How do you find inspiration for new electronics projects?

[Xena E]

4GHz oscilloscope/TDR using 2N3904 class transistors in the signal path.

Oooh... that'd be some weird messed up distributed amplifier ...

 

Where's me bag of BC 107's

 

E

Surely you need to use OC44s???

Thanks!

Possibly cutting edge tech in 1957, but now sadly unavailable, and I don't have any 

[ebastler]

Every time I sit down at my bench I'm confident that inspiration is about to strike at any moment and I'll be embarking on something fun within a few minutes.  Hours later - disappointment  .

For me, project ideas typically are not generated at the bench, but come from outside, often "out of left field" -- either from other areas of my own interest, or from talking to others who point out an interesting topic or an unmet need they have.

E.g. something around the house ("I really want a remote sensor for that drainage shaft"), some technical oddity someone explained to me ("Did you know that there were computers which used acoustic delay lines as their main memory?"), a device someone wish they had ("A desktop display which shows my aging relative the events and appointments for today, which I can update remotely").

[coppice]

4GHz oscilloscope/TDR using 2N3904 class transistors in the signal path.

Oooh... that'd be some weird messed up distributed amplifier ...

 

Where's me bag of BC 107's

 

Not as interesting as that; far more prosaic Yes, BC107/177 are fine

It is permissible to have fast transistors in a trigger circuit, but TDR doesn't need that.

BC107 is an impressive date code to find on a transistor.

[wobbly]

Thank you to everyone who has responded to my topic so far, I really appreciate your time and advice.  I've started looking through some EE magazine archives (going back to before I was born!) and I've already earmarked some projects of interest. 

someone explained to me ("Did you know that there were computers which used acoustic delay lines as their main memory?")

Funny you should mention delay lines!

A few weeks ago I came up with an idea for a novelty project just for fun - a kinetic sculpture perhaps...  A delay line memory of sorts but with a delay in the region of seconds, rather than milliseconds.  And with rather low total storage volume; perhaps 64 bits to start with.

My idea was to construct a simple disc out of plywood, perhaps 12" in diameter.  The disc can be driven by something like a stepper motor via a belt-drive or a cog on the edge of the disc.
Near the edge of the disc, 64 evenly spaced holes are drilled of about 6mm each.  Into each hole is pressed a single glow-in-the-dark plastic BB-gun ball.

The idea being that each BB pellet would act as a single bit of memory - if it was glowing bright then that represents a 1-bit, if it was dark then a 0-bit.
The disc would spin continually and a REFRESH head would read the state of each passing BB and reinforce it's state by shining a bright UV LED on it (or not).  I was hoping the BBs would glow well for at least a few seconds before needing refreshing.

I got really excited by this idea and got as far as buying a few different varieties of BBs.  To my disappointment, I wasn't able to contrive a detector that was able to discriminate between a bright and a dark BB.  Even in a dark room.  Sounded simple, turned out to be really hard for me at my knowledge level - which isn't all that high, turns out.

[ebastler]

Funny you should mention delay lines!

A few weeks ago I came up with an idea for a novelty project just for fun - a kinetic sculpture perhaps...  A delay line memory of sorts but with a delay in the region of seconds, rather than milliseconds.  And with rather low total storage volume; perhaps 64 bits to start with.

My idea was to construct a simple disc out of plywood, perhaps 12" in diameter.  The disc can be driven by something like a stepper motor via a belt-drive or a cog on the edge of the disc.
Near the edge of the disc, 64 evenly spaced holes are drilled of about 6mm each.  Into each hole is pressed a single glow-in-the-dark plastic BB-gun ball.

The idea being that each BB pellet would act as a single bit of memory - if it was glowing bright then that represents a 1-bit, if it was dark then a 0-bit.
The disc would spin continually and a REFRESH head would read the state of each passing BB and reinforce it's state by shining a bright UV LED on it (or not).  I was hoping the BBs would glow well for at least a few seconds before needing refreshing.

I got really excited by this idea and got as far as buying a few different varieties of BBs.  To my disappointment, I wasn't able to contrive a detector that was able to discriminate between a bright and a dark BB.  Even in a dark room.  Sounded simple, turned out to be really hard for me at my knowledge level - which isn't all that high, turns out.

That is a really neat idea. I have been toying with the idea to build a series of demonstrators for various old memory technologies. (Already have a project name, "Memory Lane"  ) Thought about waves on ropes or "slinky" spirals to visualize acoustice delay lines, for example.

It should be possible to get the glow-in-the-dark bits to work. Probably needs some careful tweaking of wavelength, exposure time and intensity -- maybe a flash lamp or LED flash? I'll give that some though, thank you for the idea!

[jpanhalt]

@wobbly
I wonder whether those BB's would decay fast enough to get the effect you want.  Fluorescence, of course, decays much faster, but probably way too fast.  Lifetimes are usually in the ns range or shorted; although, some compounds in the milliseconds range have been developed (first set of references).

Phosphorescence is often too long, but there is a lot of work on shorter lifetimes (second set of references).  I agree that phosphorescence is a more likely way to get your project to work.  Neither reference for phosphorescence is particularly relevant, but they may give you some ideas of where to look.

Long fluorescence:
https://pubs.acs.org/doi/10.1021/cbmi.3c00091
https://pubs.acs.org/doi/10.1021/acscentsci.5b00327

Fast phosphorescence:
https://www.picoquant.com/applications/category/life-science/phosphorescence-lifetime-imaging-plim
https://link.springer.com/article/10.1007/s00214-023-03020-w

@ebastler
Unfortunately, I am old enough to remember the first memory calculator a professor in our physical chemistry section at university had (Dean Robinson).  It was acoustic based.  I suspect working examples of those machines might be expensive today (http://www.vintagecalculators.com/html/calculator_memory_technologies.html ).

[Kjelt]

If you are a natural curious person don't limit yourself just to electronics.
If I may suggest have a peek at mechanics as well, actuators, transducers, linear and  steppermotors, servo's, robotarms, sensors, vaj settings, it will open a new world and it is called electromechanical engineering aka mechatronics. Then it also allows you to come up with many new projects.
For my self, I get the most interesting new ideas when I am on a holiday outside of the country, no access to the lab. The moment I get bored , the magic inside my brains start to happen.

[nali]

I like the thinking about the disc - not sure how well it'd work in practice though as the dots would pretty much blur together with the rotation. It's similar to a science project for kids I'm involved in where one of the displays is a wall of luminous materiel and kids can "write" on it using a LED keyring torch.

If you're thinking along those lines meybe try one of those "propeller displays" i.e. a single string of LEDs rotated and modulated to show the time / some text / a picture. Yes you can buy them for peanuts but it's a nice mix of electrical & mechanical design to get into.

Otherwise, I'd go for something you'll use later, like a nice bench power supply for example? Make it with modularity in mind so you can add external CC/CV control, power logging etc.

[CatalinaWOW]

I think jpanhalt has identified your core problem with the memory project.  But maybe it is just a chance to turn a liability into an asset.  Make this a very slow, but high capacity memory.  Putting the BBs on a very large disk (or belt) that moves slowly enough that dynamics aren't an issue.

You might aid your detection problems with a differential observation technique.  Mount the beads in pairs, keep one in the dark and measure the brightness difference.  Ambient light will always be a killer.

[wobbly]

I'm glad my "Optical RAM" idea has captured so much positive attention, thanks all!
I will start a new thread in the "Projects" forum about it so people can chime in.  I'm going to revisit this project and use a different detection method... I have some "TSL235R" Light-to-Frequency Converter devices, which appear to be extremely sensitive and very fast to react.

Mount the beads in pairs, keep one in the dark and measure the brightness difference.  Ambient light will always be a killer.

Good idea, to have one "control" bead perhaps in the centre of the disk, where it's likely to receive the average amount of ambient light.  Perhaps even use the bead itself as a kind of lens that the sensor looks through, then the sensor won't be seeing the ambient light directly, but only as influenced by the beads!

I'm thinking that I'll inevitably have to run this machine in a fairly dark room, which will make it easier to shoot a video of it running.  Ok, I'm off to start the new thread in the Projects forum!

[wobbly]

I like the thinking about the disc - not sure how well it'd work in practice though as the dots would pretty much blur together with the rotation.

Thanks .  That's the other benefit of using a stepper motor to drive the disk - I can start and stop the motion quickly and with a lot of precision.  Timing is definitely going to be a critical part of this build!

[Benta]

Most obvious ideas are already available as ready-made products and well-documented.
I feel attracted to secret or sometimes "illegal" things, because you really have to reverse-engineer them, and it's an intellectual challenge every time. Note that this is only for my own satisfaction, not for profit.
My first bigger hobby project of this kind was breaking the Macrovision VHS copy protection (the boxes you could buy back then only did a partial reconstruction). Old stuff now, and no longer interesting.

Right now I'm playing with the "Keyless-Go" systems for cars. A nice demonstration/prank would be reparking your friends' cars. My target is to get the cost under $50 for a system, which is actually turning unambitious.
It'll be a one-off fun project just for myself, I'm not a car thief.

Is this odd? Perhaps, but I enjoy that kind of thing.

[Benta]

Sorry, but... thread hijacking your own thread?
@wobbly, you've already got a thread running on your hobby horse:
https://www.eevblog.com/forum/projects/optical-delay-dynamic-ram-(oddram!)/
Why force it into this one?
If someone's interested in your optical disk, there are now two threads to follow. 

[Kim Christensen]

Replace the lunchroom clock at work with a modified movement. Make it so that time slows down during break times and catches back up during working hours.   
Or if you're the boss, you could do the opposite. 

[Postal2]

... My first bigger hobby project of this kind was breaking the Macrovision VHS copy protection ...

Oh, I did that with a GRUNDIG VCR (VS200?). There was a signal somewhere in there that was turned on during fast forward. The signal had to be on all the time - and the VCR synthesized the synchronization very well.

[jpanhalt]

Right now I'm playing with the "Keyless-Go" systems for cars. A nice demonstration/prank would be reparking your friends' cars. My target is to get the cost under $50 for a system, which is actually turning unambitious.
It'll be a one-off fun project just for myself, I'm not a car thief.
Is this odd? Perhaps, but I enjoy that kind of thing.

There are some things best not done.  Some cars are worse than others: https://www.mprnews.org/story/2024/03/11/kia-and-hyundai-owners-continue-to-report-car-theft-after-free-security-upgrades .

[Rerouter]

My projects usually fall into a few flavors

- Making something because the price of a new one is out of skew
- I want to learn something
- It seems fun (Usually helping power level someone elses project)

I fully get Benta's thought process
There was a good year where I was treating automotive software like suduko, cracking CRC, encryption and mappings, not for an end goal, but just as something fun to work on.

[pcprogrammer]

Sorry, but... thread hijacking your own thread?
@wobbly, you've already got a thread running on your hobby horse:
https://www.eevblog.com/forum/projects/optical-delay-dynamic-ram-(oddram!)/
Why force it into this one?
If someone's interested in your optical disk, there are now two threads to follow. 

If you had read this thread properly from the start you would know that the idea for the optical memory disc started here and that wobbly started a new thread to explore it further, so no hijacking done here.

[wobbly]

Replace the lunchroom clock at work with a modified movement. Make it so that time slows down during break times and catches back up during working hours.   
Or if you're the boss, you could do the opposite. 

That's a fun one!  Reminds me of "Vetinari's Clock" from the Discworld books.  Many people have made them and this page has a good description of it: https://github.com/akafugu/vetinari_clock

[tggzzz]

Replace the lunchroom clock at work with a modified movement. Make it so that time slows down during break times and catches back up during working hours.   
Or if you're the boss, you could do the opposite. 

That's a fun one!  Reminds me of "Vetinari's Clock" from the Discworld books.  Many people have made them and this page has a good description of it: https://github.com/akafugu/vetinari_clock

Here's another, retrofitted to an early '70s 50Hz digital clock. https://entertaininghacks.wordpress.com/2015/02/23/vetinari-digital-clock/

Half a century, and still working

[RJSV]

   Like some other posts here I often put a money saving angle, into the equation.
So here I'm posting a scetch, of a little solar powered 'shopping bike'.   But also, I had proactively bought a used SCOOTER, as a kind of 'back-up' so that any inspirational feelings would have encouragement, (and some reminder).

   For typical weekly shopping, I would need more capacity, than my backpack, and better weight capacity.   Thus you can see, in my picture, shopping could utilize a rather large bin, with 2 wheels, in the back.
Once the concept was established, I could explore various battery and controls, which are the heart of the electronic designs and inspiration.
    Security always a concern, I had a paradox, in that making the CART too good would invite theft.   Plans were to use a typical bike lock chain.

   Even better, were plans for a Self-Driving cart, that could follow along behind, as I walked to supermarket!

   So you see, I tried to consciously create an environment, where more direct inspiration could happen, with less barriers.
Generating money, or at least saving some costs are inspiring too.

[Benta]

Right now I'm playing with the "Keyless-Go" systems for cars. A nice demonstration/prank would be reparking your friends' cars. My target is to get the cost under $50 for a system, which is actually turning unambitious.
It'll be a one-off fun project just for myself, I'm not a car thief.
Is this odd? Perhaps, but I enjoy that kind of thing.

There are some things best not done.  Some cars are worse than others: https://www.mprnews.org/story/2024/03/11/kia-and-hyundai-owners-continue-to-report-car-theft-after-free-security-upgrades .

What you link to is a completely different issue.
I'm playing with Keyless-Go systems, which is a basically flawed concept in itself. Believe me, no car having this feature is safe. I do not need physical access to the car or any part of it at all.
I'm just demonstrating the stupidity of car marketers (and to a certain extent NXP's), offering needless and dangerous "comfort systems", nothing else.

[jpanhalt]

@Benta,

I am equally concerned about all the keyless features on new cars.  My 2024's electronic "features" bother the heck out of me.  It's as bad as a Bill Gates OS, "We know what you want to do better than you do."  I hate it when settings
 revert back to some default on every restart.

Keyless start is at the top of my list.  My point was not whether they are crummy designs but whether anyone would want to move someone else's car as a "joke."  Imagine the shock of not finding your car?  What if someone knew you were not the owner and saw you get in it?  In America, a lot of people are armed.  At the very least, it is like using an electric shock on a friend as a joke or some of the TikTok silly challenges that end in death.  I am not a stick in the mud so far as humor.  I just don't find anything potentially dangerous to another person as funny.  My example of the Kia/Hyundai was not meant as an exact example, but rather as an example of a really poor manufacturer's implementation.  It was one factor in why I got a Mazda instead of the Kia/Hyundai brand.

[Kim Christensen]

I am equally concerned about all the keyless features on new cars.

My coworker has one of those cars. He and his wife carry their fobs in their pockets. One day, my coworker dropped off the wife and continued onwards to another destination. He didn't know that his fob was still at home and that the car had been started via his wife's fob. His car continued to run without a fob until he got to his destination and shut off the car.
Then he was stuck. 

[jpanhalt]

Exactly.  I learned of that "feature" at the Kia dealer.  1) The car should stop when the fob is out of range; and 2) There needs to be a manual override with a real key should the RF communication get disrupted.  My Mazda fob has a manual key hidden in it.  I don't know whether that key is somewhat unique.

EDIT: I can't imagine the logic behind remote start.  What if you could do that to an airplane?  That's why it is at the top of my most hated list.  And, I have survived 17 Minnesota Winters getting into a very cold car.