Process of taking a project from idea to working box?

[add4]

I find that the most difficult part of pedal building is the forward and backward planning practical:
Before starting a project, I think i should
- know the controls the pedal will have. To me that means breadboarding the circuit to know what options I want to drop.
- Make a layout of the pedal, be sure all the components will fit together without touching each other
- Make a circuit layout and etch it
- Make the art for the pedal, print it and make a drilling guide for it, drill
- Paint, label, clear coat
- fit all the pots, jacks, switches and other bezels in
- populate your board
- connect the pots, jacks and other things to the board
- test and hope it works

But in reality at the moment i find i'm reacting to issues rather then planning stuff out (only 2 builds completed so far, so i'm learning a lot ), i usually first build the cirtuit and connect everything, then desoler things to modify it to my tastes (LOTS of lost time). Then i think of a layout, see that a few wires are too short, others too long, drill, then find out the components fit, but they barely don't touch each others, and the inside of the pedal is a mess :p

What is your 'process' to make a project happen neatly without lost time and energy?

[defaced]

Here's a recent build I did that required a bit of engineering to finish.  This is sorta an outline to how I do stuff, though different projects require different levels of detail.  For example, I wouldn't have modeled the box layout had I mounted the controls on the top, but since I wanted this form factor and space was limited on that face, I had to make sure stuff fit and use the right tools during construction otherwise the box would have been scrap.  

Board and layout: http://www.diystompboxes.com/smfforum/index.php?topic=83930.msg808556#msg808556

Enclosure design, drilling and assembly: http://www.buildyourownclone.com/board/viewtopic.php?f=8&t=33177

At first all the steps seem daunting, but if you take them in chunks, it's alot easier.  As you can see by the dates on these posts, this project took a while to complete.  Probably close to 15 hours, but most of that was productive time with little back-stepping.  

I will say that if you're doing your own board layouts, you will save yourself alot of long term heartache by using a program like Eagle or Diptrace which connects a schematic to the parts in the layout.  This will keep you from (hopefully) forgetting to connect stuff.  However, this will increase the learning curve a bit and you will have to re-create schematics you find online in these programs before you can do a layout.  Also, expect your first boards to be prototypes and if they work, then use them.  Learning to do good layouts is a skill unto itself.  

[Gurner]

Here's how I do it (not for stompboxes, but the methodolgy will be the same)...

1. decide what I want the thing to do (requirements gathering) ...sketch out the 'building blocks' that'll be needed, then delve deeper into thje detail of block to be used.
2. make it happen on breadboard....get a gut feel for  the overall result, whether it's viable...if so, press on....
3. revisit step 1, now that I have the thing working on breadboard (bells & whistles time, which often aren't apparent until you have the thing working on breadboard)
4. incorporate bells & whistles decided upon in step 3.
5. test profusely (as obviously it's one helluva lot easier to remedy any issues at the breadboard stage)
6. board layout (Eagle)
7. PCB isolation mill a prototype board
8. Populate board - 'loose wire' the switches, pots etc
9. board testing ...with 'progammer wires attached ...or at least pads to get them on easily (firmware program will likely need tweaking)  
10. If more boards requiired - etch/batchpcb etc.
11. Commision - job done.

...I can't emphasise how much testingverification I do!

[deadastronaut]

yep testing is everything...

breadboard everything  put a true bypass 3pdt toggle on your breadboard so you can instantly hear bypass and on......
tweak to your taste..test test test...your ears may be different on different days...what sounded cool yesterday may still need tweaking today.....

nothing worse than building something you don't like, and have to pull it apart to mod it....

design pcb,check it over a few times... etch , populate..test test test again...

when happy then design a box/ drill, put it all in the box.....then test test test....

lots of time is eaten up, but its worth it in the end when you have a pedal your completely happy with..not a dust collector on a shelf...

then have a beer, and soak in the satisfaction that you now have a great useful pedal...

[slacker]

I do pretty much what Gurner posted, with the exception that I use veroboard rather than PCB. Most things don't get any further than stage 2 because I realise the circuit's not for me or it doesn't do anything different enough than something I've already got. Even fewer things get past stage 5.
Point 8 is really important in saving time and making things easier on yourself, make sure the circuit works before you box it up.

Personally I don't worry too much about lost time, this is a leisure activity that I enjoy doing, so in some ways the longer things take the better  

[boogietone]

For box layout, I use Visio to place physical components for spacing. This helps immensely and provides a drilling guide that can be taped to the box.

[PRR]

> only 2 builds completed so far

Build more.

When you first learned guitar, first two complete songs, you were better than when you started but not yet a Guitarist.

Build more.

> without lost time and energy?

I think that's part of the learning.

Don't have to make EVERY build "perfect the first time".

Think of it as practice. Playing scales and riffs and changes over and over in private before you go on stage.

[smallbearelec]

I discuss some of my thinking process in this article:

http://www.smallbearelec.com/Projects/FurFace/FurFace.htm

[R.G.]

... without lost time and energy?

See if you can find a copy of "The Sensuous Gadgeteer".

Time and money are only lost if you have already learned all you can about making the widget in question and are only trying to meet a production/money schedule.

For things where you are learning as you go, thinking about productivity and efficiency is good only to the point that it does not get in the way of your learning and doing a good job. One saying I liked from The Sensuous Gadgeteer was to consider the finished product to be just the garbage of the work. If all you want is the finished widget, it is usually faster to work more hours and buy it.

You're flirting with what has been formalized as Project Management. I was extruded through a set of coursework from the PMI. It's an interesting way to look at things, and is hard-as-nails on the topic of how to not waste time and energy (energy being directly interchangeable with money). The big secret? Sit down and mentally write down in excruciating detail every step of the process from planning to finished goods, including in big organizations the persecution of the innocent and promotion of the guilty. In addition, write down the known unknowns (what you already know you don't know as in "I don't know how to do powder coating", for an example) and make some allowances for unknown unknowns; things that will come up that you don't already know that you don't know.

You're getting as responses what the PMI calls a "work breakdown structure". Best advice? Read them all, then spend a day or two envisioning your own, mixing the ones you've received and visualizing it in your head. Every project winds up converting some known unknown to a known, and an unknown unknown to a known unknown or a known. I like to think of it as education. 

[PRR]

> See if you can find a copy of "The Sensuous Gadgeteer".

Out of print (a long time).

$12 to $25+ at ABE.com used-book marketplace.

[add4]

Time and money are only lost if you have already learned all you can about making the widget in question and are only trying to meet a production/money schedule.

For things where you are learning as you go, thinking about productivity and efficiency is good only to the point that it does not get in the way of your learning and doing a good job. One saying I liked from The Sensuous Gadgeteer was to consider the finished product to be just the garbage of the work. If all you want is the finished widget, it is usually faster to work more hours and buy it.

You're flirting with what has been formalized as Project Management. I was extruded through a set of coursework from the PMI. It's an interesting way to look at things, and is hard-as-nails on the topic of how to not waste time and energy (energy being directly interchangeable with money). The big secret? Sit down and mentally write down in excruciating detail every step of the process from planning to finished goods, including in big organizations the persecution of the innocent and promotion of the guilty. In addition, write down the known unknowns (what you already know you don't know as in "I don't know how to do powder coating", for an example) and make some allowances for unknown unknowns; things that will come up that you don't already know that you don't know.

You're getting as responses what the PMI calls a "work breakdown structure". Best advice? Read them all, then spend a day or two envisioning your own, mixing the ones you've received and visualizing it in your head. Every project winds up converting some known unknown to a known, and an unknown unknown to a known unknown or a known. I like to think of it as education. 

Very nice advice here..
After 3 university diplomas in theoretical fields, that's something i'm learning in the process of making pedals: i will make mistakes, things don't have to be perfect at the first try, making them right comes from trying, undoing them, and try again until it's less bad. eventually things will become good.
I really have the 'if i understand how to do it, it should be perfect' sickness. working on a cure at the moment.

Thanks for the book reference, i'll try to grab it! thanks for the link too, PRR!

Interesting insight and advices from you all, i'm waiting for components to start new projects, i'll try to envision the process with precision and see how it goes.
Seems that experience comes from experience ...

By the way, the help i receive on this forum is invaluable! thanks to you all.

[StephenGiles]

And always remember that you don't have to use a tiny box - if it's a big circuit, think rack!!

[boogietone]

What R.G. said.

Nonetheless, "Project Management" does not have to be quite a soul desiccating chore. Keeping decent notes, arranging your workspace, and thinking about your workflow from idea through to finished box are generally good ways to improve the next iteration. Widgets come off an assembly line at a rapid click only after being thoroughly thrashed about in someone's garage for untold hours and endless frustration.

[Electron Tornado]

There's some great advice in the above posts.

I'll also stress keeping good notes - organized, detailed, legible. That will be a big help on your current project, as well as any future projects. Another benefit is that if you have to step away from a project for a period of time, your notes will help you pick up where you left off.

Every project is an opportunity to learn.

[Minion]

After building building maybe 8 guitar amps and about 8 pedals and 20 mic preamps and a couple EQ"s and other assorted electronic gagetry I have come to the conclusion that I suck at planning so i don"t even try anymore ......

I just figure out what the first step is and do that step and then figure out my next step , If I plan ahead and get all the parts together and schetch out every step I usually end up having the augment my plan because of unseen difficulties which causes me to spend more money and time than it should .......

Being able to change your plans and being flexable with your project is is for me the most important thing that I do that leads to a successfull project ......