So anyway, we should have a contest

[T3sl4co1l]

What do you think?  Something simple, like a radio, or light flasher for instance?

Other neat ideas:
Amplifier
Oscillator (various types)
Generating waveform of so-and-so shape

Of course, the more interesting (technically) ones require an oscilloscope to test, which may reduce the number of contestants.

Judging?  Prizes?

Tim

[Lightages]

I think the idea has merit. The contest could just be a consensus between members on how to do the contest but design by committee will go nowhere. I believe that for this idea to float, one individual needs to start the contest, provide or look for a sponsor for a prize if one is to exist, and one person needs to judge the result.

Because of my assertion, I don't think the idea will work without Dave stepping in and controlling the contest, or one of his sponsors. Perhaps if everyone agreed that someone who volunteered to do the judging was the sole judge, then perhaps. I can only see this ending in tears and fist fights.

[kalel]

Perhaps make the basic design rules be as simple as possible, with necessary parts easy to obtain (or make). That way, there would be both less and more complicated entries depending on ability and creativity.

Prizes, I guess they would have to be community driven. But if it's a heavy part such as bench gear, shipping can be a big issue, so in that case the contest might be limited geographically. That would also reduce the pool of entries. But I think it's not easy to make a worldwide one.

Judging, what about making that community driven as well? Instead of one (or two, three) person deciding based on what they personally like, entries could be voted on by everyone. As long as it's made in a way that doesn't invite some voting abuse (also less likely to happen if the reward is not something of high monetary value). If that's difficult to avoid, then a few judges could be selected (from anyone that applies to be one and has a good standing on the forum) to vote on the entries themselves.

Just a few thoughts, I can't say for sure if they are good ideas for this particular contest.

[buck converter]

I would't leave it to the community to set the rules. I don't think Dave needs to be involved either.
If I were to run something like this, there would be a date where everyone says if they are participating, and PM the organizer their basic idea. One  or two weeks later, the must submit there project in a 24 hour window. This way people can have the same idea without the "you copied me".
The single judge would choose who gets the prize, which could just be and old multi-meter or something. Community voting could end in a mess from duplicate accounts to fist fights.

contest idea: ORIGIANAL electric clock.

EDIT: This is just what sounds reasonable to me, but i do not have time / resources to run a contest.

[DmitryL]

ORIGI - what ?

[buck converter]

ORIGI - what ?

I actually have a few ideas myself, would enter it if could find the time and the contest is to make an original clock.

[DmitryL]

Original indeed! ... where is the corresponging smile ?

[Cyberdragon]

An electric clock you say?

How about start with some sort of magnetic tuning fork like an Accutron watch has but bigger. This would be driven by some vacuum tubes which provide the oscillation, amplify and shape the signal to square wave pulses. This would then be frequency divided down by a few more tubes until it is just low enough to be handled by relay logic. The output of the relay dividers would go to stepper switches which would handle the counting of the seconds for time and date and drive a nixie tube display. This would be set up inside of a transparent case so you could see (and definitely hear) it operating.

[buck converter]

An electric clock you say?

How about start with some sort of magnetic tuning fork like an Accutron watch has but bigger. This would be driven by some vacuum tubes which provide the oscillation, amplify and shape the signal to square wave pulses. This would then be frequency divided down by a few more tubes until it is just low enough to be handled by relay logic. The output of the relay dividers would go to stepper switches which would handle the counting of the seconds for time and date and drive a nixie tube display. This would be set up inside of a transparent case so you could see (and definitely hear) it operating.

Now that is  original . By original I meant that the display it self is something new. Thanks for giving me a new view of things.

[Cerebus]

Perhaps make the basic design rules be as simple as possible, with necessary parts easy to obtain (or make).

Constraints are good, they often spark good engineering.

Might I suggest one possible set of constraints for parts, a set that has the dual advantages of cheapness and universal availability. That set is:

• Silicon bipolar transistors, discrete (limited to common small signal types such as 2N3904 2N3906, common power transistors if the basic design challenge requires (e.g. PSU))
• Resistors, Best grade permitted in design no better than 1% metal film/oxide, 100ppm/C
• Capacitors, Al electrolytic, ceramic or polyester, nothing exotic or hard to source (e.g. PTFE, polystyrene)
• Diodes, Silicon
• LEDs if necessary

With bonus points if your design continues to work within specification if random small signal transistors (TUP, TUN) are substituted for the ones in your design.

Further bonus points available if your design continues to work within specification at temperatures of 0C and 70C.

[Cyberdragon]

An electric clock you say?

How about start with some sort of magnetic tuning fork like an Accutron watch has but bigger. This would be driven by some vacuum tubes which provide the oscillation, amplify and shape the signal to square wave pulses. This would then be frequency divided down by a few more tubes until it is just low enough to be handled by relay logic. The output of the relay dividers would go to stepper switches which would handle the counting of the seconds for time and date and drive a nixie tube display. This would be set up inside of a transparent case so you could see (and definitely hear) it operating.

Now that is  original . By original I meant that the display it self is something new. Thanks for giving me a new view of things.

You could also attach pointers and dials to the stepper switches to act as a secondary display, below the nixies. These could be lit by LEDs powered from rectified filament voltage.

[vk3yedotcom]

Constraints can induce ingenuity. Eg

Bronze:  'Two transistor challenge' with discrete components, no more than two active devices with <30 parts all up

Silver: 'One transistor challenge' with no more than one active device and <15 parts

Gold: 'Two terminal challenge' Unlimited number of parts but each only with two terminals (ie flashing LEDs and buzzers OK but no transistors)

[Cyberdragon]

Gold: 'Two terminal challenge' Unlimited number of parts but each only with two terminals (ie flashing LEDs and buzzers OK but no transistors)

It is entirely possible to build a simple computer using only two terminal components. Those of us who know the "secret component" should accept this challenge. Unfortunately I don't have alot of the "mystery component" lying around, so I'll only participate in theory or small circuits, but those who do or are willing to buy them in bulk (they should be cheap, but component quality will be a factor in these types of circuits) go for it!

[WZOLL]

+1 on the electric clock. I could even take it to school and get a free trip to the White House. 

[BrianHG]

As for a simple transistor as the most complicated device contest only, I would have been able to do something in my beginning years.
I have become so intertwined with MCUs over the years, I cannot think of developing anything without them.

As for MCU, like PIC based project ideas, I have a few of those...

Through some trickery, I can imagine making a clap-on-clap-off audio switch, or, whistle on/off using around 4-6 transistors with a MIC + a relay for the outlet switch.  Just having a dual or quad op-amp here would make life easier...

[bd139]

Constraints can induce ingenuity. Eg

Bronze:  'Two transistor challenge' with discrete components, no more than two active devices with <30 parts all up

Silver: 'One transistor challenge' with no more than one active device and <15 parts

Gold: 'Two terminal challenge' Unlimited number of parts but each only with two terminals (ie flashing LEDs and buzzers OK but no transistors)

I like the minimalism side of this. However I think someone has already won gold on that!: http://wwwhome.ewi.utwente.nl/~ptdeboer/ham/neonclock/



I like the idea of being restricted to one core part i.e. you're only allowed to use 2n2222's as per: http://www.k8iqy.com/qrprigs/2n240/K8IQYs%20Original%202N2-40%20Article.pdf

[Cerebus]

As for a simple transistor as the most complicated device contest only, I would have been able to do something in my beginning years.
I have become so intertwined with MCUs over the years, I cannot think of developing anything without them.

All the more reason, I'd have thought, to follow the Pease dictum and make the programming language du jour: solder.

[Cyberdragon]

Constraints can induce ingenuity. Eg

Bronze:  'Two transistor challenge' with discrete components, no more than two active devices with <30 parts all up

Silver: 'One transistor challenge' with no more than one active device and <15 parts

Gold: 'Two terminal challenge' Unlimited number of parts but each only with two terminals (ie flashing LEDs and buzzers OK but no transistors)

I like the minimalism side of this. However I think someone has already won gold on that!: http://wwwhome.ewi.utwente.nl/~ptdeboer/ham/neonclock/

CRAP You spilled the beans on my "secret component"! Yes, neon indicators can function as trigger switches that, when sometimes combined with silicon diodes, can form logic gates allowing pretty much any digital circuit to be constructed. There have even been calculators at least theorized, not sure if they were built. But if anyone wants to tackle this, have at it.

Also, that clock violates the rules, which are ONLY two terminal components, and that uses nixies! It is possible to build a neon clock/calculator using only two pin components from the plug to the output. The lamps themselves are already lamps, just use them to illuminate some numbers in columns for the output, with ones between for decimals/dividers. As for power, just use a capacitor dropper (or maybe just a resistor in 120V countries since it needs to be pretty high anyway), and then rectify and filter it. 

EDIT: I think an exception could be made to allow a three prong plug if a metal case was used. Safety first! Also exceptions for components that may have multiple terminals but only two connections, such as buttons and gas regulator tubes.

[bd139]

You could use those "long" neons to make a 7 seg display and drive it from a diode ROM. There's way more headroom at 100v+ for diode drops so you could get quite clever.

I can picture myself getting electrocuted building that though since I just gave myself a mains zap off a VTVM.