It's not easy being a beginner........

[Noidzoid]

Turned out square Allen keys exist:
https://duckduckgo.com/?q=square+allen+key&iax=images&ia=images

And they're called Robertson drives.
https://en.wikipedia.org/wiki/P._L._Robertson

That is genuinely interesting, thanks  (I am adding this bit in brackets to assure you this is not sarcasm as it is easy to be misunderstood in situations like this.)

So a square drive key is called a Robertson and was manufactured by the same as although Allen may have previously come up with the idea, could not make it feasible to produce.

So:

Square drive key = Robertson
Hex drive key = Allen

Correct?

[tautech]

Turned out square Allen keys exist:
https://duckduckgo.com/?q=square+allen+key&iax=images&ia=images

And they're called Robertson drives.
https://en.wikipedia.org/wiki/P._L._Robertson

That is genuinely interesting, thanks  (I am adding this bit in brackets to assure you this is not sarcasm as it is easy to be misunderstood in situations like this.)

So a square drive key is called a Robertson and was manufactured by the same as although Allen may have previously come up with the idea, could not make it feasible to produce.

So:

Square drive key = Robertson
Hex drive key = Allen

Correct?

Yep correct, yet if you speak of Hex or Square drive everyone instantly knows what you're talking about. 

[Noidzoid]

Turned out square Allen keys exist:
https://duckduckgo.com/?q=square+allen+key&iax=images&ia=images

And they're called Robertson drives.
https://en.wikipedia.org/wiki/P._L._Robertson

That is genuinely interesting, thanks  (I am adding this bit in brackets to assure you this is not sarcasm as it is easy to be misunderstood in situations like this.)

So a square drive key is called a Robertson and was manufactured by the same as although Allen may have previously come up with the idea, could not make it feasible to produce.

So:

Square drive key = Robertson
Hex drive key = Allen

Correct?

Yep correct, yet if you speak of Hex or Square drive everyone instantly knows what you're talking about. 

My thoughts exactly.  Which is what frustrated me in the first place when reading Hex, Allen and squarish in the same definition.

I don't know much about electronics but if you saw a wave on a scope that was supposed to be a square wave, would it be ok if the corners were wobbly? Would it be ok if they were squarish? Or would it mean something was not right and need addressing?

[tautech]

For your enlightenment and entertainment 
https://www.eevblog.com/forum/testgear/show-us-your-square-wave/

[Noidzoid]

For your enlightenment and entertainment 
https://www.eevblog.com/forum/testgear/show-us-your-square-wave/

Aha! I see lots of squarish waves.

I will continue reading though I think it will be a while before I understand it all.

Thanks

[rstofer]

This particular subject, square waves into a bandwidth limited scope, is going to take a while to truly understand.  The math of the Fourier Series is a bit tedious and not at all important at this point.

The short version: A square wave is composed of a series of sine waves at odd harmonics (1st (fundamental), 3rd, 5th...) from DC to daylight (a really high frequency).  The amplitude of each sine wave is related to 1/n where n is the harmonic number.  As a result, by the time we get to the 9th harmonic, we're talking about a relatively small but nonzero contribution.

So, take a 100 MHz square wave and put it through a 100 MHz scope.  All you will see is a 100 MHz sine wave because the next harmonic is the 3rd (300 MHz) and that will never make it through the analog front end of the scope.  You absolutely won't have any 5th harmonic.  So all you see is the 1st harmonic, a sine wave at 100 MHz.

Repeat the experiment with a 10 MHz square wave and it will look pretty good because the 9th harmonic will pass through the front end.

Attached is the output from the Analog Discovery Spectrum tool.  It shows the spectrum of a 2V p-p square wave of 1 kHz that is symmetric with 0V (+1 to -1).  As a result, there is no spike at 0 Hz (the DC offset that would be present if the signal was 0V to 2V) and it shows the contributions out to the 19th harmonic and the amplitude is still not 0.

This is a five minute experiment with the Analog Discovery 2 which is why I think so highly of it.

At this point it is sufficient to know that you can't see a 100 MHz square wave on a 100 MHz scope.

[Paul Moir]

And they're called Robertson drives.
https://en.wikipedia.org/wiki/P._L._Robertson

Being a bit pedantic here, but a square drive is not a Robertson.  A Roberson is a trapazoid truncated pyramid, which is what makes it awesome, while square drive and Allen are not and therefore not awesome.  Think lathe key.   (For Noidzoid's sake:  Canadians are a little patriotic about our screw.  While the rest of the world foolishly ignores them, you'll find them ubiquitous here.)

[Dave]

I decided to go with The Art of Electronics for my main book.  Turns out it goes through things at breakneck speed and does not always do a great job of explaining them.

As an example its explanation of Thévenin's Theorem left me completely confused.  I guess its really meant to be used in a University setting where you are also attending lectures.

The vast majority of The Art of Electronics is layman level stuff. It's a 1200 page book, there is only so much they can pack into it.

[Noidzoid]

And they're called Robertson drives.
https://en.wikipedia.org/wiki/P._L._Robertson

Being a bit pedantic here, but a square drive is not a Robertson.  A Roberson is a trapazoid truncated pyramid, which is what makes it awesome, while square drive and Allen are not and therefore not awesome.  Think lathe key.   (For Noidzoid's sake:  Canadians are a little patriotic about our screw.  While the rest of the world foolishly ignores them, you'll find them ubiquitous here.)

I do believe they are used in some differential plugs.  Excellent as the key doesn't jam in the hole. Name not known to me as Robertson (unit now) but well aware the use. And yes, lathe keys rarely get stuck unless something fouls it.  Nothing wrong with your well placed pride for your  countryman it is an excellent design.

[StillTrying]

The vast majority of The Art of Electronics is layman level stuff.

I think you could get quite far just knowing all of AoE's subjects in layman's terms.

It's a 1200 page book, there is only so much they can pack into it.

Is it just me that can think of at least 3 solutions to the only 1200 pages problem.

[NivagSwerdna]

TBH I was more concerned with the definition of High Signal.  I think you need another book.  Art of Electronics can be had cheap second hand every now and again.

[rstofer]

TBH I was more concerned with the definition of High Signal.  I think you need another book.  Art of Electronics can be had cheap second hand every now and again.

There was a time when logic levels were compatible with vacuum tubes and even analog computers had swings from -100V to 100V and that was just their linear region, the supply voltages were much larger - perhaps +-300V.

We went through a long period where 5V TTL was king and although the switching threshold was much lower (logic 0 < 0.8V, logic 1 > 2.0V, we just called it 5V logic.   But since the output voltage could be as low as 2.4V, there wasn't all that much margin between 2.4V minimum output and 2.0V minimum required input.

Tocay 3.3V is common, 1.8V is coming along.  Lower voltages are in process.

https://www.analog.com/media/en/training-seminars/tutorials/MT-098.pdf

So, the definition in the text is sloppy because a logic one is not just a small delta V above 0V because the logic 0 threshold has to be some value > 0V but it's getting to the point where the thresholds are quite low and quite close together.  How far can you divide 1.5V?  Maybe a logic 0 threshold of 0.5V and a logic 1 threshold of 1.0V or something like that.  Pretty darn close to 0V for the logic 1 threshold.

The book is sloppy and everything should be taken with a grain of salt (or more).

If you wonder why logic voltages are getting lower, it is because of switching speeds.  There is an equation that goes like i = C dv/dt  -- the instantaneous current is equal to the capacitance of the circuit times the time rate of change of voltage (how long does it take to change voltage level).  We want dt to be small because we want to switch in near-zero time for speed.  We have only minimal control over C but smaller device geometry helps reduce it so the only thing we can do to reduce current (and heating) is to restrict dv, the swing in logic voltages.  We have to reduce the heat because we have so many gates changing state on a particular clock edge and we do that with smaller geometry (to reduce C) and lower voltage swings (to reduce dv) and this allows us to kick dt even lower (logic is faster) and speed is good!