How do you guys stop breadboarding?

[stallik]

I have never built a Tubescreamer!

Despite the fact that this circuit forms the basis of so many custom pedals, I've never managed to get my breadboard build converted to a real build. I must have modded it a thousand times and mostly, it sounds great but then, the next mod does something to make it better (or worse so gets downgraded again) and so on and so on.

I must have had this circuit or a variation of it on  the same breadboard for about 4 years. I use it all the time but never consider it complete so just how do you decide "thats worth casting in stone"? Is it just a case of "that'll do?"

[willienillie]

eh- this comment wasn't really relevant.

[jimitrader]

Cheers! What i do is go down to the local hardware store and get an 8 foot long sheet metal u shaped thing..they are the same size as a 1590BB box minus the bottom and front/back sides. I then i cut them off with a dremel tool. then it can be a temp holder for final testing before doing the 1590bb ...use the tester one as a template later ..that way if i change my mind a few times until i settle on a setup i like im not ruining a good box...way cheaper that way...and i have only done one with 1590bb because i don't play out and they work just fine with no bottom or f/b ? for me this works..as i am still learning i like to stay on the cheap..lol...i have made about 12 pedals so far and still have a few feet left...anyway hope that gives you some ideas..rock on! JimiTrader
also this site ( https://effectslayouts.blogspot.com/ ) has boards and stuff with a way cool list of projects that are verified..and print out for doing your own pcbs...its way faster that way than breadboard first?

[stallik]

Thanks guys, there are some cool ideas there but I was thinking not so much 'does it work' , more 'is this different enough from the original to call it something new'

[Phoenix]

I take a bit of a different approach to prototyping, and typically avoid the breadboard.

If I want to try something new I almost always go straight to perfboard/pad-per-hole/matrixboard. This avoids annoyances with dodgy breadboards and potential noise issues from unshielded circuits as I'll put it straight in an enclosure, which is nice.
This forces me to largely "commit" to a circuit, and also hones my intuition for layout as I rarely do any planning other than mechanical (board-mounted-pot/switch location), prefering to "wing-it", and almost never bother with steps like DIYLayoutCreator.

I do use plated-through-hole perfboard so that I can more easily make changes (much less likely to lift pads compared to single-sided boards, and allows soldering from both sides if I paint myself into a corner), so tweaks are definitely still possible. I'll often mod until something breaks (introduce some problem that's too much effort to track down on a lashed-together prototype), before starting over on a more "final version". As items like pots and switches are unlikely to move, this means the enclosure is still useable for the second iteration, so lost parts investment is usually minimal, just rip out the now non-functional prototype and slap in the new one. Any of the more pricey parts like pots and switches are usually easily salvaged if you're thrifty/on a budget and provides desoldering practice anyway.

While for anything I'm going to build multiples of or has significant surface mount parts I'll design a PCB for, I've built many many circuits this way, including items I use every day like pieces of test equipment. White+pink noise generators, audio probes, oscilloscope differential probes, signal generators, pulse testers, zener testers, production test-jigs, etc.

Might be just the thing to kick-you-in-the-pants and commit?

[deadastronaut]

i have projects on many different breadboards for months sometimes....i hear ya.

when is it time to call it a day and complete?.....never. it can always still be tweaked.

you just have to bite the bullet and say ''thats it'' thats what 'I' wanted....done.

[Joncaster]

I'm having similar thoughts with a spring tank reverb. It's on both of my breadboards (psu and circuit), in a shoe box while I work on my amp.
I can't take it off the breadboard till I finalize it, but I don't really know what the finishing touches are.
What does my perfect spring reverb even sound like? For all my other circuit tunings, I've reached that point where it has a 'thing', a musicality and feel that let's you create through it. Without that "I know it when I hear it" feeling, I'm not committed enough to perf it and box it.
It's damn close, though. And as an aside, I highly recommend trying a compressor in a spring driver (ala ESP).

[GibsonGM]

I generally wait til I get something that gets me so excited I have to crank my amp up and just wail on the thing for a hour or more - after I've already tweaked to a point of nervous exhaustion.     Then, while I'm still pumped up, I solder it up and get it in an enclosure.   

By the time this wears off, it's too late.  So I have to make another circuit and repeat the process!

[Mark Hammer]

I take a bit of a different approach to prototyping, and typically avoid the breadboard.

If I want to try something new I almost always go straight to perfboard/pad-per-hole/matrixboard. This avoids annoyances with dodgy breadboards and potential noise issues from unshielded circuits as I'll put it straight in an enclosure, which is nice.
This forces me to largely "commit" to a circuit, and also hones my intuition for layout as I rarely do any planning other than mechanical (board-mounted-pot/switch location), prefering to "wing-it", and almost never bother with steps like DIYLayoutCreator.

I do use plated-through-hole perfboard so that I can more easily make changes (much less likely to lift pads compared to single-sided boards, and allows soldering from both sides if I paint myself into a corner), so tweaks are definitely still possible. I'll often mod until something breaks (introduce some problem that's too much effort to track down on a lashed-together prototype), before starting over on a more "final version". As items like pots and switches are unlikely to move, this means the enclosure is still useable for the second iteration, so lost parts investment is usually minimal, just rip out the now non-functional prototype and slap in the new one. Any of the more pricey parts like pots and switches are usually easily salvaged if you're thrifty/on a budget and provides desoldering practice anyway.

While for anything I'm going to build multiples of or has significant surface mount parts I'll design a PCB for, I've built many many circuits this way, including items I use every day like pieces of test equipment. White+pink noise generators, audio probes, oscilloscope differential probes, signal generators, pulse testers, zener testers, production test-jigs, etc.

Might be just the thing to kick-you-in-the-pants and commit?

I actually adopt a similar sort of approach.  I have more than enough breadboard real-estate available, and I bought myself a bunch of those jumper wires with the pins on the end for breadboard use, but I never use any of it.  I either perf or pad-per-hole it.  I think my bias against breadboard is that I can never get the arrangement of components to visually flow the way the schematic does.  That's the same grievance I have against stripboard: it's hard to think with.

The trouble with that method is that I can end up with a lot of stuffed-and-wired boards that don't get installed into an enclosure, and don't have any identifying marks on them to indicate what they are, when I dig them out of the bin of shame a year or more later.  What IS that circuit?  Hmm, nothing particularly distinctive about using a pair of dual op-amps.  I don't see any diode pairs.  A 10k pot and a 100k pot.    Any ideas?  Nope.  Guess it goes back in the bin. 

[davent]

How do you draw the line between better and just different/fresher than the previously rendition?
dave

[Mark Hammer]

Me?  Well I try to fit perfed-and-completed circuits into separate ziplock bags that I can write on with a marker.

[davent]

How do you draw the line between better and just different/fresher than the previously rendition?
dave

Me?  Well I try to fit perfed-and-completed circuits into separate ziplock bags that I can write on with a marker.

In the tweaking process, is it really better than before or is just different and fresh?
dave

[Mark Hammer]

My mods are generally not intended to be "improvements to tone", but more often functional options.

[patrick398]

This may just be me, but i feel like i never have a proper 'feel' of how a circuit works when it's on the breadboard. I usually breadboard for a while, then when it's basically 'finished' i'll go to PCBs. It's only when it's mounted in an enclosure that i feel like i can get a true feel for how it sounds and how the controls interact etc so i tweak more from there

[amptramp]

A long time ago, I was forced to use breadboards at work and vowed to never use them again, a vow I have never broken.  If you find a breadboard that fits electrolytic capacitor leads, resistors will fall out, particularly the 1/10 watt ones that I have over a thousand of.  Transistors and IC's may or may not fit.  And nothing seems to stay connected correctly for very long.

I have started using prototyping boards with two busbars under the 0.3" wide IC's and three-hole pads on either side.  If you need more than three pads, four- and five-pad versions are available but the most common idea is to space the IC's apart and and use extra three-hole pads between them which can be connected with component leads and insulating sleeving if the pads are not adjacent.

[kraal]

Hi,

Here is how I decide when to "produce" a finished version of my ideas.
Maybe it can help you organize yours.

I always apply naming and versioning best practices taken from development as follows:

• Each project has a (code) name and a described objective (why does the project exist, and what do I want to achieve with it) that is SMART: (specific, measurable, achievable, relevant, time-bound.)
• Then I work on it and create versions that match the following format X.Y.Z (X being the major version representing a set of features required to reach the defined objective, Y being the minor version which represents changes made to how the features of X are implemented, and Z is the fix version number which represents "bug fixes"), for instance "1.3.1"

Having a name, an objective and a clear way to create versions helps me detect when I'm about to start something new, which in turn informs me that I'm about to finish something (or start working on a completely new objective).

• If it's a new idea, I create a new project,
• when I add a new feature to an existing idea  I increment X (and reset Y and Z),
• when I try a new way of doing things, I increment Y(and reset Z),
• when I fix an issue, I increment Z.

Then the "produce or wait" decision is taken as follows:

• If it's a new version and if it's not mandatory, it tells me that I can "produce" the current version,
• If it's mandatory, production will wait.

Additional note:

I personally don't use breadboards a lot (only when trying something completely new), not that I don't like it, but I don't have much space and not everything can easily be done on a breadboard (especially with you need a smd component).

I draw the schematics of everything I try in KiCad, try to simulate as much as possible with ltSpice (when possible), document a lot what I do, look at the impact of modifications, create 3D models, then design and order PCBs.
Then I populate the PCBs (with sockets for the parts I'm not sure), mod them when needed, then improve the schematics or the PCB design.
Sometimes I add a dip switch and "free pins" when I want to be able to try other values later.
All my project are versioned using git (makes easy to create branches to test new ideas, PCB layout, etc.)

In my case your question could be "when do you stop simulating / starring at your 3D model" :-) The answer is the few rules I have described above + I try to explain what I'm working on to my family / friends, if I get a "it looks cool" that's also a good hint :-)

Kind regards,

M.

[stallik]

^ that's exactly what I do when writing software. Wonder why I never thought to apply it to builds.

Some great suggestions guys. Think I'm going to commit to whatever is on the breadboard now, then build subsequent versions.

As an aside, whereas most 'based on a TS' circuits I see appear to do more, my attempts almost always do less but do that bit quite well.

[stallik]

Finalised my first TS. It's a double. 2 different circuits, A/B or both. Fun thing is that when auditioning the clipping diodes - long leads into a socket, I noticed that my fluorescent desk lamp introduces a pleasant octave down on one circuit. I'm using a DD to give me room for the foot switches but it's still too small to stuff the lamp in and have it switchable

[amptramp]

Finalised my first TS. It's a double. 2 different circuits, A/B or both. Fun thing is that when auditioning the clipping diodes - long leads into a socket, I noticed that my fluorescent desk lamp introduces a pleasant octave down on one circuit. I'm using a DD to give me room for the foot switches but it's still too small to stuff the lamp in and have it switchable

That is one thing about semiconductors - they are light sensitive.  A number of people have designed fuzzes where they put the clipping diodes on the outside of the box so they can see when they are conducting.  Bad move.  Any stage lighting will modulate the signal and LED's are the worst - they have a high enough leakage already, you don't need to add photoconduction that is modulated at line frequency.

[StephenGiles]

I don't do smd so breadboard is fine for me.