12
\$\begingroup\$

My breadboard wiring is disorganized and this makes complex circuits hard to debug and so on.

Are there any standard rules for how to arrange the wiring?

Something like "always leave x holes between components" or "always put resistors before switches rather than after".

Or does everybody just make their own rules as they go along?

Any links to resources would be helpful.

\$\endgroup\$
  • 6
    \$\begingroup\$ If there's nothing critical about it, make it look pretty or organized for whatever floats your boat. If there is something to worry about, then worry about it and make that right. My only bright line rule is "make it work well with the right features at a cost you can afford in a size and weight that works for a power budget you can dissipate and otherwise tolerate." If you are having trouble debugging, then work on an appearance that helps you there (other things allowing.) \$\endgroup\$ – jonk Aug 9 '17 at 6:41
  • \$\begingroup\$ @jonk: answer-quality comment :) Would you mind actually converting it to an answer? I'd like to expand on it, but it's hard to refer to a comment in an answer... \$\endgroup\$ – Marcus Müller Aug 9 '17 at 6:56
  • \$\begingroup\$ @MarcusMüller Looks like there is a pig pile already here (no offense meant to pigs or answerers.) Still want it, separately? \$\endgroup\$ – jonk Aug 9 '17 at 7:28
  • 1
    \$\begingroup\$ only if you care enough. I was going to make the point that "making it easy to debug" is a) a core argument against overly large breadboard designs (what with P(#loose contact>0) for N contacts becoming large) and that it's a good idea to modularize, find things that work, then solder these as fixed, "trustworthy" units, and continue. \$\endgroup\$ – Marcus Müller Aug 9 '17 at 8:04
  • \$\begingroup\$ +1 for the point about modularizing. i had been wondering how larger projects were handled. \$\endgroup\$ – thatsagoal Aug 9 '17 at 10:32
18
\$\begingroup\$

Face all ICs the same way.

Input goes to the left end of the board, output comes from the right end. current flows downwards

Much like the conventions in drawing circuit diagrams.

If any rules cause a problem ignore them as neccessary.

\$\endgroup\$
  • 2
    \$\begingroup\$ I always input on the right, but then that's just because I am right handed and all my bench supplies sit to the right ;) I second this - face everything same way and try to keep the wires as short as possible so you don't get too much 'spaghetti' , they have kits of premade wires on ebay and such that are helpful time savers. \$\endgroup\$ – Rendeverance Aug 9 '17 at 11:54
  • 2
    \$\begingroup\$ Not just ICs - take the trouble to design the board and place all the components so the labels are the same way up (and components with color coding all read in a consistent direction). It will save you a lot of time in the long run. \$\endgroup\$ – alephzero Aug 9 '17 at 19:39
15
\$\begingroup\$

First and foremost, solderless breadboards are rarely a deployable format for electronic circuits. They work pretty well for certain (not all) kinds of prototypes. If you have a circuit that you intend to have any sort of lifetime, your plans should include some way to get your circuit off of a solderless breadboard, and into something more permanent.

There are a number of things you can do to make your life easier.

Things like color codes really help. Pick standard colors for Vss, Vcc, and ground. Stick with them. As mentioned by others, place pin 1's for all IC's in the same orientation. For inportant locations, make a little flag with a stub of wire and some tape to make a label. Sometimes it's useful to populate all your IC's first, and get them all powered and grounded neatly before you implement anything else.

The biggest thing you should do, however, is to break your circuit up into functional subcircuits in your head, and plan in advance how you intend to test whether a subcircuit is working. Then, build it, test it, leave it alone, and move on to the rest of the circuit.

Usually, I do this from left to right -- inputs moving along towards middle stages, moving along toward output stages. That's a generality, however, and sometimes it makes more sense to deal with output stages first -- usually on the right side of the board.

If you're doing microcontroller design, this can get confusing, because the architecture is more of a central hub thing, with one unit receiving and sending ins and outs. In this case, you need to have a microcontroller setup that isolates various subcircuits, one at a time to let you test them. It should generate whatever test signals you need. Be sure to remember some reset circuitry.

Of course, breadboards are not robust. You move an oscilloscope probe, and short a cap to a resistor that's not mounted as stably as it should be. That's just the nature of the beast. When a circuit gets too big and complex, there are diminishing returns to leaving it on a solderless breadboard, mostly because of unintentional stuff like surprise shorts on a circuit that worked 5 minutes ago. Take the time to put down your working subsections on soldered breadboards. Better yet, build and test a subcomponent, plop it into a PCB design, and move on to the next subcomponent. Spend a few dollars and a week to get your prototype PCB, populate it, and enjoy your functional circuit.

\$\endgroup\$
  • 1
    \$\begingroup\$ +1 for functional blocks. When you're in the early-prototyping phase and the eventual board-layout is something for later, having every function occupy a different part of the board keeps things simple and easy to debug. If path lengths aren't a problem, even multiple boards can be used. \$\endgroup\$ – Mast Aug 10 '17 at 7:28
  • \$\begingroup\$ "Breadboards are not robust" - I'll say. As part of my CS degree, I had to implement an entire computer (Z81, ROM, RAM and logic chips, outputting to a CRT) on breadboards. Unfortunately, the output was never quite right, and a little pressure on the RAM would completely change the output. \$\endgroup\$ – Baldrickk Aug 10 '17 at 9:50
  • \$\begingroup\$ "... rarely deployable ..." -> :-). BUT - Tidy as possible. Add sponge rubber overlay that compresses wiring & components. Test deploy with eg board pressing on sponge. Ensure all wires are routed and shaped to sit nicely when compressed. Ensure components do not short when pressed down and that intended contacts remain made. Once "stable", insert into suitably sized clamshell instrument case and close onto foam and breadboard and screw securely shut. | A "reasonably good" degree of reliability can be obtained. With due care good enough for demos, shows and possibly even (gasp) ongoing use. \$\endgroup\$ – Russell McMahon Aug 16 '17 at 2:03
  • \$\begingroup\$ @RussellMcMahon, I'd never accept a circuit like that. Possibly for demos, but my experience is that if it's gonna go bad, it's gonna happen at a demo. If it's too complicated to place it in some more stable arrangement, it's probably too complicated for a solderless breadboard in the first place. To top it off, I only use good breadboards, and it would be a sin to lose them in a box where I couldnt use them for what they were meant for \$\endgroup\$ – Scott Seidman Aug 16 '17 at 2:15
7
\$\begingroup\$

Do not be lazy organizationally when starting a project (especially a big one). I usually tape my breadboards and all other components to a cardboard panel to make it more organized. I also only use jumper wires if I have to -- use these low profile wires instead; it will make debugging really easy.

enter image description here

Also as you are making your project, reflect every change you make in an Eagle (or other circuit design program) schematic. It will make your life easier in the future, especially if you want to turn your project into a PCB. Good luck on your projects.

\$\endgroup\$
5
\$\begingroup\$

For breadboard, unless your circuit is very simple or you are in an evironment that requires it (school), it is often much easier (and even cheaper if your time costs money) to simply design and order a pcb.

For solderboards, you can get a bit more complicated, but be careful. I personally have used Eagle CAD to design solderboards (because i did not have enough time for the pcb option), and it made everything much easier.

For both cases however, my main advice would be to give yourself some space. dont try and fit everything into a corner to "save some space", in case of a rework, you would find yourself with some kind of ununderstandable vertical monstrosity.

\$\endgroup\$
  • \$\begingroup\$ Why the downvotes? \$\endgroup\$ – Sclrx Aug 9 '17 at 10:37
  • 1
    \$\begingroup\$ +1 for not fitting everything in the corner... This was my common mistake when I started creating own PCB's. Leaving some space to insert additional LED, filter or signal output is easier than keeping it small and trying to fit it between elements (unless that's really needed), resulting in piggybacking lots of them which then caused troubles with finding errors. When you're sure everything works as expected, you can optimize the layout and, eg. switch to SMD components. \$\endgroup\$ – Mark Aug 9 '17 at 13:19
4
\$\begingroup\$

Before starting with your breadboard, use a tool like VeeCAD or Fritzing which support breadboard and stripboard routing. It's much easier to find a good routing solution using software, instead of rewiring the breadboard again and again when parts of the circuit end up being too far away or don't fit.

\$\endgroup\$
  • \$\begingroup\$ While Fritzing does support breadboard and stripboard, it does not support solderboard where the pins aren't connected per row by default. It also doesn't support placing wires/components at odd angles and a slew of other things that made me stop using it fairly quickly. \$\endgroup\$ – Mast Aug 10 '17 at 7:32
1
\$\begingroup\$

Breadboard 'contructions' can be made good enough to use other than on bench so as to be, with due care good enough for demos, shows and possibly even (gasp) ongoing use. See at end.

While not recommended until much experience gained (or at all :-) ) I have made useful prototype benchtop (ie not portable) assemblies with say 6 large breadboards. In a few case with several microcontrollers intercommunicating. Plus interface to external high current and voltage circuitry (NOT on bread board per se.). Building the whole assembly on a sheet of ply or similar allows it to be lifted off the workbench and stored, so that any one of a number of circuits can be worked on when required.

Best are 4 power rail boards with power supplies along rails. Add extra vertical power or large connection rails if needed. Keep ICs all one orientation except in extreme special cases.

Learn how many "spare" columns between ICs works for you for interconnects.

Drawing the circuit with breadboard limitations in mind THEN breadboard layout before starting helps for major circuits. Minor stuff often OK "on the fly". I have not used layout software such as Fritzing but I'd expect that the lack of ability to simulate all real world practices would be excessively limiting.

Note that some components may "fit" in holes but damage breadboard "springs".
eg 1N400x diodes may qualify. For such components solder a wire to each lead and cut just long enough to fit BB.

Be aware of inter-row capacitance issues and contact resistance issues.

Be aware that low cost BBs may be inferior BUT that quality may or may not improve with cost.

Breadboards CAN be used in a "portable" role with reasonable success if they are enclosed in an outer ":shell" and the wiring and components are restrained from moving under normal handling forces. I have had good results by doing the following.
Build as tidy as possible.
Add sponge rubber overlay that compresses wiring & components.
Test deploy with eg board pressing on sponge.
Ensure all wires are routed and shaped to sit nicely when compressed.
Ensure components do not short when pressed down and that intended contacts remain made.
Once "stable", insert into eg suitably sized clamshell instrument case and close onto foam and breadboard and screw securely shut.
By this means a "reasonably good" degree of reliability can be obtained. With due care good enough for demos, shows and possibly even (gasp) ongoing use.


A good example of "how not to do it" - even though it worked OK.

enter image description here

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.