Why unfinished projects?

[RoGeorge]

Some projects will stay forever in the 'bucket list', some will be started but stuck on the 'to be finished' phase, some will be silently dropped or forgotten, some projects will fail, and some will be finished.

What are the root causes for not finishing a started project? (if you are the type who use to always finish your projects, please explain how to do that)

[Messtechniker]

Basic problem solved - loss of further interest because of no specific application
for the solution at the moment. Solution may come handy sometimes several
years (or even decades!) later. For this, proper documentation and labelling
of the protoboard is the key.

[Rerouter]

For me, there is a few reasons,

A number of my projects are "because I can", Building something interesting, to learn all the quirks and tricks involved with such a task, e.g. I've never done BGA, someone posted a Kicad project archive asking how to escape
 a 12x12 grid one, well I spent 30 minutes and broke it out on 3 layers, respecting grounds,

So now I know how to escape BGA's respect high edge rate return currents, and the fun of 4/4 spacing

Many others get sections ripped out and reused, I've built a few diagnostic / testers over the years, and they are full of reusable, refined and idiotproof chunks,

Others get skipped because I learned enough to discover i was fighting an X / Y problem, or lacked the correct search terms to find a cheap and well made solution.

And a few are simply access to IP preservation, working with some really handy piece of electronics, So I sit down and replicate it in case they later decide to stop selling them, This has become more important recently as a number of small one man band operations have stopped due to old age, (Some of the best RF wizardry I knew of was pumped out by a 97 year old who still hand assembled things up until march when he fell and was whisked away to a home)

[Mr. Scram]

Basic problem solved - loss of further interest because of no specific application
for the solution at the moment. Solution may come handy sometimes several
years (or even decades!) later. For this, proper documentation and labelling
of the protoboard is the key.

I've noticed that's a huge factor. As soon as I've figured out how to do it and am confident enough it'll work and I can pull it off, my interest level drops significantly. I like a challenge, but I don't really like menial tasks. When it's my job I take pride in also doing the latter properly, but when it's my own project that isn't really a factor.

[jm_araujo]

For me the root cause for not finishing is getting distracted by the next project (which should be an option!).
It's mostly because of the ninety-ninety rule: https://en.wikipedia.org/wiki/Ninety-ninety_rule . I find the second 90% pretty boring and get distracted by the next big shiny project easily.

[ebastler]

For me the root cause for not finishing is getting distracted by the next project (which should be an option!).

+1. You could argue that it is a special case of "loss of interest" -- something else becomes more interesting, at least for the time being. I have pushed projects on the back burner for two years, and then revisited (and actually completed) them later.

[Conrad Hoffman]

Wow, all of the above! Some I can't even explain. For example, I bought parts to build the Poor Man's Spectrum analyzer at the Dayton hamfest decades ago. I built the boards, temp wired them together and everything worked perfectly. I did the chassis work, the paint and mounted everything. Everything is there, ten-turn pots, BNCs, a meter and power supply. All I have to do is the fairly simple chassis wiring to connect everything. It's been on the shelf for 20 years or so, can't tell you why. I did finish my Cordell THD analyzer long ago, but that one took me several years. Just collecting the parts was half the battle, mostly the complicated rotary switches.

[hans]

First of all? What's "finished"? Is that an usable board, or one that is completely tidied up and that you could take to a job interview to show you can do product design?
For me it's the latter. And it usually gets stuck on firmware and polishing.

I build a board, bring up the hardware, write an alpha firmware and fix all botches made.
Then I quickly lose interest. I'm often interested in making concepts work, I think this is where I get most satisfaction out of it.

When I have a real usage for a project, the boards often stay in a patched up state. Although I'm fully aware that most stable hardware designs have to go through 2+ revisions, I don't see the point in spending money twice on extra PCBs, components just so I can avoid a few fixes. For most of the designs I'm doing these won't add functionality or improve performance.

It's hobby after all. Although I know it's a job for many people, but I see the art of "design for manufacture" as mostly busy work.

[TomS_]

Ive been going through this a bit myself this year.

Late last year I started working on a project, and was fully invested in all of my spare time. I made significant progress on the software side, and designed Rev A of a lot of the hardware. Earlier this year (where "its almost done" is how I would describe that project) a new interest caught my attention and now I am more or less fully invested in that instead, with the other project tantalisingly close to "ready for testing".

I still want to complete my other project though, I suspect maybe this second project is a bit of a distraction - its one of those curiosities that you develop, and just have to investigate and get out of your system before you return to your other project.

Prior to this I did some other projects which I saw through from start to finish serially.

There are other projects I would like to do, but they arent nearly as interesting as my "current two".

[TheUnnamedNewbie]

For me there is often also a "no longer need it". I would start a project because I want the result of something (eg, powersupply) but then before it is finished get something that works and beats my own thing in every way. Exampleso of this were my power supply (suddenly found myself owning 5 supplies), milliohm meter (got a 6.5 digit meter, does the trick also), voltage source (got a good deal on a working source), the list goes on.

Also, once the interest to do something wears of, I kinda forget about it. I will remember it at times I can't work on it, but when I do hve time I start thinking and wanting to work on other things.

[Mr. Scram]

For me there is often also a "no longer need it". I would start a project because I want the result of something (eg, powersupply) but then before it is finished get something that works and beats my own thing in every way. Exampleso of this were my power supply (suddenly found myself owning 5 supplies), milliohm meter (got a 6.5 digit meter, does the trick also), voltage source (got a good deal on a working source), the list goes on.

Also, once the interest to do something wears of, I kinda forget about it. I will remember it at times I can't work on it, but when I do hve time I start thinking and wanting to work on other things.

I've stopped trying to build things that I can buy. I need to undervalue my time ridiculously to make that an interesting proposition. Of course, doing a project for the sake of it is still viable or when something is simply unaffordable.

[SiliconWizard]

Basic problem solved - loss of further interest because of no specific application
for the solution at the moment.

I'd second that. This is a very common cause of abandoned projects for engineers. You have a problem at hand that you are challenged to (or challenge yourself to) solve, and this is what gets you interested. Once the problem is solved, you just move on. You have built a "brick" of knowledge or technology that you can reuse later on.

Another common cause is procrastination. It's very pervasive. There's a lot of causes for this, one being loss of interest, so kinda back to the first point.

Yet another cause is the "going from prototype to product" syndrom. You have prototyped something that works. Now to make it a finished project, you need all the finishing touches. You then realize that after all your efforts and what seemed hard work to get to the working prototype, there's even more work ahead to finish it. You didn't expect that, and the finishing part doesn't look very attractive compared to the initial phase, so you give up.

[TheUnnamedNewbie]

Yet another cause is the "going from prototype to product" syndrom. You have prototyped something that works. Now to make it a finished project, you need all the finishing touches. You then realize that after all your efforts and what seemed hard work to get to the working prototype, there's even more work ahead to finish it. You didn't expect that, and the finishing part doesn't look very attractive compared to the initial phase, so you give up.

I think this is related to the fact that some (I think, don't know how common this is) don't really care about the end goal but care about the joy of designing a thing, and then mainly the fun of the electronics. Once you get past that point of working prototype, most of the hard electronics design work will be done and it's now, as you said, a matter of finishing things off. That is just not as fun, hence we abandon the project.

[DaJMasta]

Usually it's another project pushing it to the side, or a combination of some sort of time sink outside of projects that makes me sort of get out of the habit of doing it and then a new one crops up.  Not really lost interest as much as it is getting excited by the next thing.

I also enjoy the hardware part much more than any software part, so I've got a few projects that are basically at that point.



That said, I do come back to them, so they sort of get shelved for a while and then I cycle back.  Many of my repair projects follow the same general trend, too.

[kony]

Add burnouts to the poll. Best part being one does not ever recover fully, despite the reconvalescence taking easily over year. 

[nuno]

Add burnouts to the poll. Best part being one does not ever recover fully, despite the reconvalescence taking easily over year. 

+1!

And maybe turn the poll into multiple choice?

[b_force]

For hobby projects it's always a combination of all.

[JS]

Hard to get components here, now might be getting better and easier to import but still a pain waiting 2 months and paying a lot extra for a component.

Missing components is my main reason I guess.

JS

[Mr.B]

Most of my unfinished hobby projects are due to my lack of ability with enclosures.
My latest unfinished project is a nixie clock controlled by an ESP8266, fetching time from an NTP server, automatically correcting for NZDST.
Did the design.
Layup the PCB.
Blank PCBs fabricated in China.
Assemble PCBs.
Write test code.
Tested - all OK.
Write final code.
Working clock - A thing of beauty.
Project stops...

I really want a nice wooden enclosure, but I don't have the tools to make a professional looking one.
The project will sit in its box until I finally resign myself to the fact that it will never have a wooden case and to just get on with a 3D printed case.

This sort of thing happens to about 33% of my projects.
Projects that do get finished are usually because I have deliberately designed them to fit a COTS enclosure.
Then they turn out looking a bit industrial...

[thermistor-guy]

Some projects will stay forever in the 'bucket list', some will be started but stuck on the 'to be finished' phase, some will be silently dropped or forgotten, some projects will fail, and some will be finished.

What are the root causes for not finishing a started project? (if you are the type who use to always finish your projects, please explain how to do that)

Failing to stay focussed and to persevere.

My next few projects are research projects, on sensors and calibration. If my current project works out, it should generate three papers (say, IEEE standard) immediately, and a more complicated fourth paper after those. IF it works. My intuition outpaces my math. It's taken me a few years to teach myself the math I needed, so I could prove the main theoretical result that will underpin the first paper. Until recently it was just a guess. Now it's looking more solid, but it's still not a sure thing. My next major subtask is to get the experimental data to verify the theory.

There's math to learn, theorems to prove, test setups to build, experimental technique to validate, software to write. Software for  test instruments, for data logging, for data extraction and analysis, for optimization, for simulations. All of this is in progress. There is a mass of detail to take care of, and it's easy to feel that the goal is just too far away, when actually it's not.

I've learned to be more persistent. Going to the gym has helped me with that. Eat, sleep, train, repeat - with discipline and insight, and regardless of mood, easing back where necessary to avoid injury. The results come, not immediately, but they do come. This is what success feels like.

Projects = brain gym.

[@rt]

There aren’t any. I’m only allowed to work on two at a time, so if I want to do something new, one of the current ones have to be complete.

[Mr. Scram]

There aren’t any. I’m only allowed to work on two at a time, so if I want to do something new, one of the current ones have to be complete.

Who is allowing this?

[@rt]

Myself and it’s been that way for more than a decade.

[Mr. Scram]

Myself and it’s been that way for more than a decade.

What happens when you're stuck with two projects you don't care about any more?

[IDEngineer]

I think this is related to the fact that some (I think, don't know how common this is) don't really care about the end goal but care about the joy of designing a thing, and then mainly the fun of the electronics. Once you get past that point of working prototype, most of the hard electronics design work will be done and it's now, as you said, a matter of finishing things off. That is just not as fun, hence we abandon the project.

I thought a lot about this "unfinished project" problem a while back and came to roughly this same conclusion, but I generalized it to "It's just not creative enough".

One of the things I've realized about myself is that creativity is a huge part of the attraction. Boring, repetitive, mundane tasks are almost physically painful. I can feel my IQ dropping in real time. So I'm attracted to problems that require creativity to solve, and once the creative part is done my interest wanes. That doesn't mean I drop the project... indeed, my work requires me to do everything from the initial concept to the production fixtures and tooling and documentation. But once the technical challenges of the product itself are resolved, I'm actually attracted more to things like designing an efficient calibration fixture for production of the product than finishing the actual product itself!

The good news is that I recognized this in myself a long time ago, and developed the discipline to overcome the "next shiny object" syndrome. That's not to say I don't have unfinished projects, though. For all but the physically largest projects I get a new banker's box, mark it with the name of the project, and keep all of the associated components inside. This prevents components getting mixed between projects, and makes it easy to resume a project since everything is in one place. It also gives me a visual indication of how many unfinished projects I have... there's a sizeable stack of banker's boxes in view as I type this. 

[basinstreetdesign]

It's all on me.  I like to finish something when I start it.  Having said that ... a project may stop when I cannot figure out some aspect of it.  A case in point: I wanted a library of system building blocks for my OrCad Spice simulation tool so that I could quickly test an idea without having to do detailed design on every functional block.  I couldn't figure out how OrCad implements Spice linkage to models for a generic filter.  Spice help no help.  No on-line help.  No forum help.  So it just sat there.  And sat there.   

[RoGeorge]

developed the discipline to overcome the "next shiny object" syndrome

Very interesting, can you give more details about that, please?

[IDEngineer]

developed the discipline to overcome the "next shiny object" syndrome

Very interesting, can you give more details about that, please?

Umm... the short version is that my income dries up if I don't finish projects. 

I also don't let myself start playing with the next cool project until the present one is finished to the extent possible at the time. I will jot down notes, do some basic research so the ideas can be percolating in my head, but none of the hands-on fun stuff. It's just too distracting.

My son once asked me if I could be happy as a Professor at a university. My response: "No, it's not creative enough." Unless you're doing NEW things frequently, the brain starts to atrophy. Sometimes I get angry when a project runs into trouble, but on the other side of the problem I realize that the research I had to do to get there has increased my personal skill set.

[@rt]

Myself and it’s been that way for more than a decade.

What happens when you're stuck with two projects you don't care about any more?

That wouldn’t happen. Result driven motivation is natural for me. It’s more about not becoming focussed on too many things, and also being a hobby, not having a hobby budget spread wide enough that budget is a bottleneck. I am definitely a finisher, so it at least has worked for me.

[@rt]

Looking at the whole thread, hell no. If there’s a clear objective that preferably contained some fair, manageable learning curve, that’s a journey, otherwise that’s just some fairy shit. There was journey if there was a goal maybe, regardless of outcome. Anyway, that would explain not getting something done. It didn’t matter, or goal was unreasonable, that’s why. Like anything else.. going to the gym, maintaining relationships, keeping your vehicle in check, etc.

[CatalinaWOW]

All of the above, plus competition with other projects getting started.

I can truly say that a bare majority of my projects do get finished, though some take many years, and some have morphed greatly on the way.

[TerraHertz]

Unfinished projects? This is my area of expertise!

With me the main factors are:
* External priority interrupt happened. Some urgent time demand (or a series of them) that can take months, and by the time that's done a project that was going well before is now put away, buried, lost interest, etc. This has been a big factor for me in the last 6 months.

* Unexpected difficulty with some detail of the project. Current 'trivial' project as example. I thought I'd just quickly convert an old LCD monitor to a small white-light panel. (Which I do actually need for something.)  Remove LCD glass, hack the power supply and CCFL driver board to just run at power-on, put a sheet of plain glass to replace the LCD. Thought it would take two days max. Turned out the power board has multiple interlocks, and getting it to start reliably without the original CPU/graphics card, was a pain. Traced out the whole circuit, but the ccfl switchmode control IC startup and fault shut-down nuances are a bit hard to understand. So annoying! The project got put aside, and is in danger of being permanently unfinished, due to other chores pushing it down the queue. Trying to make myself re-attack it currently.

* Project dependency trees, getting too deep and complicated. Wedging of any one detail jams up the entire tree. Also spinoff projects do the same.
  For eg:
  My vacuum chamber project (part of a much larger project, that's taking decades.)
  - Need to machine a large diameter aluminum ring, as an adapter on the chamber. (And other future similar parts needed.)
   - So need ability to make it/them myself.
    - Current lathe too small to do it, so buy a bigger lathe. (Became a multi-stage saga, but now achieved.)
     - To set up the big lathe, my 'forge shed' needs to be finished (big swing doors, benches, shelving, etc.)
      - That shed construction is stalled due to a problem I hit with welding large, unusual hinges for the swing doors.
       - Need to be able to MIG weld the stainless steel hinge tubes, without distorting them.
        - For a long time I couldn't figure out how to do that. Now I do have an idea, and recently got a needed part.
          Now actually doing it is getting close to top of queue.
          But there's also _another_ building project in-progress and urgent - an extra storage area 'lean too' - that's
          stalled due to running out of money to buy the wall exterior cladding. Not for much longer I hope.

  (That top level project also has a spinoff - wanting to learn how to do color photo/printed anodizing of aluminum control fascia panels. Just because I want it to look cool. Another whole list of awkward.)

* Emotional aversion. A project that you want and need to do, and should be doable, but some component of it really, really gives you the shits. And so you keep putting it off. In my case, couple of examples are:
  - A major writing project. To do with legal matters; prep for a potential legal case. But there are aspects of it that I find very unpleasant to even think about. So it's extremely difficult to force myself to sit down and work on it. I can write voluminously when needed, but not on _this_ one.

 - Any technical project that involves dealing with PC OS quirks. In particular MS Windows, and problems in the class of 'there is no way you will ever understand exactly why this shit is happening, and any solution is likely to just stop working for no apparent reason. With all hours of effort to that point potentially wasted. And if you do even hope to understand, it will involve hundreds of hours of Net-hunting-flailing, with no guarantee of any useful result. And anything you do learn will be obsolete in a year. And/or you will have to accept conditions that you find morally completely unacceptable.

I have another whole project tree stalled due to that last one.
 * A project hacking a simple, cheap document scanner to serve as an 'edge scanner', for doing large books.
    That involves some quite fine micro-surgery on the sensor array (chopping the end off, right at the end of the actual sensor.)
   - So I want to take pics of that process, for a writeup. I have good microscopes, but lacked a good microscope camera.
    - Buy a nice hi-res camera (done)
     - It's USB3, and my workhorse PCs are all WinXP (hence no USB3 support.)
      - Repair and option up a big 1 TB Raid array machine, to use as USB3 video server. (done)
       - Needs Win7. But due to learned absolute distrust of MS, I refuse to install standard Win 7 for long term use. (don't even mention Win8/10.)
        - So need an NT-Lite customized version os Win7. Have been meaning to learn how to do that for a while. Buy the NTLite tools. (done)
         - And need a working Win7 machine to run those tools on. Fine, I have plenty of spare PCs. Set one up, install the hated plain Win-7.
          - Get to the 'online authentication' shit. (One of many MS bullshits I flatly refuse to accept/suffer.)
            - Find a means of killing that. Install. Learn how to use it.
             .... and that's where I'm up to with that. Haven't turned the machine on for months. Partly due to other time demands and projects, but also due to emotional aversion. In a sane world, this crap would not be necessary.
          This machine will also be the host for my 3D printer, so that project is in limbo for the same reasons.


* The 'opportunity too good to miss' effect. Where you are going great on some project, then some unrelated opportunity comes up that you simply cannot pass up, but to take it will involve quite a lot of work. So all current projects get pushed down on the stack, while the 'new thing' takes over all your time. For eg recently I was going great on a 'vacuum hold-down hotplate' construction, then had to divert attention to picking up a large 3-phase 415V 30KW UPS from 600Km away. And associated conflicts with the auction seller, who decided they wanted an extra $960 after (due to their oversight) I won it in auction for $1. All sorted now, but that hotplate project... soon to be resumed, I hope.  http://everist.org/NobLog/20180312_hotplate_that_sucks.htm

* Project ideas, that never even get a whiff of starting. Like I have a blank 1U panel in the small instrument rack on my desk, that I want to build some useful things into. I have a list of ideas to include. Nothing all that hard. Just never have time. http://everist.org/NobLog/20160228_bench_rack.htm

* The 'can't find needed data/schematics', and 'totally unexpected disaster' effects. I have a nice HP 1000 minicomputer, that I started a refurbish project on. Got to the point of wanting to test the power supply, before trying a power up with it attached to the system. Turns out the schematics for this late-model unit power supply are not online. Got in contact with one HP antiques archivist who might have it. Jon Johnston (sp?) He was overseas on a climbing expedition. I left it for a while. When I next inquired, found he'd returned, then gone off on another climbing trip, and perished in an accident. Don't know who has his collection.
 
* The ridiculous 'can't find one stupid little essential part' effect.
I have a HP 5061A atomic clock. Sans the power cord, and the rear panel power connector is an unusual 3 pin round military screw-in thing. Any I find on ebay are too expensive for my budget vs the value of trying the machine. Which is probably well past use-by date on the physics package tube, but I'd like to try it. Because it's a historical piece I don't want to just hack in a standard IEC socket. Also don't want to nastily run it with some improvised kludge. But can't find a genuine matching power connector. So it just sits on the 'dodo' bench.
(That project also needs an accessory for degaussing the physics tube, so that's a second bit of unobtainium.)

* Getting old, and failing concentration due to stress/life issues.
 I have several software projects, that in younger years would have been only mild challenges, and very enjoyable. But I find that now I can't concentrate like I used to, so the necessary learning curves and spending time dedicated to the projects, are much more difficult. Not helped by all the other urgent chores and projects.

* Unproductive time sinks. Eevblog being one of them, but far from the worst. Mainly the state of the world is so 'edge of my seat' at the moment, I can't resist spending a major part of each day online, just keeping up with developments.

[james_s]

I keep a mains cord with alligator clips around for just such occasions. If you don't want to do that you could remove the original socket without actually modifying anything and just feed a cord through the hole.

[ElektroQuark]

In my case, waiting for components to arrive always ends with starting another project from the TODO list.

[TerraHertz]

I keep a mains cord with alligator clips around for just such occasions. If you don't want to do that you could remove the original socket without actually modifying anything and just feed a cord through the hole.

I have exactly that same mains cord (only with a double pole switch in-line.) Don't want to use it for fear of a short in the socket, damaging the pins. It's a very rare, cool historical instrument, high sentimental value too. So, no kludges.

I can't remove the original connector without damaging the historical accuracy, since it's riveted in. Could run a cable from inside. But I'd much rather just find the right connector. I'm considering making one - have pins that will fit, can machine a shroud, and just need to find a way to make an insulator insert to hold the pins. Maybe of epoxy? But need to get the pin positions exactly right, and there's a keyway. So really needs to be molded in-place. With some form of release, so don't get epoxy on the original socket & pins.
It's totally not urgent, as I'm busy with other things. Every now and then I look for a suitable connector. It can stay that way till I find one.

Incidentally, to illustrate 'too many projects: pic is a stack of papers that is my 'to do' list.
Generally they are line items, many different chores per page.
Below the yellow post-it tag, is older sheets that are a mix of random stuff to do, only some sorted by topic or project. Some items are crossed off, but most not.
Above the tag, are new sheets that I wrote out recently after finishing the UPS-fetch saga. All sorted by class of item, different projects, etc. Most of those sheets are pretty full, very few items already crossed off. There's _some_ duplication of tasks there with ones on the lower sheets.
Sometimes I think I should just give up, spend the rest of my life bushwalking or something.

[BravoV]

As a hobbyist, in the old days, when only one new project maybe for months or even year, it was done until completely finished.

Now I'm too spoiled with too many "new" stuffs coming. As the rate of incoming new projects either planned or well prepared that made old ones sort of half cooked and sometimes left abandoned. 

[james_s]

You could put wires on the pins and then cover them in heatshrink so they don't short. I've also used hot melt glue to make custom connectors for temporary use.

[RoGeorge]

* Project dependency trees, getting too deep and complicated. Wedging of any one detail jams up the entire tree. Also spinoff projects do the same.
  For eg:
  My vacuum chamber project (part of a much larger project, that's taking decades.)
  - Need to machine a large diameter aluminum ring, as an adapter on the chamber. (And other future similar parts needed.)
   - So need ability to make it/them myself.
    - Current lathe too small to do it, so buy a bigger lathe. (Became a multi-stage saga, but now achieved.)
     - To set up the big lathe, my 'forge shed' needs to be finished (big swing doors, benches, shelving, etc.)
      - That shed construction is stalled due to a problem I hit with welding large, unusual hinges for the swing doors.
       - Need to be able to MIG weld the stainless steel hinge tubes, without distorting them.
        - For a long time I couldn't figure out how to do that. Now I do have an idea, and recently got a needed part.
          Now actually doing it is getting close to top of queue.
          But there's also _another_ building project in-progress and urgent - an extra storage area 'lean too' - that's
          stalled due to running out of money to buy the wall exterior cladding. Not for much longer I hope.

Sometimes long dependencies chains happen to me, too. I noticed that the length of the chain can be severely shorted, or even completely eliminated if I drop some of the hidden constraints. I call them 'hidden constraints' because sometimes I am not aware those constraints are in effect until the constraints start to severely impact the project.

In the above example, the hidden constraint might be "if I have the skills, then I must do it all by myself, with local resources". To brake the dependency chain, some alternatives will be to pay somebody else to do the large diameter Al ring, or to rent some time on a bigger lathe.

In the second dependency chain, the one with the power cord, the hidden assumption might be "it must be impeccable", as in the work (here the restoration) must be as perfect as possible.

I don't know for others, but for me those two constraints left me with a lot of unfinished projects.

[TerraHertz]

Sometimes long dependencies chains happen to me, too. I noticed that the length of the chain can be severely shorted, or even completely eliminated if I drop some of the hidden constraints. I call them 'hidden constraints' because sometimes I am not aware those constraints are in effect until the constraints start to severely impact the project.

In the above example, the hidden constraint might be "if I have the skills, then I must do it all by myself, with local resources". To brake the dependency chain, some alternatives will be to pay somebody else to do the large diameter Al ring, or to rent some time on a bigger lathe.

Sure. In this case I went with buying a bigger lathe, because:
 * I've always wanted one. Size of my small lathe is often a frustrating constraint.
 * There are actually TWO large diameter adapters needed, and it's an experimental rig so highly likely I'll need to modify them as system development proceeds.
 * There are likely to be other large things needed for the system in future. Unknowable.
 * Cost of the whole lathe ($900 + $300 to transport it) is not much above likely cost just to get one such ring made.
 * It's cool. Older than me! A fun restoration project in itself.

But you are right. For instance my wish to learn and set up to do color printed anodizing, just for cool looks of a project, is maybe too much. I may reconsider that. It's in the path atm, since I don't yet urgently need the panels finished. The main obstacle is that my workshop 'kitchen' where the anodizing rig would go, is still stacked up with boxes. Waiting for that storage space...

'Renting time on a lathe' - never heard of such a thing. I can imagine any commercial company would simply refuse, for liability reasons. And I don't know anyone else with a large lathe.

In the second dependency chain, the one with the power cord, the hidden assumption might be "it must be impeccable", as in the work (here the restoration) must be as perfect as possible.

I'm not normally so picky. In this case the high 'antique' value of the gear in original condition, weighs in with no great hurry to proceed, and small chance it may ever be workable. The tube has low hours of use, but they also shelf-age.

[tpowell1830]

Unfinished projects? This is my area of expertise!

With me the main factors are:
*etc....

Very well written and postulated. I find all of this in my projects as well, I call it "the circle of life".  One project depends on another which circles back to a previous and finally all boils down to lack of funds...

[palpurul]

In my case there are many reasons just like everyone else. Before I start a project I try to figure out if it's realizable. I browse digikey to look for available and cost effective parts. If it's obvious that I am going to need very exotic parts, I don't do it I just move on. When things look somewhat realizable I start doing it.

The first reason is obviously TIME and COST
I always underestimate the time it takes to solve small, unglamorous problems. I always focus on the main thing that makes the heart of the project and I end up realizing that I actually figured out how to solve the easiest problem about the project. It's amazing to find out sometimes pesty capacitor selection more important than opamp or any active component selection. These little things are sometimes easy to overlook and it takes time to fix and requires design iteration, hence more time and money.

The second reason is that it takes me some time to realize that there is not much point doing it . Most of my ideas are rubbish I don't want to accept that it is rubbish at first because it gets me excited and I just want to do it, and after some time excitement fades away, I become honest to myself and accept that it's a rubbish idea. These ideas are like challenges rather that projects, a different and unnecessarily difficult way to do something  . The moral of the story is be honest to yourself don't get excited easily. 

[dkonigs]

I've stopped trying to build things that I can buy. I need to undervalue my time ridiculously to make that an interesting proposition. Of course, doing a project for the sake of it is still viable or when something is simply unaffordable.

When I was getting started, this was a big factor. Nearly everything I could build was something I could also buy, far cheaper, and far better. Then I stumbled upon the one big exception: Mobile robotics!

On one hand, mobile robotics provides a great platform for a seemingly endless list of sub-projects.
On the other hand, this virtually guarantees you'll never finish and eventually get bored with it.

So I said "No more mobile robot projects! I need something I can actually finish!". (And in exchange, I now have a very difficult time actually coming up with project ideas I'm motivated to do.)

[james_s]

One factor though is you can gain a lot of education and satisfaction by building it yourself. Even if you can readily buy something, you can't say "I built that myself!" about it. If building it won't teach you anything new though that sways things back toward the buy it route.

[RoGeorge]

Random thought: Unfinished projects might happen because we perceive the world on a logarithmic scale.
(Weber's Law - https://www.britannica.com/science/Webers-law )

How is that?
From the poll, one of the main cause for unfinished projects is "loosing interest". Loosing interest is the same as "not rewarding any more". Reward is proportional with the perceived progress. For the same amount of work (and a steady progress), our logarithmic perception will need more and more work to achieve the same reward. This is not sustainable for long projects. For longer projects, all we can do is put constant work. But constant work (so constant progress with the project) will be perceived as less and less rewarding because of the logarithmic perception. See the chart at 04:34, where for a constant delta x (x=effort=progress) the y become smaller and smaller (y=perceived_progress=rewarding=interesting).



The same mechanism could be used to explain or justify all kind of things, like: why some get bored in marriage, why some become addicted, or even why immortality would be "dead boring". 

[zitt]

Nuff said:

Simply put; I get it working good enough for a prototype.
There has to be a light at the end of the tunnel for me to get it to 100%. Products that end up at Pinball-Mods.com/oscom are a good example of light at the end of the tunnel.

[james_s]

I have quite a few "delayed finish" projects. I'll start something, work on it for a while and then either get bored, hit a point where I get stuck, or get distracted by some other project. Many times I eventually get back around to it, pull out a project I shelved several years earlier and finish it up. The funny part is often something I was stuck on turns out to be very easy the next time I have a go.

[CopperCone]

$$$$$$$$$

[@rt]

Does that curve map to your experience though? It doesn’t for me.
I don’t often enjoy starting something, most enjoy seeing things coming together and beginning to work (software or hardware),
and begin to lose interest rapidly once whatever the objective was in the first place is actually happening.

[Neukyhm]

Why unfinished projects?

Well it's not my fault that the things I make tend to explode when turning them on 

[Teledog]

Usually start projects assuming that I have all the parts on hand ..then finding out..not!
If it's imperative that it's done ASAP, then it's Digi-Key, if not, then the 1-3 month fleabay wait.
Bag the parts in a zip-lock bag along with all the documentation & put it in the TODO/archive tote bin..along side its brethren

[thermistor-guy]

... The funny part is often something I was stuck on turns out to be very easy the next time I have a go.

I wouldn't say "very easy" but yes, this often happens to me, whether it's a practical, theoretical, or creative task. It's very much the case with advanced maths: the first time I study a new topic, it's one painful struggle, and I still don't quite get it. The second time I study it, it starts to gel.

I used to exploit this as a student: I would read ahead on the assigned material, and feel lost. But during the lectures, on my second exposure to it, the material would start to make sense, and I could see where the lecturer was heading. That first exposure, as bewildering as it felt, was nevertheless essential.

In Tae Kwon Do there is a similar phenomenon (so I'm told). When you first reach black belt level as a student, you have followed what you have been taught, but have yet to gain a deep gut-level understanding. Then as a black belt, you teach yourself Tae Kwon Do again, from the basics, but this time you understand what you are doing and why. Mastery seems to require learning it twice: the first time from instructors, the second time from yourself.

[james_s]

Usually start projects assuming that I have all the parts on hand ..then finding out..not!
If it's imperative that it's done ASAP, then it's Digi-Key, if not, then the 1-3 month fleabay wait.
Bag the parts in a zip-lock bag along with all the documentation & put it in the TODO/archive tote bin..along side its brethren

I always seem to miss one resistor value or some other dumb thing when I'm ordering parts, then I spend hours looking through my stashes for a 2 cent part.

[Jwillis]

I have several projects going on at once .Generally they get finished as the parts become available.If something goes unfinished it's because it has served its purpose, a better design is used or a catastrophic failure. I find my failures as enlightening as the successes. 

[dkonigs]

Usually start projects assuming that I have all the parts on hand ..then finding out..not!
If it's imperative that it's done ASAP, then it's Digi-Key, if not, then the 1-3 month fleabay wait.
Bag the parts in a zip-lock bag along with all the documentation & put it in the TODO/archive tote bin..along side its brethren

I always seem to miss one resistor value or some other dumb thing when I'm ordering parts, then I spend hours looking through my stashes for a 2 cent part.

Once upon a time I wasted way too much money going to Radio Shack and giving them 25 cents (multiplied by the size of the package) for that part, because I didn't want to pay shipping and/or wait another week.

Unfortunately, not so feasible of an option today.

[AngraMelo]

The enclosure part drives me nuts.

Although Im still a beginner and I dont have any particular interesting designs of my own (mostly copies of other people's). I breadboard it, do some measurements to see if it will fit for my application, make the PCB (usually a crude one, etched at home with peroxide and drilled by hand in a unbelievable shaky way), populate it, wire everything up, separate the enclosure aaaaaaand Im already looking at the next project.

I really dislike the process of making it visually appealing. I dont know why.

[RoGeorge]

One way to avoid this might be to start by building the enclosure first.
Once the enclosure is ready, start building the electronics.

[Undweeber]

My project got to a point where it needed to get significantly more complicated. I am okay with that as I wanted it to be the best possible, but then I got bored I do come back to tinker with it though, and I do plan on continuing.

[Bendba]

I would say that I have different kinds of unfinished projects.


As a hobbyist, some projects didn't keep up with the learning curve, what seemed to be a challenging project turned into something trivial half way through the project and I lost interest.
Or only one part of the project was a challenge but the rest is basic, problem solved, moving on and the basic parts never get done.


Some project are more "not yet finished" rather then unfinished. Usually because of financial constrains, cannot afford the parts to finish the project, yet. Usually it is enclosure, potentiometer's and knobs. I have over a dozen board perfectly working with trimmer pots where I should have a nice ten turn pot, and with dodgy connectors. Some get used on regular basis.
One day, I'll put them in a nice enclosure (I started keeping old DVD players, planning on using them as enclosure. They are easily stackable, often have already enough buttons on the front panel and sometimes a usable display.)

And then there are some "couldn't be bothered finishing" project that I started when I was lacking really interesting projects to build, and then I found something that is much better, more exciting to build and left the old project as it was. I wouldn't call them waste of time as I still learned something from them.

[james_s]

I seem to have trouble with the enclosure part too. Either I can't find an enclosure I like well enough, or I just don't seem to get around to doing all the required cutting and drilling. I must have at least a dozen projects floating around that are finished and fully functional devices simply lacking a nice tidy box to go in. Part of the problem is that I'm picky about the enclosure and want my projects to look really clean and professional so I can occasionally be paralyzed by indecision.

[dkonigs]

Not having a way to build the "physical part" of my projects (e.g. enclosures) is why I finally broke down and bought myself a 3D printer about 6 months ago. Sure, it may not be the best way to make a professional-looking "plain old box", but it really helps and really opens up your options.

[james_s]

3D printers have exciting promise, but the hobby ones seem to have stalled at the point were they're ok for printing out trinkets but it takes some doing, either a lot of refinement or a lot of post-printing rework to get something that even approaches the quality of a low end injection molded enclosure. A couple of my friends have them and they seem to spend a lot more time tinkering with the printer than printing useful objects.

[JS]

Cases, always cases. Mechanical part of the projects are the more boring, I've ton of working PCBs that never were enough for a case or I just forgot about them.

  I'm considering a 3D printer but not in my budget right now as I'm buying a lot of stuff for the lab. Just got an scope, soldering iron, a new DMM in the way, few builds to help around here, etc. Still in the need of a air soldering gun and some more test gear.

  Still the 3D printer, while really useful for many many things, doesn't help much for low noise applications where you want as much shielding as possible. Using internal shielding at the PCB could be helpful but still adding more mechanical problems to the mix.

JS

[threephase03]

Well, I have had this problem in the past. I decided to try something different I've started to record my progress on youtube - not fancy videos just quick shoot and uploads. I think it will push me to make progress. See how it goes. 

[SiliconWizard]

Still the 3D printer, while really useful for many many things, doesn't help much for low noise applications where you want as much shielding as possible.

You can always use adhesive copper foil tape and shield the 3D printed enclosure with it. (This kind of stuff: https://www.3m.com/3M/en_US/company-us/all-3m-products/~/3M-EMI-Copper-Foil-Shielding-Tape-1181/?N=5002385+3293719476&rt=rud )

For those not willing to invest in a 3D printer and having occasional needs, there's quite a few companies out there that can 3D print your design for a very reasonable price. That's what I've been doing for a while.

[IDEngineer]

One project depends on another which circles back to a previous and finally all boils down to lack of funds...

I've long told people that I either have the time to finish projects, OR the money to finish projects, but seldom both simultaneously.

[IDEngineer]

3D printers have exciting promise, but the hobby ones seem to have stalled at the point were they're ok for printing out trinkets but it takes some doing, either a lot of refinement or a lot of post-printing rework to get something that even approaches the quality of a low end injection molded enclosure. A couple of my friends have them and they seem to spend a lot more time tinkering with the printer than printing useful objects.

This is why, when we purchased a 3D printer, we bought a fully assembled and *supported* machine (in our case, a Lulzbot Mini), despite my son begging me to get one of the cheap eBay kit versions because he wanted the experience of assembling it. The Mini has generated literally hundreds of parts since we got it and the only maintenance has been the printhead fan, which required replacement a couple of months ago. Frankly, they made a poor choice of a tiny squirrel cage fan that appears designed to fail - I bought two so I'd have an extra. They've since revised the design to use a more reliable fan and I expect to upgrade at some point.

You need to go into 3D printing with realistic expectations:

1) Every filament is different (even the "same type" of plastic from different vendors). It takes a bit of experimentation to work out the best settings.

2) Every machine has its quirks. For example, the temperature sensors have reasonable repeatability but very poor absolute accuracy. 220C on one printhead could be 30C different from another, which takes you right back to point #1 above.

Neither of these points is unusual. Most machine shops have "the guy" you go to when working with a specific machine, a guy who has spent enough time with it to know the subtle details of getting its best performance. This is no different, other than it's YOU learning those details.

Even with these points, having a 3D printer has been life-altering. Much of my work involves a lot of custom mechanical design in addition to embedded hardware and firmware. With the 3D printer I can get started early on the mechanical parts instead of waiting on someone else to generate a prototype. I have done 5-6 revisions a day on a given part, tweaking in the details, and by the end of the day have it down to the point where a machine shop can begin cutting the injection mold - a process that in the past could take several slow and frustrating weeks. This has made me give serious consideration to one of those PCB prototyping machines... it's hard to overstate the value of very quickly iterating a design in near-real time, the rate of progress is almost an unfair advantage. It's made such a difference in the mechanical realm, I'm wondering what would happen in the hardware realm.

Don't dismiss the value of having a 3D printer around. Your design work will move MUCH faster. And it's handy afterhours too... my wife's minivan's glove compartment door latch stopped locking closed, and I was able to crank out an improved part with better reinforcement. Had it working in about an hour. Imagine what that would have cost had I taken it to the dealer, or ordered the OEM part(s). 

[Messtechniker]

Cases, always cases.

True enough. My way around this is to reuse the cases of dead/no longer used
 ATX power supplies after ripping out the innards. Saves a bit of mechanical work
since some cutouts can be reused, like the mains power inlets, or the earthing
arrangement for example. If any large fan cutouts are present I cover these with 3 mm
plastic sheets cut to size and reusing the former fan mounting holes as shown in
some examples below. All are oneoffs.

[JS]

Cases, always cases.

True enough. My way around this is to reuse the cases of dead/no longer used
 ATX power supplies after ripping out the innards. Saves a bit of mechanical work
since some cutouts can be reused, like the mains power inlets, or the earthing
arrangement for example. If any large fan cutouts are present I cover these with 3 mm
plastic sheets cut to size and reusing the former fan mounting holes as shown in
some examples below. All are oneoffs.

No doubt, I have a bunch I picked up from the university dumpster, this is on my bench right now getting a μSupply inside.

JS

JS

[CopperCone]

Lol I hated having to deal with fucking mechanical design so much that I got a buncha metal working equipment, so I can make nice custom welded enclosures AFTER I design a PCB.

It's alot of money and work to make a nice box but you get shielding and good ground behavior, otherwise you are left with tricking out altoids tins and trying to design into commercially available dimensions which is super annoying.

Still need a damn milling machine and lathe so I can make RF enclosures though, the other stuff just covers you for up to maybe VHF, but IMO its essential for things like power supply boxes, general instrument boxes, etc...

I just can't stand like exposed RF cans and some kind of fishing net made of coaxial cable out on peoples tables.

[CatalinaWOW]

While a mill or lathe can get some beautiful enclosures, you can get the performance without going that far.  Some combination of soldering, brazing and gasketing can RF seal your sheet metal boxes really well.  It is harder to make precise tuned cavities, but not outright impossible, and there is always the casting option.

[Sudo_apt-get_install_yum]

The problem I face when not finishing projects is a combination of different factors.
Most of the time it’s a result of lack of time, I start a project when there isn’t much to do  at work or in school and then a huge project starts at work and I work late every day and don’t have time to finish the project.

Once work slows down witch can be after 1 month or 6 months I loose interest in the project since most of my projects are started because I want to learn something new or because I have a need for the "product" I’m making I no longer have a real use for the project after such time.

I also remember a few years ago when I was a student with no money I used to buy almost all of my electronics from eBay and it took ~30 days to get the parts and if the things were wrong or I updated the design and changed components I had to wait another 30 days and with a lot of things in school it was easy to deem the project not worth it.