemonkey]

I'd just like to add that at my previous job, I used to use laser print on opaque draughting film for prototype boards that I etched myself. I found that the toner stuck much better to it than the clear over head projection type film and as a result got much better results.

[sleemanj]

Btw, I'm using the negative acting blue colored film type and no idea what brand it is, whats yours ?

"Whatever the chinese vendor sent this time brand" :-)

Depending on who I get it from, generally it's one of two kinds, one appears blue unexposed, and the other a much lighter blue, almost turquoise, unexposed.  When fully exposed they are both dark blue.  There's no practical difference in the two for hobby purposes, they appear to expose at about the same rate, develop the same, perform the same under etching....

[BravoV]

LOL ... great to hear that ! Definitely will try your re-exposure technique next time, thank you. 

[sleemanj]

And the more i look at silk screen emulsion the more interesting it gets.

You can get liquid photo resist paint (negative acting) (just the first one I found on ebay, no idea what the seller is like), it's basically the same as the UV Soldermask ink, except they thin it with "banana oil", aka isoamyl acetate or amyl acetate.

Amyl acetate might be tricky to get, apparently ethyl acetate works the same and is the active component in acetone-free nail polish removers, so probably easier to come by [ bonus points for somebody who tries just using  ethyl acetate nail polish remover to dilute it and see if it works with the various contaminants ]. 

I'm not sure if Kerry Koppert is a member of the EEVBlog, I hope he doesn't mind me copying his email from the Homebrew PCBs mailing list about his experiments with this liquid photoresist...

I tried to source some
banana oil (Amyl Acetate) in New Zealand to no avail (I used to buy it
as a kid 50 -60 years ago for model aeroplanes) but I tracked down some
Pentyl Acetate but the supplier was out of stock. He suggested Ethyl
Acetate, N-Butyl Acetate or ISO Propyl Alcohol so I got some Ethyl
Acetate. It doesn't smell as nice as Banana Oil but it works. Next I
tried painting a board with a brush. It sorta worked but the covering
was uneven so I pulled out my El Cheapo Air Brush and tried that. Hey it
works and give a pretty impressive result.

I think develop in sodium carbonate or perhaps weak hydroxide,  pretty much the same as for the dry films.

[EEVblog]

Will it fit onto an A3 diagonally? If so you can use a photocopier onto an A3 transparency.

[EEVblog]

The problem is that 508mm is larger than your usual 18" (457mm) panel size, so most manufacturers will charge a premium for it.
As with the A3, can you rotate the board diagonally to fit onto an 18" panel? If necessary, do this on your PCB file itself and define routing around the board (or cut it out yourself).

[EEVblog]

http://project-pcb.com/
will let you enter 508x108mm, and it's $167 + postage for 1, $216 + postage for 5.

PCBcart are the same price and also accept that dimension

[EEVblog]

Just for kicks I entered 5000mm into PCBcart and it still accepted it!
But their max dimension is 600x700mm
http://www.pcbcart.com/cart.php?page=capability___standard_pcb
So no problem.

[Dave Turner]

For those trying tracing paper for the first time note that there are two sides to tracing paper one of which shinier than the other. This is the side that the design should be reverse printed on and laid face to face with the resist. This results in slightly less diffusion than any other combination thus giving a crisper edge to the traces.

Whilst I said reverse printed I guess that this is the norm for the printed output of the bottom layer from most PCB layout software.

[BravoV]

it's basically the same as the UV Soldermask ink, except they thin it with "banana oil", aka isoamyl acetate or amyl acetate.

Amyl acetate might be tricky to get, apparently ethyl acetate works the same and is the active component in acetone-free nail polish removers, so probably easier to come by [ bonus points for somebody who tries just using  ethyl acetate nail polish remover to dilute it and see if it works with the various contaminants ]. 

Aha .. Banana Oil, I think I saw it at our local industrial chemical store.