[dorkbotpdx-blabber] perfboard!

Tue May 19 23:46:42 EDT 2009

thanks, paul, that's a lot of good advice!  i wish you'd do a video
of some of your wire techniques, 'cause i'm having a hard time
visualizing them.

do you think that teflon wire is available anywhere in town?

-m-

On May 19, 2009, at 2:19 PMTuesday, Paul Stoffregen wrote:

> Wire wrapping is easier if you have problems soldering, but  
> solderless breadboards are by far the easiest way.
>
> Unfortunately, wire wrap sockets are pretty expensive, since they're  
> made with machined pins instead of stamped and folded metal like low- 
> cost sockets.  Wire wrapping doesn't work as well when you wrap on a  
> round component leg, like a resistor or capacitor.  It's really  
> designed for the square posts on those special sockets.  The 4 sharp  
> corners really bite the tightly wrapped wire and make it strong.
>
> I personally prefer soldering with the pad-per-hole perf board.  The  
> boards I use have a ground plane on the top side, which is nice  
> since you don't have to route a ground wire all over the board.  You  
> get a nice low impedance ground everywhere, which helps if you're  
> using any high speed digital chips.  The downside to the ground  
> plane is you have to be careful with parts like crystals or certain  
> terminal blocks with exposed metal on the bottom side, but usually  
> it saves a lot of time because ground needs to connect to just about  
> everything.  Everywhere where a ground connection is needed, I just  
> fold a piece of bare wire over needle nose pliers and stick it down  
> 2 holes and bend together on the other side, and solder both  
> sides.   A larger tip on the soldering iron or higher temperature  
> setting really helps when soldering to the ground plane.  I do all  
> those first when I build something.
>
> Originally I was using some pretty spendy material from Vector.  It  
> was so expensive that years ago I just had 100 boards fabricated by  
> Sunstone.  At the time their prices were much lower and it worked  
> out to be a fraction of the cost of buying from Vector, and I get  
> the nice pads and ground plane feature!  My boards are 5 by 3 inches  
> and rarely do I use a whole one.  I usually chop them into just-the- 
> right-size pieces using tin shears.  I still have plenty of extras,  
> if anyone is interested.
>
> It's a boring way to start wiring, but I always make a point to wire  
> all the power connections first (well, after making all those ground  
> contact points).  Usually I use bare wire, often left over component  
> legs, which are strong tin plated steel wire that bends sharply  
> compared to soft copper wire.  Because ground is routed on the top  
> and bare wiring running all around for the power makes adding more  
> connections easier.  If there's more than one power supply line,  
> some planning is needed, but the common case of a single power line  
> is very easy because ground never gets in the way of routing it all  
> over with just bare wire.
>
> Then I wire up all the mundane signals... the ones that have no  
> question about where they need to go.  Things like crystal and  
> capacitors, 0.1uF decoupling caps, required pull resistors, and so  
> on.  Again, it's kind of a boring way to start because I'm always  
> thinking about the more interesting stuff.  But doing the boring  
> things first means they end up on the bottom and the wires that  
> later get changed around are on top.
>
> 0805 size surface mount parts work great with 0.1 inch pad-per-hole  
> perf board.  They fit perfectly between 2 pads, and if the pins of a  
> socket aren't soldered, you can often press one between 2 pins.   
> Most of the stuff I build this way has all the resistors and  
> capacitors on the bottom side.  It's also very easy to add an extra  
> resistor or capacitor that you didn't realize you needed at first,  
> because there's usually some free pads nearby, even if you've packed  
> a bunch of sockets/chips/whatever pretty closely spaced.
>
> For other surface mount parts, you really need adaptor boards to 0.1  
> inch headers.  Years ago, I made a big panel with almost every type  
> (also back with Sunstone had reasonable pricing), which ended up  
> costing what I would have spent for only a half dozen of the normal  
> ones.
>
> The other trick I use is Teflon insulated wire.  It's more  
> expensive, but well worth the money.  The insulation doesn't melt.   
> It doesn't shrink either.  Before someone told me about this stuff,  
> I was always having problems where I'd solder a wire and it would  
> accidentally touch another nearby wire underneath and melt though  
> both layers of insulation and short the two together.  I had lots of  
> other ways of shorting things too, all due to insulation melting or  
> shrinking.  I believe the Teflon wire itself it rated for 200 deg C,  
> but you can wrap it around your soldering iron and it will only  
> distort a bit but not melt through.
>
> With all soldering, leaving the iron in contact for a little while  
> longer dramatically helps get all the material up to a uniform  
> temperature (as opposed to quickly touching the iron and solder and  
> withdrawing both at the same time).  The Teflon insulation gives you  
> that extra time, which would horribly melt many cheap thin PVC  
> insulated wires.
>
> I use Kester 44 solder.  The 44 rosin works a LOT better than most  
> others for prototyping.  It is tin-lead solder, and for prototyping  
> the lower melting temperature is much, much easier to work with than  
> lead-free solder.  For circuit boards I use Kester 331 and  
> additional liquid flux, both of which are an organic water soluble  
> fluxes that MUST be washed off with water.  For pads-per-hole  
> prototypes, Kester 44 (or similar) is really the best choice.
>
> I mostly use 30 gauge wire.  I'll typically strip one end (usually  
> an inch or two of bare wire), lay it on the board with the  
> insulation right where one end will solder, then I pull it around  
> the path I intend while pressing the end still.  I'll route a tiny  
> bit of extra length past the other solder location, then pinch it  
> tightly with my fingers at that spot on the wire.  Then I use the  
> wire strippers to strip the insulation and work it down the wire,  
> but NOT completely off.  If it breaks free suddenly and goes all the  
> way off, I cuss like a sailor and try again, but usually I leave  
> enough bare wire that I can push the desired length of insulation  
> down to the point where only a tiny bit of than 1-2 inch bare wire  
> is left exposed.  Then I solder that end, route the wire on the  
> board, and solder side that's still connected to the spool.  Each  
> time, the 1-2 inch of bare wire gets a little shorter.  There is  
> very little wasted wire, and the result is pretty accurately  
> measured wire lengths for a nice-looking wiring job.  I usually add  
> about .1 to .2 inch extra length so all wires take a somewhat curved  
> path, which helps later if you need to push them around to access a  
> resistor, capacitor or pad underneath.
>
> When wiring, if I'm pretty sure a wire will definitely connect to a  
> pin no matter when else I change, and especially if another wire  
> might connect to that pin later, I usually go to the trouble to bend  
> a small hook on the wire.  I use wide blade tweezers, which give  
> most of the strength of needle nose pliers but are easier for doing  
> such small work.  I lay the hook over the pin and squeeze it tightly  
> to grip the pin, then solder.  If it's the end of the wire (still  
> attached to the spool), I wrap the wire half way around the pin,  
> clip it off with cutters, pinch with tweezers, then solder.  Later  
> if the solder is reheated, the wire will tend to stay in place, kind  
> of like a wire wrap (though a 180 degree hook squeezed on seems to  
> have more gripping power than wrapping all 360 degrees... it seems  
> impossible to wrap anywhere nearly as tight as a real wire wrap too  
> does).  Bending hooks is time consuming, but when you add a second  
> wire the first one stays put.  Of course, I often just heat the  
> solder and press the wire in, especially if it's one of those last  
> wires (on top of all others) that I'm likely to change as I fiddle  
> with the design.
>
> The big advantage of pad-per-hole soldering is it's the cheapest way  
> to prototype, even if you spend a little extra for a spool of the  
> Teflon insulated wire.  You also get a pretty durable finished  
> construction that's not much larger than a good PCB layout would  
> have been (if using through-hole parts) and there's not much  
> pressure to take it apart like with solderless breadboards, and you  
> can just start wiring things up without having the design complete,  
> unlike making a circuit board.  The main downside is it takes more  
> time to connect all the wires by soldering, especially if you do  
> those extra little things like bending the hooks on most wires.
>
> If you've seen things I built this way at the meetings, well, that's  
> how I do it.
>
>
> -Paul
>
>
>> i've never done wire wrapping, tho i've seen it done.  it always
>> looks like a complicated loom for gem sweaters.  is it an easier way
>> to build a complicated circuit than soldering?  what are the start-up
>> costs?  you need a wire-wrapper, right?  and some pins?  what else?
>>
>> i'm down with soldering, but i like to know my options.
>> i already found the padded perfboard at the lombard Radio Shack.
>>
>> -m-
>>
>> ---
>> HELP!  A Bear is Eating Me!
>> http://helpabeariseatingme.com
>>
>> _______________________________________________
>> dorkbotpdx-blabber mailing list
>> dorkbotpdx-blabber at dorkbot.org
>> http://music.columbia.edu/mailman/listinfo/dorkbotpdx-blabber
>>
>>
>
> _______________________________________________
> dorkbotpdx-blabber mailing list
> dorkbotpdx-blabber at dorkbot.org
> http://music.columbia.edu/mailman/listinfo/dorkbotpdx-blabber

---
HELP!  A Bear is Eating Me!
http://helpabeariseatingme.com