Conductive Ink

Thanks for this. Do you recommend glass because the solvents would attack the barrel of a plastic syringe? It seems McMaster-Carr only has plastic, and don't say which flavor of plastic, but the ones I've seen seem to be polyethylene, which AFAIK is resistant to most common solvents.

Am I blowing smoke up my own ears here? ;-)

Thanks, Rich

Well, kinda, yeah, but because of the parameters of the project, I'm looking for a technique where I can do a layout, test it, fix the layout, test it again, and so on until I get it right. Having a board made, or even etching it myself, would stretch a few days' work out into weeks or months.

And the thought of etching it myself makes me want to run away screaming. :-)

Thanks, Rich

Actually, if it were to come to this (rather than just send the finalized layout to a board house), I already have some paper in mind; it's the stuff that was recommended by Canon for color when I bought my BJC-210 (clay coated), assuming somebody is still making it. :-)

Thanks! Rich

Well, I don't trust simulators; there are too many other variables in real life. My background is "only a tech", so I'm used to slapping together a prototype, testing it, rearranging parts, testing it, and so on until I get what I want - of course, I'd design (well, the client and I are going to collaborate on the basic design) out part variances, like the way they design circuits where beta isn't the critical parameter, and so on.

But my little voice is telling me that the physical layout of this thing would play at least as important a part as component selection, circuit topology, and so on.

I guess, if I got ahold of a batch of cheap boards, I could do it with resist pen, etch (YECCH!), test, and so on; I just want to minimize the turnaround time from one design/layout to another.

The lazy man's way is usually better - it leads to much more efficient use of resources. :-)

Thanks, Rich

Actually, this stuff:

claims that you can solder to it after it dries, albeit it has to be cool and quick. I'd need to find some low-melt solder.... Nah, the stuff says it can take 350F for 5 seconds; 63/37 melts at 365F, which might be a little close to the edge, but I've been soldering on and off since most of the folks around here were still a gleam in Mom's eye, so I'm confident I can figure it out. And we all know how thirsty silver is for liquid solder. :-)

OOh! Gotta remember to pick up some liquid flux!

...

Oh you're more than welcome! In fact, though, some of the respondents have recommended against the type that lays down the fat blobs, but someone[1] recommended just pouring it into a syringe. I think this decision will be easy. :-)

Thanks, Rich [1] (sorry for not citing you, whoever it was.)

Rich Grise hath wroth:

Sorta. Glass usually works with most everything. Glass is also much easier to clean and the replaceable tips are a must. I have an ever shrinking assortment I obtained from medical equipment supplier about

20 years ago, when it was much easier (and cheaper). No clue where to get them today.

Plastic syringes come in many different compositions. Some are resistant to solvents, others are not. However, the usual failure is NOT the plastic barrel. It's the rubber seal at the business end of the plunger that gets eaten. I picked up a box of what was apparently plastic veterinary vaccination syringes that fail miserably with anything that attacks the rubber plug, which seems to be everything I ever wanted to put in a syringe instead of the insipid non-clogging patented contrivances that always clog and die.

Actually, some of the fancy ones are polycarbonate and even look like glass. However, it's not the plastic barrel or plunger that's the problem. It's the soft rubber seal. With glass, you usually have a precision ground plunger and clear barrel. The seal is good enough that it doesn't require the rubber thing.

Looking at one syringe in my pile, it's inscribed: Tomac Star Interfit, unground barrel, Made in U.S.A. for American Hospital Supply, Div of A.H.S.C., Evanston, IL. It's probably 20 years old. I couldn't find it with Google. I'll ask at the local hospital if there are any industrial suppliers (non-sharp non-hypodermic needles) and what to ask for. That may be a problem as I've already been accused of being a nuclear terrorist and drug dealer when I ask such questions.

Cigarettes are bad for you, whether you inhale through your mouth or stick them in your ears. See safety instructions on the package.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

Using simulators adds an artistic layer to electronics design. It's the art of working with behavior that's missing. With real circuits, it's the art of working with complete behavior + surprises :) Both fall into the art of working with imperfection.

When I use pcb software for prototypes, I make boards with any predicted changes. Extra footprints, extra copper that can be dremeled to shape, cutting points, parallel chip resistor footprints for tweeking, test points, part swapping pads, jumper pads, components spaced out for easy assembly, test and swapping...etc.

One time I made a pcb that had 5 circuit variations to do the same function..Let the best one win..

If I had to dodge etching, I'd rather use adhesive copper tape than conductive ink.

D from BC

What do you mean SMT? SOIC-8? Or some sort of QFN? CSP? Specify. I've done some work with just a one-sided copper clad board, I sliced it up with a grinding wheel for ground/power areas, and just tied the signals together with 30 gauge wire. This was for a 32MHz VCXO.

Oh, I used Surfboards from Digikey too.

Of course, they're a whole 2 dollars. :)

For more RF-style stuff, I just engrave strips into copper clad boards.

You could also use silver conductive epoxy. It's quite good, we've pushed 3GBps signals in a test splice with the stuff.

I've found that *software* and *other-than-drawn-by-hand techniques* make doing things a second time (with corrections) MUCH easier.

As D from BC mentioned, allowing the ECAD software to do automated checks can make life less stressful.

Hand-drawn == Less precise (and SLOW if you have to do it twice) I've already mentioned error checking.

If the thought of Ferric Chloride is bothering you, there are transparent, non-staining alternatives:

"PCB FAQ" by Vincent Himpe 1995 "Persulfate" marked at the 75% mark. Lots of other good stuff there.

*-*+Vincent.Himpe+Sodium.Persulfate+re.crystallize+21.Jun.1995+needs.heating+ammonium.persulfate+Choose-*-*+Turns.dark.blue+*-*-cools-down-*-*-*+sci.electronics+transparent.solution&strip=1

**DIY PCB Fab** by Tom Gootee 2003, 2007 Etchants ("Muratic", "Persulfate"); "glossy"; and lots of (cheapie) techniques. *-*+Hydrochloric.Acid+*-*-stain-*-*+Glossy+brush+*-*-*-transparent+Persulfate&strip=1

Techniks' "Press'n'Peel" at $2 a sheet is a worth consideration.

I'll bet I can match you for sloth--but several sayings spring to mind: "Penny-wise and pound-foolish." "If you don't have the time to do it RIGHT, how will you ever find the time to do it OVER?"

You haven't mentioned the complexity nor the pin spacing--but no matter; all told, I found ECAD and etching to be MUCH faster and FAR better than whatever is in 2nd place.

50 years of empiricals back me up.

thinking

[trim]

project,

fix

Having

days'

Hi Rich, that was exactly my frame of mind too .....

i **dreaded** the idea of etching a board !

but i was quite pleasantly surprised at the ease of etching (especially using the somewhat less toxic {muriatic acid + drugstore peroxide}) etchant

i could only find a PLCC replacement for my DIP40 IC and the only **clean** way i knew was to make some adapter PCB

now that i read about "dead bug" i came up with some hybrid ideas maybe you already considered or not

Ideas (with "dead bug" - ishness) Create a PCB mask with lots of SMT component pads (only the pads) in the position you like , then etch that board of pads, then put your components down on the pads and use your conductive pen to draw traces between the pads on the board where you like (i think you can erase with acetone)

-or- some variation on this theme that gives you the components nailed down and flexibility to wire in various ways like putting several solder pads next to SMT components to put wires or jumper pins or (well you know more about the various connector stuff out there than I do)

any way some ideas, hth, robb

stuff:

Rich, some more details: I tried a silver-conductive ink pen (Circuit Works, # CW2200 - MTP, micro tip), available at Fry's. The traces were to too fat. The smallest I could make was roughly 0.040", which means that a wire intended to connect to one lead on a SOT-23-6 transistor would often glob onto and also connect to the adjacent pin, shorting the two pins.

To make anything finer I had to lay down a fattie trace, then scratch out any excess with a needle. That was hard work, and traces often chipped or tore in the process.

On the plus side, the stuff is amazingly conductive.

If you still wanna try, auto parts places also sell that conductive ink used to repair those printed-on-the-glass rear-window defroster elements...

Cheers, James Arthur

(That'd be Jeff Liebermann ;-)

On Mon, 15 Oct 2007 10:54:09 -0700, D from BC wrote: ...

OK, where can I get some? McMaster-Carr, Digi-key, and Mouser all have it as narrow as 1/4" - I'd need something more like .06 (1/16) or so.

Or do you buy the wide stuff and slit it? =:-O

Man, I miss the days when you could walk into the hobbyist store and buy anything you could think of.

Thanks, Rich

[about easy breadboarding]

"Many hands make light work." "Too many cooks spoil the soup."

"Look before you leap." "He who hesitates is lost."

etc, etc. ;-)

And of course, "slow and steady wins the race", assuming the rabbit stops for a nap on the way. :-)

It won't be very complex; as far as parts selection, it's TBD; I'm just trying to look ahead.

Thanks, Rich

I got stuck too with the same problem.. I ended up doing time consuming trimming. I bought a nice roll of copper tape from a glass artwork shop for $4.00.. D from BC

Jeff Liebermann snipped-for-privacy@cruzio.com posted to sci.electronics.design:

The google search string you want is: dispensing syringe. They have been used for soldering pastes and adhesives for at last 20 years now.

JeffM jeffm snipped-for-privacy@email.com posted to sci.electronics.design:

*-*+Vincent.Himpe+Sodium.Persulfate+re.crystallize+21.Jun.1995+needs.heating+ammonium.persulfate+Choose-*-*+Turns.dark.blue+*-*-cools-down-*-*-*+sci.electronics+transparent.solution&strip=1

*-*+Hydrochloric.Acid+*-*-stain-*-*+Glossy+brush+*-*-*-transparent+Persulfate&strip=1

Either you started at an amazingly young age or you are at least a little bit older than me. (i can only claim 4 honest decades and i started young)

Just an idea that came up from the intersection of a couple of things, and which may not work...

IF you can work at the size of a credit card, or only slightly larger: [83.9x51mm - 85.6x54mm - 92x60mm - 98.5x67mm ]

"ID card" or "Badge" printers are typically thermal dye sub or resin printers set up to print onto a rigid card substrate. The "Resin" material would probably make an effective resist - you just get the cheaper ribbon that prints nothing but black. You can get printers that will double-side, and they are typically 300DPI.

Probable problems with the concept: They are also designed to print onto a very smooth plastic card, and I don't know if they would manage an effective job of printing onto copper - not quite as smooth, and a far more effective heat-sink. Some do have driver settings for cranking up the heat, but I don't know how far that would go. The cards are typically only 0.030 inch thick, I don't know how thick the printers would go - it probably varies by printer.

But if worked, spit a card out in 30-60 seconds and drop it in etch.

Of course, you'd have to precut the PCB material to card size. But it seems a lot more (potentially) practical than the discussion earlier this year of trying to get a typical laser printer to print direct to copper.

--
Cats, coffee, chocolate...vices to live by

Thanks! Rich

JosephKK wrote:

The 50 years of empirical are not all my own. I am willing to learn from the missteps and successes of others. To clarify: Going back to the adoption of PCBs and moving forward, I see the *etch away the copper you don't want* paradigm widely used. I don't see the *lay down individual conductors using a fluid* notion. The empirical evidence is that those who have tried the later have quickly abandoned it.