Do you personally use a plastic solderless breadboard?

NO!

Parts soldered to gold-plated FR4.

I buy Radio Shack 276-170 linear breadboards in bulk for linear work. For RF I do "Dead Bug" or "Manhatten" style work on very thin microwave PCB.

I do not use the solderless breadboards, because of the element to element capacitance, and because the reliability of any given connection is horribly bad.

Steve R.

That depends, a lot, on what is being prototyped, the packaging of the components, performance expected from the circuit, etc.

E.g., too many leads makes many microcomputer (not microCONTROLLER) designs impractical to prototype in that manner. Likewise, high operating frequencies, low supply voltages (i.e., poor noise margins), high power signals, etc. all can make that sort of breadboarding more hassle than it is worth.

OTOH, a slow analog circuit is ideal for that sort of approach. Just be aware of the limitations that come with it (as well as connections that can be flakey, come loose, etc.)

If you're ever going to *do* anything with the design (i.e., commit it to foil), then it is usually better to just prototype

Consider: if you will eventually commit the design to foil, then there will come a time when you have to make your first

With that in mind, let your *first* prototype be the *foil* prototype and deal with design/layout/routing issues in that first shot -- skipping one iteration (ideally). This has the added benefit that you can (relatively easily) create a second copy of that first prototype -- not that easy with other approaches!

Of course, you have to be reasonably confident in the design before you begin!

Is this one of those surveys in advance of one of those kickstarter campaigns? I seem to recall filling one out months ago, but I can't remember if it really was related to breadboarding. I know the people behind the survey had no real idea of the market.

And then, just last week, I saw something about a kickstarter campaign of a little plastic gizmo that amounted to a microUSB connector with leads that fit the breadboard. Have we dumbed things down that much? Forty years ago, when those breadboards were new and I got one, I'd just wire some bits of wire to things that didn't fit into the breadboard holes. No need to buy something.

That said, one uses what fits the situation.

Those white breadboards are great for ICs since they are layed out for ICs. They aren't so great for discrete components.

So you put your IC breadboarding on the white breadboards, and then hou use scraps of copper circuit board for the rest, soldering to the copper when you need a ground connection, and keeping everything in the air above the board. Works fine for breadboarding, you can use components salvaged fro something with small leads, or (if the circuit allows it) use parts without cutting their leads, so they can be reused later. I used the same piece of circuit board for years, it's even better after a bit of use since the solder is already on the board, you just need to heat it a bit, maybe add a bit of solder, and solder the component lead. This arragnement makes it really easy to change things, or scrap it all. Perfboard is lousy for making changes.

If you build compactly, you can just make the "breadboard" the finished product, hide it behind a panel somewhere.

Michael

I learned the hard way that they are horribly bad even at (audio)

I use PCBs for prototyping. There are ways to get them inexpensively if you aren't in a hurry.

If I'm in a hurry it means I am being paid for the work and the costs of a quicker prototype is part of my quote.

I occasionally use one of these for simple and low frequency stuff:

ChesterW

Never. Dead bug all the way, with SMD breakout boards and some uC/CPLD/interface boards from vendors.

Cheers

Phil Hobbs

Depends on the nature of the project. For some things, these are fine - if you don't loan them out to idiots and are not one yourself. I used to maintain a labful of them, and we had to treat them as largely disposable due to idiocy making them disfunctional after a relatively short time. My personal one, bought new, has held up much longer.

Many other options for things where the stray capacitance of these is a problem have been discussed in the group in the past. Those are generally of either the "build up sections of insulator and copper on top of a copper-clad board" or "cut islands into a copper-clad board" flavor. When Mr. Larkin is diverted from politics long enough to post about electronics he often has links to pictures of this sort of thing.

Perfboard with or without little copper rings is another option for some types of work, and you can sometimes even make the little copper rings work with 3-terminal surface mount devices, since through-hole devices are becoming scarcer.

Some types of things you simply have to spin a board to prototype effectively, especially with the rise of surface-mount and the shrinkage of pin-pitch.

The best intentions often don't quite result in a neat breadboard:

I have a small collection of those white breadboards and bags of jumpers. That last time I used them was about 20 years ago for a simple low frequency logic circuit. They're useless for anything over a few MHz, or a few amps. Screwing around with all the little jumper connections takes more time than building a "dead bug" circuit on a PCB. The only real advantage to using one of those white breadboards is that there's no soldering involved. However, I can solder, so that's not much of a benefit.

Dead bug or 3D bug on a PCB. For discrete components, a single sided PCB with the copper layer on top acting as a ground. Components are mounted on the PCB using drilled holes. I use a small router bit or end mill to clear out the ground plane around un-grounded leads. Interconnections are done on the unclad bottom of the PCB using plastic sleeves over component leads.

For SMT, I use small carrier PCB's to bring out the leads so that they can be soldered. They're also handy for easily replacing the SMT device after I blow it up.

For RF prototypes, I either etch a PCB approximation of the strip lines, or pay the price and have the PCB fabricated. Ocassionally, I'll build up the strip lines using copper tape on a single sided PCB. Xacto knife to make changes and fix mistakes.

The uglier the breadboard, the better it works.

That's so beautiful that it belongs in an art museum after you finish your prototype! It seems that SMT makes solderless breadboard a thing of the past.

I've used "Dead Bug" before. "Manhattan style" was a new one for me. One of my old ARRL Handbooks talked about "hacked Manhattan style" (my words).

Hacked Manhattan doesn't use raised islands. Instead one takes a piece of copper plated board and uses a hacksaw to carve grooves through the copper while leaving the phenolic "substrate" mostly intact. The horizontal and vertical grooves are spaced a quarter inch apart to form a quarter inch square grid. At the end it looks like a piece of quarter inch grid paper with grooves where inked lines ought to be. Each quarter inch square of copper forms its own little island.

Did Radio Shack give its 276-170 a makeover? The RS website shows what looks to be a nylon solderless breadboard. At other sites the 276-0170 looks like a plain old perfboard with a nlyon breadboard hole pattern.

I do "Dead Bug" or "Manhatten" style work on very thin microwave PCB.

acitance, and because the reliability of any given connection is horribly bad.

A few months ago somebody recommended _Troubleshooting Analog Circuits_ by Bob Pease (RIP) as good reading material for John Larkin's new engineer. I bought the book (along with the other books that people mentioned that weren't already in my library). (Thank you everybody.)

It turns out that Bob Pease didn't like nylon breadboards for much the same reason that everybody here doesn't like them (stray capacitance and what not). Matter of fact, Pease said that he'd never seen them used at his workplace before.

Pease also didn't like SPICE. Then again, nobody's perfect. ;)

Although your approach requires more time and effort up front, it ought to speed things along later and create a thoroughly thought out design.

I've done something similar. I've used a nylon breadboard to prototype, then migrated all components to a breadboard perfboard, and finally hid the whole contraption in a bud box. ;) What goes on inside the bud box, stays inside the bud box.

This may be the only way to prototype high density SMT.

That was my approach until I read _Troubleshooting Analog Circuits_ by Pease. For solderless breadbroads Pease recommends perfboard and Digikey A208 and A209.

Pease was one year ahead of me at MIT... he's always been a little weird ;-)

I didn't trust Spice myself until the late '70's when I was able to get a GenRad programmer to fix the BJT B-E capacitance (in the Fortran source code)... then I became a believer as product matched simulation perfectly. ...Jim Thompson

Are you sure you want to follow Bob Pease's example?