Add something "fun" and educational to EDAboard?

[d123]

Hi,

Worst suggestion of the week: Any chance of adding sponsored competitions, etc., here where "winners" get a free evaluation board or capacitor set or similar?

Or, how about a "spot the mistake" "fun puzzle" section with e.g. schematics of known circuit types set for different levels of ability/knowledge with deliberate errors, or just plain bad circuits that are habitual design errors, for us all to think about and discuss and learn something from?

Possibly not formats this website wishes to pursue for any number of valid reasons or not a way of keeping bums on seats and gaining new members, who knows?

 

[betwixt]

You mean "who can design a three input XOR gate with fewest components" kind of competition?

Well, if you want to sponsor it and judge it, I'm sure WTWH Media would be interested.

As far as "spot the mistake" goes, we spend much of our time warning people about mistakes on other web sites already!

Brian.

 

[d123]

Hi Brian,

Thanks for answering.

1) You mean "who can design a three input XOR gate with fewest components" kind of competition?

2) Well, if you want to sponsor it and judge it, I'm sure WTWH Media would be interested.

As far as "spot the mistake" goes, we spend much of our time warning people about mistakes on other web sites already!

Brian.

1) Yeah, why not. Your XOR suggestion sounds pretty good, actually. Also things like writing a short piece of code to fulfil some function or other, or having snippets with silly mistakes in "spot the mistake".

2) Sure, I'm loaded. Surely WTWH deal with distributors and manufacturers and have sponsored content of some kind or other. Isn't this "enter now" modus operandi how semiconductor manufacturers publicise new products/devices from what I see?

3) True, along with my own poor circuit offerings and the related questions or those worrisome threads asking "What is a pullup - do I need to join a gym to make one? What is a Watt/What is a resistor? Can I create these things in code?" .

I meant typical recurrent things such as a schematic with incorrect transistor biasing (e.g. 100k in series with base/gate and 1k to ground) or not providing sufficient gate drive to NMOS and so on.

Was it The Fast Show where a character who always made a fool of themseves punchline was "I'll get my coat now."?

Anyway, thanks. I'll get my coat now.

 

[betwixt]

Well... I can suggest it to WTWH (the owners of Edaboard) but first you have to have a bash at my first question!

Brian.

 

[d123]

Hi,

Well... I can suggest it to WTWH (the owners of Edaboard) but first you have to have a bash at my first question!

Brian.

:grin: Okay! That's obviously very hard... but the weekend approaches and I love getting all OCD-ish and frustrated with things not working as wished for in the simulation tool. I doubt I'll succeed but it sounds such fun to think through and slog through whether or not you forward the suggestion. Get back to you on that one if I get anywhere un-embarrassing with the challenge.

 

[BradtheRad]

You mean "who can design a three input XOR gate with fewest components" kind of competition?

I too enjoy seeing 'novelty' circuits which use an unexpected approach to do an electronic function. The 4QDTEC website has many examples of this...
For instance XOR gates, one made from 2 transistors and a diode, the other from 1 transistor and 2 diodes. Although they're 2-input type they might be spun off to make 3-input type.

www.4qdtec.com/dlc.html

 

[betwixt]

I think some of us are quietly chuckling over this.
hint: Start with a truth table.

Brian.

 

[KlausST]

Hi,

like this:

There are 10 types of people in the world: Those who understand binary, and those who don’t.

Klaus

 

[Akanimo]

Hi,
...

There are 10 types of people in the world: Those who understand binary, and those who don’t.

Klaus

Oh yes. Really. [LAUGHS]
And somebody would ask "Why 10 and not 2".

I wonder where you got this from Klaus. Really funny.

 

[betwixt]

Well here are some more. They are old, at least pre 1993 because that's the date on the file and I claim no responsibility for writing any of them. Some are outdated and very USA centric but funny never the less.

Brian.

 

[d123]

Hi,

Thanks for the link to the cheat sheet Brad, it's a starting point. A demoralising starting point when you consider the preliminary thinking I did last night (at least 15 transistors so I must be on the wrong track):



This looks really hard (for me) (without cheating) from first considerations, I suppose suggesting the SN74LVC1G386 Single 3-Input Positive-XOR Gate doesn't count as "fewest components" as the emphasis is on the word design?

Nice joke Klaus, I don't quite understand it but it made me laugh when I read it during lunch. Is it about the 1 and the 0? I'd have asked the same as Akanimo - Why not two? And Brian, I'll have a read of the jokes later, always welcome, thanks.

 

[KlausST]

Hi

Nice joke Klaus, I don't quite understand it but it made me laugh when I read it during lunch. Is it about the 1 and the 0? I'd have asked the same as Akanimo - Why not two?

Because of the tiny difference:
* in binary "two" isn´t "2"
* in binary "two" = "10"

Klaus

 

[d123]

Hi,

Thanks for explaining Klaus, much appreciated. It's this one, isn't it:

**broken link removed**

because I don't deal with this stuff I can never remember the order (0000, 0001, 0010, 0011, 0100, etc.). I copied the image from Number. It explains it all for - I think - children, but could serve as an introduction for anyone interested.

In general, and this is not directed at you, Klaus, or anyone specifically at all, in case the following is in any way misunderstood:
As an aside, when I started the thread I didn't mean I personally want to e.g. enter competitions or think I can troubleshoot complex schematics as my knowledge is limited in most/all areas, that would have been deluded. Wouldn't want original motivation for the suggestion to be misunderstood. A man who wants to make a pF counter with a current source, two 555s and some CD4026s/4033s + SSDs is aware of their limitations. I just wondered if it were an idea worth considering so the focus isn't always on one person's circuit problem, and I like this forum.

My most recent "great" idea for a circuit I was considering was turning the EU mains AC 50Hz into a 1Hz signal with a 230VAC:15VAC transformer, a comparator, a couple of logic gates and two CD4017s, then dropped the idea as the ICs needed downriver of all that to count hours would be at least 5 or 6 more 16-pin CMOS ICs - too big to be worth making. Only reason to begin with that design was that it has been bugging me for a long time that the US has 60Hz for cheap clocks and here in Europe we have an inconvenient 50Hz and I couldn't think of a way to get 1 second out of 50Hz before last week. Under no delusions whatsoever, a hobbyist when time - and especially money - permit is just that.

 

[betwixt]

CD4059 can divide any frequency up to about 3MHz by any number between 3 and 15999. It is in current production.

Brian.

 

[asdf44]

The competitions could be subjective and judged by a simple forum poll.

Not much work required, just come up with the challenge.

 

[Akanimo]

CD4059 can divide any frequency up to about 3MHz by any number between 3 and 15999. It is in current production.

Brian.

In one of my projects, I had some difficulty dividing a 1 Hz clock to 4 hours with Duty cycle of 2min/4hrs. The clock division needed not be very accurate though so after some thoughts, I realised I could do it with a CPLD with leftover macrocells and pins to cycle signals back and forth the two ICs if I combined it with 555 oneshot. I never knew about CD4059 until you mentioned it on this thread and I just checked its cost and I'm happy that implementation was a less expensive one. I don't do microcontrollers. It would have been even easier with uC I believe.

 

[betwixt]

Frequency division with microcontrollers is VERY easy, particularly if the output doesn't have to be a square wave. Almost all MCUs have on-board counters, all you do is count the input pulses until the division ratio is achieved then toggle an output pin. For square waves the count is half the total required then the output is reversed so you get the final divide by two. Cost wise, an MCU is even cheaper than a single CMOS counter these days and you get the option to select different division ratios if required.

Brian.

 

[KlausST]

Hi,

Many microcontrollers even have built in timer/counter periferal, that just needs to be correctly set up.
After this it needs zero processing power (software) to do this task.

Frequency dividing may use internal clock as well as external clock. --> very low part count.

Klaus