useful for a circuit board using scissors to cut out and paste the connections in any layout or configuration I might wish or dream up

If that's your aim, and the 'silver' mylar turns out to be aluminised mylar, then you might like to consider instead using adhesive-backed copper foil. I recently bought a 2 metre length of 23mm wide foil, sold as a slug deterrent, from my local pound-shop.

 

What is a "pound-shop"? That's not a term we use on this side of the pond and I can think of a couple things it might equate to.

 

That should be obvious. A pound-shop is a dollar-store.

 

I could be wrong, but I don't remember noticing silvery stuff in membrane keyboards I think it might be some kind of conductive polymer, and not a very good conductor at that. The shiny mylar used for helium balloons and gift wrap I believe has aluminum deposited on it. I'll delete this post if someone knows for sure.

If it goes black in long term exposure to air - its silver.

Silver sulphide etc - breaking wind after eating eggs can do it.

 

There has been a bit of interest in using a modified Tollens reagent to print electronic quality and quantity of silver on all matter of substrates. Here is the original description: http://pubs.acs.org/doi/abs/10.1021/ja209267c

Prof. Lewis is the lead author. If you search on her you will find additional links, including to a startup that is using the ink to print inside devices made with 3D printers. I wrote a somewhat more extended abstract of the article (attached).

John

Thanks for the information. I'll check the links out soon!

 

@Rolland B. Heiss

If you wish to produce your own printed circuit boards please be advised that copper-clad PCB 'blanks' and ferric chloride etchant are inexpensive, abundant and available (even) in small quantities -- Instead of 'resist ink', however, you are best advised to use 'spirit' (i.e. toluene, turpentine, ketone, etc...) 'based' model paint applied with a 'detailing' brush -- thus you may successfully produce boards featuring very fine traces...

Best regards
HP

 

@Rolland B. Heiss

If you wish to produce your own printed circuit boards please be advised that copper-clad PCB 'blanks' and ferric chloride etchant are inexpensive, abundant and available (even) in small quantities -- Instead of 'resist ink', however, you are best advised to use 'spirit' (i.e. toluene, turpentine, ketone, etc...) 'based' model paint applied with a 'detailing' brush -- thus you may successfully produce boards featuring very fine traces...

Best regards
HP

Apparently the ink in Pentel marking pens makes a passable etch resist.

But you pretty much don't get nice edges on your PCB tracks.

 

Ah, okay. That does make sense. I was

@Rolland B. Heiss

If you wish to produce your own printed circuit boards please be advised that copper-clad PCB 'blanks' and ferric chloride etchant are inexpensive, abundant and available (even) in small quantities -- Instead of 'resist ink', however, you are best advised to use 'spirit' (i.e. toluene, turpentine, ketone, etc...) 'based' model paint applied with a 'detailing' brush -- thus you may successfully produce boards featuring very fine traces...

Best regards
HP

I know that I could never get very fine traces using that method, but that reflects my limitations with a paintbrush, no matter how fine. But even if I could, it's hard to imagine the time it would take to be worth it. My thinking is that the fine lines seldom benefit you unless you have to cram a bunch of stuff into a small area ('seldom', not 'never'), but the time it would take to apply the resist by hand to that much stuff in that small an area -- and that's assuming that you make NO mistakes along the way -- would drive most people batty (I know it would me).

 

I know that I could never get very fine traces using that method, but that reflects my limitations with a paintbrush, no matter how fine. But even if I could, it's hard to imagine the time it would take to be worth it. My thinking is that the fine lines seldom benefit you unless you have to cram a bunch of stuff into a small area ('seldom', not 'never'), but the time it would take to apply the resist by hand to that much stuff in that small an area -- and that's assuming that you make NO mistakes along the way -- would drive most people batty (I know it would me).

In addition to all around superior results, I found the paint-on technique nice for bringing traces out between DIP pins, etc... (very handy indeed for ye-olde [48 Pin- DIP] Z8000 [segment addressed] CPU and its attendant 'cadres' of MMUs, PIOs, etc...) ---- My 'disenchantment' with 'resist ink' owes to a limitation whereby masking much narrower than 500 uM tends to result in granular (or altogether missing) traces (due, presumably, to the contrariant relationship of satisfactory coat thickness to narrow trace width) --- Note, however, that a good artist's detailing brush is essential! (I recommend sable).

A word about photo-resist techniques --- If you've the breath for it - Have at it! -- For my part it's far too much folderol for negligible 'return' -- unless, of course, the intent is production multiple boards...

Best regards
HP

 

I have a friend that does a toner transfer process and routinely does 10 mil width/space and has done 6 mil with good results. He says that things start going south at 4 mil.

What kind of resolution (width/space) do you achieve with hand-painted lines?

 

In addition to all around superior results, I found the paint-on technique nice for bringing traces out between DIP pins, etc...

A word about photo-resist techniques --- If you've the breath for it - Have at it! -- For my part it's far too much folderol for negligible 'return'

I simply disagree with you...probably because my eyes don't work that well at 60+ years old. I couldn't possibly produce good 10 mil traces and 10 mil spacing with my physical assets!

 

I have a friend that does a toner transfer process and routinely does 10 mil width/space and has done 6 mil with good results. He says that things start going south at 4 mil.

What kind of resolution (width/space) do you achieve with hand-painted lines?

Minimum width = 250um (i.e. ~10 mil), minimum spacing = 300um (i.e. ~12 mil) --- Again, the foregoing is a statement of minimum limits -- a few hours of that leaves me myopic for the balance of the day!

I simply disagree with you...probably because my eyes don't work that well at 60+ years old. I couldn't possibly produce good 10 mil traces and 10 mil spacing with my physical assets!

A 'mag lamp' might help that! --- Also 'decaff' and 'T-Total' abstinence (for several days prior) are obligatory!

Please don't get me wrong, it is not my intent to suggest that this method is easy, merely that it is preferable to 'resist ink pens', photo-resist techniques (Re: 'one-offs') and, especially, manual application of conductive material...

Best regards
HP

 

and that's assuming that you make NO mistakes along the way

It was once my misfortune to discover that Ca. 100 hours painstaking work was wasted upon a (hither-to unreckoned) mirror image!!! --- Preservation of my radial arteries required an "out of sight - out of mind" policy (Re: the X-Acto tools) that afternoon!

Best regards
HP

 

I seem to have misunderstood. I thought you were advocating manual application of conductive material.
Apologies.

 

I seem to have misunderstood. I thought you were advocating manual application of conductive material.
Apologies.

No, manual application of 'resist' paint (Specifically model paint)...

The trouble with conductive applications is that they are either objectionably resistive, friable, or impossible to apply precisely (and, frequently, all of the above)...

Best regards
HP

 

Thank you for clarification. Still outside my limits of manual skill.

 

Thank you for clarification. Still outside my limits of manual skill.

In case you are interested:

Succinctly, a copper clad board (fashioned of dielectric material) is masked (by any of various methods) with a with a pattern of 'resist' compound -- Thence immersed in an etchant wherein all unmasked copper is chemically removed from the dielectric substrate --- The end product being a PCB having copper traces corresponding to the resist pattern...

Best regards
HP

 

In case you are interested:

Succinctly, a copper clad board (fashioned of dielectric material) is masked (by any of various methods) with a with a pattern of 'resist' compound -- Thence immersed in an etchant wherein all unmasked copper is chemically removed from the dielectric substrate --- The end product being a PCB having copper traces corresponding to the resist pattern...

Best regards
HP

Been there, done that, shipped the product, cashed the check...but I did NOT manually draw the conductors OR the spaces between them!

 

It was once my misfortune to discover that Ca. 100 hours painstaking work was wasted upon a (hither-to unreckoned) mirror image!!! --- Preservation of my radial arteries required an "out of sight - out of mind" policy (Re: the X-Acto tools) that afternoon!

Best regards
HP

I have a hard time justifying 100 hours of work just to lay down the resist artwork on a board. Even if my time is only worth $0.50/hr, I could probably have the board professionally made by a bare-bones proto-house and have it in my hands well before I got done laying down the artwork.

Which is why I prefer photographic approach. I haven't made a board for many years, but when I did I simply printed the artwork on a laser printer to an overhead transparency, taped it to a piece of thin plexiglass, and then clamped that onto a presensitized board. It was trivially easy to do complex, double-sided boards. I would tape the two transparencies to two pieces of plexiglass and put the two together (without a board) to align them and drill two holes in opposite corners (usually in the positions for the eventual mounting holes for the board). Then I would put in the board on one of them and drill the holes in it. The holes were just big enough for a small standoff to slip into so I got very good registration between the sides. I routinely did 10/10 boards using that. I did one 5/5 board and had decent results, but at that level the degree to which plexiglass is UV opaque started causing problems with bleed-through of the toner due to the long exposure times. I had similar effects on lower resolution boards, of course, but with a wider track it doesn't cause significant impedance increases and/or opens.

Plus, if I needed to make a second board for any reason, including my screwing up the first board at some point (or, in one case, because I taped the overlays upside down so that I had a mirror image problem similar to you), I just make a new board.

 

I have a hard time justifying 100 hours of work just to lay down the resist artwork on a board. Even if my time is only worth $0.50/hr, I could probably have the board professionally made by a bare-bones proto-house and have it in my hands well before I got done laying down the artwork.

My projects were wholly avocational (and, hence, not time critical) then too 'twas the early 1990's when services such as you describe were both scarce and rather difficult to 'find'...

I haven't made a board for many years, but when I did I simply printed the artwork on a laser printer to an overhead transparency, taped it to a piece of thin plexiglass, and then clamped that onto a presensitized board. It was trivially easy to do complex, double-sided boards. I would tape the two transparencies to two pieces of plexiglass and put the two together (without a board) to align them and drill two holes in opposite corners (usually in the positions for the eventual mounting holes for the board). Then I would put in the board on one of them and drill the holes in it. The holes were just big enough for a small standoff to slip into so I got very good registration between the sides.

Inasmuch as my DIY PCBs were of original (i.e. my own) design, I was obliged to do the artwork anyway, so... might as well do it only once --- Moreover CAD software is (or was) insufficiently flexible Re: layout -- For instance when I wished to employ 'strip-line' (i.e. printed transmission line) techniques to my (225 MHz ECL) interferometer's "front end" (spec. data acquisition) card, the CAD software insisted upon 'optimizing' (i.e. mutilating) my layout for 'trace length economy' -- while said behavior could be over-ridden, it was hardly worth the effort --- All in all it was just easier to take up brush and paint and 'get straight down to it,' as it were...

the degree to which plexiglass is UV opaque started causing problems

That's interesting! So acrilic exhibits a significant degree of (UVB?) radiopacity -- very good to know for future reference! --- As a point of interest, polycarbonate possesses similar properties all the way up to 30 kev (at least) -- as evinced by the significant 'solarization' observable in (even 'new-ish') compression paddles (as well as the slight 'haze' apparent in 'raw' radiographs)... A study of the radiolucency of various polymers would be a rather useful resource indeed...

Plus, if I needed to make a second board for any reason, including my screwing up the first board at some point (or, in one case, because I taped the overlays upside down so that I had a mirror image problem similar to you), I just make a new board.

There's no disputing the point that your approach is, far and away, best! --- Should I ever again find the need to produce a complex PCB I'll doubtless go the 'photo-resist' route --- That said, for simple layouts, the 'paint & brush' approach just seems like 'getting it done' -- Indolence? Masochism? Obduration? Probably all of the above -- 'Tho, in my own defense, I'm bound to say "if it works, it works"!

Best regards
HP