Prototyping?

If you can get all or at least nearly all parts in thru-hole you can use vector board with a ground plane. If it has to be mostly SMT I'd do a quick-spin layout and have it made at one of the PCB "fast food" places.

--
Regards, Joerg

http://www.analogconsultants.com/

Thanks Joerg.

Jeorg would you mind linking me to what you mean by the vector board process you described.

When I prototype I begin with a theory, (schematic) then refine it or discard it, and start anew, so I need to make swift revisions.

Thanks Ken

It looks like this:

With some luck you can find smaller and less expensive versions. I mostly use "Eurocard" size which is 100mm by 160mm. Basically there is a ground plane on one side with enough clearance around the holes so it doesn't short IC pins and the like to that plane. The really good ones also have plate-through holes. When you want to connect something to ground you can do it right at the pin of the part.

If you expect a lot of changes spread the parts a bit because each hole can only take so many dismantling cycles.

--
Regards, Joerg

http://www.analogconsultants.com/

I like to do this:

It works at low level or GHz. The circuit is easy to visualize, can be physically labeled and commented, and you can keep each circuit forever.

John

this:

Thanks Jeorg, gotcha now. That's the kinda of pre-printed PCB's I'd got to for a limited run custom manufacture and testing. It's good stuff.

Right! that's why I'm thinking going to wing-nut peg-board technology to avoid soldering at the circuit verification process.

Thanks again Ken

That's the way to go for RF stuff. Although it does get old for larger concoction of logic chips, bus decoders and so on.

Hey: No eating of potato chips when breadboarding this way. Leaves your finger prints all over the place.

Ah, ain't that the voltage inverter you presented the other day?

Just make sure the soldering fumes are vented out. This technique is heavy on the solder.

--
Regards, Joerg

http://www.analogconsultants.com/

Mine end up looking like this:

Note that for SMT parts I usually home-etch an adapter, there are a couple in that photograph.

Sometimes I'll put a whole sub-circuit on a home-etched board, like the power supply, when it makes sense to do so (that PS was mostly SMT parts).

this:

Nowadays even mundane logic and transistors are so fast that a circuit without a ground plane can be hard to keep from uncontrollable oscillation. Also, I'd stay away from that white solder-less breadboard. I've got some here in the lab but I never use it for anything above 1MHz or so.

--
Regards, Joerg

http://www.analogconsultants.com/

For audio and low-RF you can use a solderless breadboard from Radio Shack. They have a lot of capacitive cross-talk, and the contacts wear out, and your lead lengths will be necessarily long. But if you take that all into account you can do some pretty good work -- then if you want the circuit to be permanent you can buy a PC board from Rat Shack that has the same layout as the breadboard and transfer everything over.

Folks will disagree with this approach, but it works for me.

For circuits that are not too complicated but must work at higher RF or for anything that'll be sensitive to lead length or cross-talk I use dead-bug construction on a ground plane, like John Larkin's post.

For complex circuits (basically anything with a parallel data bus) I lay out a prototype board and send it off to the local 24-hour turn PCB place.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

...and above a couple hundred MHz or thereabouts, if you have any matching or filter networks you'll end up re-tuning the real PCB anyway due to the significant difference in parasitics. Probably not a big deal, though.

I'm actually surprised that there isn't a lot of noise or loop instability problems with John's switcher... even with switchers operating at at the mid-hundreds of kHz, I've seen multi-inch wires cause problems.

There's no reason to breadboard stuff like that.

I wish I could buy gold-plated copperclad... the copper looks great when you start, but goes to hell in a few days. Hey, maybe a pcb house would plate some copperclad for us. THAT would look classy.

Too late for me. I learned to solder when I was three years old.

John

Hold the tomatoes but I have done UHF stuff on double-sided phenolic, including stripline resonators.

It's sometimes amazing what you can get away with in prototyping. Except with high power circuits where it may be a good idea to wear eye protection.

--
Regards, Joerg

http://www.analogconsultants.com/

The high-current path is pretty stiff on the breadboard. It runs about

600 KHz. Capacitor esr was a big hassle... we had a thread on that, and I wound up using a tantalum. The circuit is in use now on two different products, one +12 to -12 (for opamps) and one +12 to -5 (for ecl), and both work great.

John

Ken, if you want to build a neat prototype on Vector board where numerous parts only available in SMT this can help:

Called "surfboards", Digikey has them but they are expensive.

--
Regards, Joerg

http://www.analogconsultants.com/

Depends. When you want to do unorthodox things a CPLD or FPGA doesn't cut it.

I bet you could find one that does. Some domestic ones seem rather hungry these days, judging by the increasing amount of "specials" they send me.

Wow! I started around eight or so and thought that was early. Thinking about that first solder iron still gives me goose bumps. 230VAC, no PE (!), heating coil visible through grid holes and could easily be touched by resistor or capacitor leads if held wrong. OSHA would have had a hissy fit.

--
Regards, Joerg

http://www.analogconsultants.com/

There's still no 2nd source for that LM5112, or is there by now?

--
Regards, Joerg

http://www.analogconsultants.com/

What I created and used at GenRad, 1977-87....

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
         America: Land of the Free, Because of the Brave

this:

I hate to admit this, but I've always been color blind, and it's getting worse. Now I need a full spectrum light and magnifying glass to read the f**king resistor values. So what I'm thinking of doing is mounting my standard components on bits of cardboard and wiring them in. Regards Ken

Switch to SMT. They use numbers instead of colors. Of course, you still need a magnifying glass...