PDA

View Full Version : New pet peeve....



John Coloccia
06-01-2013, 1:01 PM
...and I feel I must share it with all the electrical engineers and teachers out there.

A transistor is NOT an amplifier. A transistor is NOT a switch. Stop saying it's an amplifier. An amplifier is an amplifier. A transistor can be used to make an amplifier, but so can other things. Not only that, it can be used as a buffer with unity gain, or it can even be used to attenuate a signal. In the case of a FET, because it's a voltage controlled device, it's not even proper to call it an amplifier even in the loosest sense. It's almost like it's meant to be confusing on purpose.

If you simply must have an analogy, repeat after me: a transistor is similar to a VALVE. Not a switch...not an amplifier....it's not like two diodes back to back (who comes up with this nonsense???). It's getting a bit tiresome explaining to people how transistors work when I first have to un-explain all of the incorrect analogies.

I just felt like I needed to get that off my chest :)

Brian Kent
06-01-2013, 1:59 PM
So what is the difference between a valve and a switch?

I'll add to your rant with an irony. Why is it that out of the room full of electrical engineers that gather at my house on Thursdays to play Dungeons and Dragon, why can't anyone do electrical work (changing a light switch)? Shouldn't they be called electronic engineers, since they can program anything electronic on the planet?

David Epperson
06-01-2013, 2:41 PM
So what is the difference between a valve and a switch?

I'll add to your rant with an irony. Why is it that out of the room full of electrical engineers that gather at my house on Thursdays to play Dungeons and Dragon, why can't anyone do electrical work (changing a light switch)? Shouldn't they be called electronic engineers, since they can program anything electronic on the planet?
Think of it this way

How many programers does it take to change a light bulb?

It cannot be done. It's a hardware problem.

paul cottingham
06-01-2013, 3:34 PM
Think of it this way

How many programers does it take to change a light bulb?

It cannot be done. It's a hardware problem.
No. To the programmer, the light is broken. Of course, if he is a windows programmer, he just changes the industry standard to darkness.

Stephen Cherry
06-01-2013, 3:46 PM
So what is the difference between a valve and a switch?

I'll add to your rant with an irony. Why is it that out of the room full of electrical engineers that gather at my house on Thursdays to play Dungeons and Dragon, why can't anyone do electrical work (changing a light switch)? Shouldn't they be called electronic engineers, since they can program anything electronic on the planet?

They can't do electrical work for the same reason that the general population can't do much of anything. As people interested in woodworking, we are the exception to the rule. Curiosity, interest and willingness to give it a try are the exception today, rather than the rule.

As for electronics:

Horowitz and Hill, The Art of Electronics

Jim Matthews
06-02-2013, 8:10 AM
I believe that the base level of technical writing has descended to English translations of Russian assembly manuals.

What starts out shoddy becomes incomprehensible, to the point of being obviously wrong.
I met the designer of a simple vacuum tube amplifier (his English is excellent), outraged at the persistent errors described in the assembly of his preamplifier.

On the assembly manual for one of his more complicated preamps he said,
"They throw this over the cubicle to the lowest paid intern who has watched re-runs of Sesame Street and expect accuracy?"
- Erno Borbely

Transistors, having three terminals have complex behavior which makes their application flexible.
They're now like plumbing or elevators, we only notice them when they fail.

I believe we've entered a dangerous realm, where most of us take for granted devices that function simultaneously beneath our notice, and beyond our comprehension.
http://www.electronics-tutorials.ws/transistor/tran_4.html

Chuck Wintle
06-02-2013, 8:34 AM
Jim,
I don't know if you ever saw an operators manual for japanese motorcycles when they first started to arrive in North America but these were just awful to read and understand. The english language can be quickly learned but mastery takes a long time.

Chuck Wintle
06-02-2013, 8:47 AM
Transistors, having three terminals have complex behavior which makes their application flexible.
They're now like plumbing or elevators, we only notice them when they fail.

I believe we've entered a dangerous realm, where most of us take for granted devices that function simultaneously beneath our notice, and beyond our comprehension.
http://www.electronics-tutorials.ws/transistor/tran_4.html

Thanks for the link to this site...lots of theory to explain the whys of electronics.

Rick Christopherson
06-02-2013, 1:13 PM
A transistor is NOT an amplifier. A transistor is NOT a switch. Stop saying it's an amplifier. An amplifier is an amplifier. A transistor can be used to make an amplifier, but so can other things. While it is marginally incorrect to say a transistor is an amplifier, it is definitely incorrect to say it is not an amplifier. One limits its function, which is much broader than just being an amplifier. The other blocks its function, which does include amplification of a signal. The same is true for calling it a switch. It is not always a switch, but it can be a switch.

I have never heard of any electrical engineer generically refer to all transistors as amplifiers (or switches) unless one was actually operating as an amplifier (or switch). I don't think you have either, otherwise you wouldn't have commented about both "amplifier" and "switch" in the same "pet peeve". When they are being operated as an amplifier or switch, I don't see a problem referencing them according to what function they are serving.

Identifying a component by the function it is performing is quite common, and saves a lot of needless explanation for that function. For example, a "pull-up resistor" isn't any special type of resistor, but the term is used quite frequently so the audience (colleagues) don't have to needlessly examine the circuit to identify that this is the particular function of that resistor.

Chuck Wintle
06-02-2013, 3:19 PM
...and I feel I must share it with all the electrical engineers and teachers out there.

A transistor is NOT an amplifier. A transistor is NOT a switch. Stop saying it's an amplifier. An amplifier is an amplifier. A transistor can be used to make an amplifier, but so can other things. Not only that, it can be used as a buffer with unity gain, or it can even be used to attenuate a signal. In the case of a FET, because it's a voltage controlled device, it's not even proper to call it an amplifier even in the loosest sense. It's almost like it's meant to be confusing on purpose.

If you simply must have an analogy, repeat after me: a transistor is similar to a VALVE. Not a switch...not an amplifier....it's not like two diodes back to back (who comes up with this nonsense???). It's getting a bit tiresome explaining to people how transistors work when I first have to un-explain all of the incorrect analogies.

I just felt like I needed to get that off my chest :)

are you over reacting just a tad?

Jim Matthews
06-02-2013, 7:02 PM
Thanks for the link to this site...lots of theory to explain the whys of electronics.

The writers of these technical descriptions leave out the most important factor in modern electronics, the smallest components
are made of compressed smoke. When the smoke gets out, the device is broken and cannot be repaired.

Myk Rian
06-02-2013, 8:34 PM
I just felt like I needed to get that off my chest :)
Must have been bugging you for a while.

Jim Koepke
06-02-2013, 10:17 PM
Having worked in electronics, it is easy to understand John's peeve.

More than once in my experience someone was having trouble with a light in their car. Some people only know one electrical term and think that makes them an expert. I have heard people describe; a broken wire, a non-functioning switch and a burned out light as a short circuit.

It seems impossible to remove this condition from their mind. I gave up trying to dislodge incorrect notions from their blocked minds years ago.

As to transistors, some are designed for switching off and on, some are designed for analog signal amplification, while others are general purpose. Some may confuse switching with amplification because a small binary signal into a switching transistor allows a higher magnitude signal to travel through the transistor to power some circuit or other device.

They think this is amplification of a signal when it is just the signal turning a switch on or off.

jtk

John Coloccia
06-03-2013, 12:10 AM
are you over reacting just a tad?

What would have been a proper reaction? :rolleyes: If Grumpy Cat can get a movie deal, surely I can vent in the Off Topic forum about lousy engineering analogies.

Curt Harms
06-03-2013, 9:01 AM
They can't do electrical work for the same reason that the general population can't do much of anything. As people interested in woodworking, we are the exception to the rule. Curiosity, interest and willingness to give it a try are the exception today, rather than the rule.
<snip>


Boy, you ain't wrong about that! At least around here, many people are competent in their niche. Take 'em out of their area of expertise and they don't know enough to pound sand in a rat hole.

ray hampton
06-03-2013, 4:57 PM
Having worked in electronics, it is easy to understand John's peeve.

More than once in my experience someone was having trouble with a light in their car. Some people only know one electrical term and think that makes them an expert. I have heard people describe; a broken wire, a non-functioning switch and a burned out light as a short circuit.

It seems impossible to remove this condition from their mind. I gave up trying to dislodge incorrect notions from their blocked minds years ago.

As to transistors, some are designed for switching off and on, some are designed for analog signal amplification, while others are general purpose. Some may confuse switching with amplification because a small binary signal into a switching transistor allows a higher magnitude signal to travel through the transistor to power some circuit or other device.

They think this is amplification of a signal when it is just the signal turning a switch on or off.

jtk

the only short under the car hood is when the high voltage goes to a ground too soon, for everything else the one wire voltage will go to a ground

Rick Christopherson
06-03-2013, 5:21 PM
the only short under the car hood is when the high voltage goes to a ground too soon, for everything else the one wire voltage will go to a groundActually, a short-circuit is any unintentional (but can even be intentional) direct connection across the terminals of any source, load, or combination thereof that effectively removes that (those) device(s) from the circuit.

For example, if the 2 terminals of a resistor are connected together, then that connection is considered a short-circuit of the resistor. It virtually removes that resistor from the circuit by bypassing it.

For some power supplies, the short-circuit amperage can actually be one of its ratings to indicate the maximum amperage it will provide at near-zero resistance. Even high power transformers will typically have this rating.

Chris Padilla
06-03-2013, 5:25 PM
More than once in my experience someone was having trouble with a light in their car. Some people only know one electrical term and think that makes them an expert. I have heard people describe; a broken wire, a non-functioning switch and a burned out light as a short circuit.

jtk

I think it is quite common to refer to broken/non-functioning/erratic-behaving electronics as "having a short" even though technically, it might be wrong although it could be correct! :)

David Weaver
06-03-2013, 5:36 PM
"Having a short" is something that I've heard here often, too. Actually, that is the go-to term for most people around here for anything other than a burnt out lightbulb.

Rick Christopherson
06-03-2013, 5:37 PM
I think it is quite common to refer to broken/non-functioning/erratic-behaving electronics as "having a short" even though technically, it might be wrong although it could be correct! :)While it may be common terminology among those that don't know any better, it is nevertheless a pet-peeve for most people that know the difference between an open-circuit and short-circuit. They do have a specific meaning.

David Weaver
06-03-2013, 5:43 PM
That gives us an idea, to respond to everyone saying "I think it's got a short" by saying "or more likely, it's an open circuit issue".

People don't usually remember what something isn't unless they're told what it is or it's likely to be.

Chris Padilla
06-03-2013, 5:43 PM
While it is marginally incorrect to say a transistor is an amplifier, it is definitely incorrect to say it is not an amplifier. One limits its function, which is much broader than just being an amplifier. The other blocks its function, which does include amplification of a signal. The same is true for calling it a switch. It is not always a switch, but it can be a switch.

I have never heard of any electrical engineer generically refer to all transistors as amplifiers (or switches) unless one was actually operating as an amplifier (or switch). I don't think you have either, otherwise you wouldn't have commented about both "amplifier" and "switch" in the same "pet peeve". When they are being operated as an amplifier or switch, I don't see a problem referencing them according to what function they are serving.

Identifying a component by the function it is performing is quite common, and saves a lot of needless explanation for that function. For example, a "pull-up resistor" isn't any special type of resistor, but the term is used quite frequently so the audience (colleagues) don't have to needlessly examine the circuit to identify that this is the particular function of that resistor.

I think this makes sense. The basis for amplifier design is to use a voltage (between two terminals) to control the amount of current flow in the third terminal. In this way, a three-terminal device can be used to realize a controlled source: the basis for amplifier design. In the extreme, the control signal can be used to cause the current in the third terminal to change from zero to a large value, thus allowing the device to act as a switch: the basic element of digital circuits.

It sure seems easier to call it an "amplifier" or a "switch" in these cases! :D

To John: Perhaps you are most irritated by non-EEs tossing these terms around like they know what they are talking about?! :p

Chris Padilla
06-03-2013, 5:45 PM
While it may be common terminology among those that don't know any better, it is nevertheless a pet-peeve for most people that know the difference between an open-circuit and short-circuit. They do have a specific meaning.

Agree...and I like to quickly point out to the misinformed that it could be an open-circuit problem as well...or both. :)

Rick Christopherson
06-03-2013, 6:06 PM
To John: Perhaps you are most irritated by non-EEs tossing these terms around like they know what they are talking about?! :pI got the impression from one of his follow-up postings.....


..... If Grumpy Cat can get a movie deal, surely I can vent in the Off Topic forum about lousy engineering analogies.....that he was complaining about these being used as bad analogies. Not only are the functions correct, but because the terms describe the actual functions of the transistors, they aren't technically even analogies.

An analogy describes something by making a comparison to something else--typically something more easily understood. For example, water pressure and flow in a pipe are common analogies used for describing electrical voltage and current in a wire. The purpose of an analogy is typically to make a more complicated topic easier to understand by comparing it to something that is more commonly understood by the audience.

Conversely, comparing amperage to the number of electrons passing a point in a given amount of time, is not an analogy, because it actually represents the same thing.

ray hampton
06-03-2013, 6:34 PM
That gives us an idea, to respond to everyone saying "I think it's got a short" by saying "or more likely, it's an open circuit issue".

People don't usually remember what something isn't unless they're told what it is or it's likely to be.

sometime a open circuit could be a short circuit if the open are caused by a crimped wire working out from the lock-down screw or a broken wire showing copper, both examples will be open connections until the wire touch the ground sheet metal

Steve Meliza
06-03-2013, 6:38 PM
You can use a transistor as a resistor or turn it into a diode, but we give them names like "diode connected transistor" to indicate what the use of the transistor is. Maybe the next time some one says a transistor works like a switch or amplifier you can tell them "No, it works like a diode", though I doubt they'd understand the ridiculousness of either oversimplification used when making the analogies.

Rick Christopherson
06-03-2013, 6:46 PM
sometime a open circuit could be a short circuit if the open are caused by a crimped wire working out from the lock-down screw or a broken wire showing copper, both examples will be open connections until the wire touch the ground sheet metalAn open-circuit can never be a short circuit because they are diametrically opposed terms. A loose wire will be an open circuit until it touches something else, and then it may become a short circuit. But it can never be both at the same time.

Edit: Conversely, a short-circuit can result in an open-circuit (if the wire melts), but once that happens, it is no longer a short circuit.

glenn bradley
06-03-2013, 7:20 PM
So what is the difference between a valve and a switch?

Or a router . . .

263717

And a router?

263718

John Coloccia
06-03-2013, 8:00 PM
To John: Perhaps you are most irritated by non-EEs tossing these terms around like they know what they are talking about?! :p

Ha ha...this actually turned into a serious discussion :)

OK, so the example of a "pullup resistor" was used earlier. In that case, though, it really is just a resistor, and it does exactly what the term implies. The trouble with thinking about a transistor as an amp....or a switch....is that the transistor on it's own doesn't amplify anything. When you configure it in a circuit to behave as an amplifier, then the circuit takes the input and amplifies the output. It's gets very confusing for a newbie, though, to go for years thinking about transistors as amplifiers and then they look at an actual schematic and they're like, "What's all this other stuff???". Well, the other stuff + the transistor make an amp....or an attenuator....or whatever else you want to make.

It's like someone asking, "What's a nail", and you tell them, "Nails hold houses together". OK, I suppose nails can be used to hold houses together when you use them to hold together a house, but lots of other things hold houses together too, plus it completely misses the question of what a nail actually is and how it functions. It would be far better to say, "A nail is a round, thin piece of metal that has a point on one end and a flat on the other. You can hammer on the flat side and drive the pointy side through a couple of pieces of wood to hold them together. You can also drive them just partially into a wall and hang your chisels from them if you want too", etc etc. That actually describes what a nail is and THEN describes how you can use a nail to actually do things.

The situation gets even worse when you start talking about the transistor as a switch. Just go out on the internet look at the "transistor as a switch" schematics. They all do the same thing....there's an LED (or a motor....or whatever) coming into the transistor, and....wait for it....a MECHANICAL SWITCH that connects to the base. You hit the switch, and the transistor conducts. I look at that and I think, "The switch is acting like a switch. The transistor is doing what it always does...it acts like a valve". The only reason it's called a switch is because you run the transistor in saturation. Well, what if you don't run it in saturation? What if you have a pot on the base so that you can control the brightness of the LED? Then you hit the switch. What happens?

Well, the LED is less bright now. Now what's your transistor acting as? Is it acting as an amplifier? What's it amplifying, exactly? Do you call the dimmer on your kitchen lights an amplifier? Is it acting like a switch? Well, maybe sometimes a transistor acts like a switch with dirty contacts, I guess. No no no. The analogy is cute but it really is inaccurate and confusing to newbies. It's also never used ANYWHERE else except as a fluff, hand-waving description. When's the last time anyone here referred to as a transistor as an "amp" in any serious context? If someone came up to me and said, "Hey, that order of amps you ordered came in", I would be scratching my head trying to figure out what amps I'd ordered.....because I'm waiting for transistors to build an amp. I'd be wondering if someone ordered some op-amps by accident.

Anyhow, I hope no one takes this too seriously. It's just some good-natured silliness :)

Pat Barry
06-03-2013, 9:34 PM
John, not sure WiIkipedia shares your definition of a transistor. "A transistor is a semiconductor device (http://en.wikipedia.org/wiki/Semiconductor_device) used to amplify (http://en.wikipedia.org/wiki/Electronic_amplifier) and switch (http://en.wikipedia.org/wiki/Switch) electronic (http://en.wikipedia.org/wiki/Electronics) signals and electrical power (http://en.wikipedia.org/wiki/Electrical_power). It is composed of semiconductor (http://en.wikipedia.org/wiki/Semiconductor) material with at least three terminals for connection to an external circuit. A voltage or current (http://en.wikipedia.org/wiki/Electric_current) applied to one pair of the transistor's terminals changes the current through another pair of terminals. Because the controlled (output) power (http://en.wikipedia.org/wiki/Electric_power) can be higher than the controlling (input) power, a transistor can amplify (http://en.wikipedia.org/wiki/Gain) a signal." It says here that its an amplifier.

Also, according to Wikipedia, the only mention of valves is this "Prior to the development of transistors, vacuum (electron) tubes (http://en.wikipedia.org/wiki/Vacuum_tube) (or in the UK "thermionic valves" or just "valves") were the main active components in electronic equipment."

Are those the valves you are referring to?

Rick Christopherson
06-03-2013, 9:37 PM
When you configure it in a circuit to behave as an amplifier, then the circuit takes the input and amplifies the output. It's gets very confusing for a newbie, though, to go for years thinking about transistors as amplifiers and then they look at an actual schematic and they're like, "What's all this other stuff???". Well, the other stuff + the transistor make an amp....or an attenuator....or whatever else you want to make.hehehe....I was actually anticipating you saying this, so I was already prepared for it. :D

Without an external circuit, a transistor is nothing more than a chunk of non-conductive material with some impurities strategically interlaced inside. It is absolutely meaningless and useless without external circuitry. So you cannot ignore any external circuitry.

Contrary to your later assertion, making a transistor act as a switch doesn't even require any external components, aside from giving it voltage references. It goes into saturation and conducts. And that is a voltage-controlled switch.

To function as an amplifier, all it needs is a resistor (or load) and the same voltage references mentioned above. The small signal voltage at the input will be amplified at the output (generically speaking because there are many different types of transistors).


Well, the LED is less bright now. Now what's your transistor acting as? Is it acting as an amplifier? What's it amplifying, exactly? Do you call the dimmer on your kitchen lights an amplifier? Is it acting like a switch? Well, maybe sometimes a transistor acts like a switch with dirty contacts, I guess. No no no. The analogy is cute but it really is inaccurate and confusing to newbies. Here is where you are actually wrong, and this is not something an Electrical Engineer would ever mistakenly say. (I am not sure if you are an Electrical Engineer or a Software Engineer). That's because you missed the logical connection where the transistor is amplifying the small-signal input at the output.

Maybe this would be more evident if you thought about the load as being a speaker instead of an LED. Instead of just seeing a light get brighter or dimmer, with the speaker, you would hear it oscillate at different tones and volumes, depending on what frequency and magnitude the small-signal input was oscillating at.

So even without any external circuitry (aside from giving the transistor voltage references), a single transistor with no other components will in fact function as an amplifier to a speaker (or other) output. It will take a small-signal input and amplify it for use at the output.

Chris Kennedy
06-03-2013, 9:48 PM
John, not sure WIkipedia shares your definition of a transistor. "A transistor is a semiconductor device (http://en.wikipedia.org/wiki/Semiconductor_device) used to amplify (http://en.wikipedia.org/wiki/Electronic_amplifier) and switch (http://en.wikipedia.org/wiki/Switch) electronic (http://en.wikipedia.org/wiki/Electronics) signals and electrical power (http://en.wikipedia.org/wiki/Electrical_power). It is composed of semiconductor (http://en.wikipedia.org/wiki/Semiconductor) material with at least three terminals for connection to an external circuit. A voltage or current (http://en.wikipedia.org/wiki/Electric_current) applied to one pair of the transistor's terminals changes the current through another pair of terminals. Because the controlled (output) power (http://en.wikipedia.org/wiki/Electric_power) can be higher than the controlling (input) power, a transistor can amplify (http://en.wikipedia.org/wiki/Gain) a signal." It says here that its an amplifier.

As I read this description, I think it is getting to John's point. It describes what a transistor is capable of, not what it is. As it says, it is used to amplify signals -- and there is it is -- it is used to amplify signals. But is that what makes something a transistor?

I will freely admit that I have no freaking clue what a transistor is. I know they exist and I have seen them, but I don't know what defines one. I still don't from reading this thread, to be honest, but what an object can be used for doesn't define what it is. A wrench can be used to tighten bolts and so can a pair of pliers. They are two different tools. Defining what something is can be terribly difficult and tricky.

Chris

John Coloccia
06-03-2013, 10:33 PM
Here is where you are actually wrong, and this is not something an Electrical Engineer would ever mistakenly say. (I am not sure if you are an Electrical Engineer or a Software Engineer). That's because you missed the logical connection where the transistor is amplifying the small-signal input at the output.
I'm not sure where I'm wrong. You CAN use a transistor in exactly the manner I described to control an LED (or whatever else), and in that case it functions as neither an amp or a switch. If anything it functions as a variable resistor and dissipates heat just like a resistor. No one would ever call that an amp.

My MS is in Software Engineer, but I spent a career doing radar, robotics and electro-optics so I do somewhat know what I'm talking about :)

re: whoever mentioned valves
Well, in the UK, tubes are called valves. I think they were originally called valves in the US too, because the first Diode tubes acted as one-way valves. The triode tubes, which came later, have similar function to transistors today. They're all just called valves now in the UK, and over here we all call them tubes. That's not what I was referring to, but they WERE originally called valves for a reason, though :D

ray hampton
06-03-2013, 10:34 PM
An open-circuit can never be a short circuit because they are diametrically opposed terms. A loose wire will be an open circuit until it touches something else, and then it may become a short circuit. But it can never be both at the same time.

Edit: Conversely, a short-circuit can result in an open-circuit (if the wire melts), but once that happens, it is no longer a short circuit.

you could be wrong because sometime lighting will strike and a radio will become a open circuit plus a short circuit

John Coloccia
06-03-2013, 11:22 PM
you could be wrong because sometime lighting will strike and a radio will become a open circuit plus a short circuit

I was drinking some Silver Patron the other night with a friend of mine, and I felt like my brain became an open circuit plus a short circuit :)

Rick Christopherson
06-03-2013, 11:48 PM
I'm not sure where I'm wrong. You CAN use a transistor in exactly the manner I described to control an LED (or whatever else), and in that case it functions as neither an amp or a switch. If anything it functions as a variable resistor and dissipates heat just like a resistor. No one would ever call that an amp.That is where you went wrong. Because you saw a mechanical switch in the circuit example you were citing, you lost sight of the fact that the transistor was in fact acting as both an amplifier and a switch, depending on whether it was saturated or not. The existence of that mechanical switch caused you to loose sight of the forest for the trees.

As a matter of fact, the existence of that switch did the very thing you tried to caution against in a previous post. It is an external circuit component that misleads to the true function of the device (transistor). If you get rid of the switch in your example, you may suddenly see the true function of the transistor with less confusion.


My MS is in Software Engineer, but I spent a career doing radar, robotics and electro-optics so I do somewhat know what I'm talking about :)I sensed from your first posting that you were a software engineer and not an electrical engineer. I didn't want to jump to conclusions, but as soon as I saw you compare your sample circuit to a dimmer, then I knew for sure. That's because it is a common mistake for someone to make that has a fairly good understanding of electrical circuitry, but not a complete understanding.

At a self-educated level, it is easy to not see that the output is an amplification of the input. I remember making the same mistake years ago, because my mind was rushing about 10 minutes ahead of my professor's lecture. So for that 10 minute period, I was "self-taught" until the professor's lecture caught up with me. The only reason why I remember that is because it contradicted my "self-taught" intuition, which is not typically wrong. So it stuck with me for all these years, and that is why I spotted it when you did the same.

John Coloccia
06-04-2013, 3:38 AM
Well Rick, you're trying to make this about my background and skill level, but you don't know anything about my background and skill level. I still don't understand what "mistake" you imagine I've made. *I* don't find this confusing at all. You need to re-read my posts because I think you've skimmed through them a bit. The point of bringing up the mechanical switch analogy is that it's ridiculous to demonstrate a transistor switching by hooking it up to a switch. While you might want to do that sometimes, it's hardly a good demonstration of why using a transistor as a switch is useful.

What's the problem comparing my circuit to a dimmer? Do you mean because dimmer's are PWM'd? That's hardly the point. Replace that with "linear regulator" if it bothers you to think of a hypothetical dimmer. I don't know about you, but I would find it awkward explaining how a linear regulator works using the analogy of the transistor as an amp.

Rick Christopherson
06-04-2013, 4:59 AM
There was nothing wrong with comparing your circuit to a dimmer. However, what it did do was simply tell me what your background was. Obviously there was a foundation to it, or I wouldn't have guessed correctly. As the original poster that posted a pet peeve, your background is very much germane to the topic. You complained about an alleged miscue that really isn't an error.

Even your example with a switch driving the transistor isn't a ridiculous situation. That's because a transistor can behave as an amplifier all by itself. From your own example, consider that the switch may be a very low-power contact, but the transistor can be a very high power transistor. Therefore, that tiny little microswitch can suddenly control a massively large amperage via the amplification (and switching) of the transistor. The switch itself could never handle that total power, but it can control the transistor that is capable of handling that power.

As for a linear regulator, if you can make one with a single transistor, then just tell me where to send the check, because I'm "IN". For most regulators that aren't using an IC Op Amp, a zener diode is typically more critical to the design than the transistors surrounding it.

The more I read, the more I am beginning to realize that what your pet peeve is really about is that you are trying to learn something new, and you are not finding good enough analogies to get you there without fully understanding the entire topic. However, transistors are not simple devices, and their non-linear characteristics cannot even be simulated with a single model, because they behave differently under different conditions. If you want to see just how complicated it can get, then do some searches for "small signal analysis" of transistors, and you will see how complex the analysis can be. You can learn your analogies in a couple of hours on the internet. But if you want to go deeper than that, it is an entire upper-level college course dedicated solely to the topic.

John Coloccia
06-04-2013, 8:25 AM
???
Well, this has moved into the absurd. I'm not trying to "learn" anything. Rick, I've been designing and building circuits professionally for over 20 years. I think it's pretty obvious to everyone that you've completely missed something.

Steve Meliza
06-04-2013, 10:20 AM
I made a linear regulator with a single transistor, send me the check. It wasn't very linear and I soon replaced it with a Darlington pair, but it worked. No zener diodes or IC op amps used either.

Switched capacitor circuits are a good example of transistors working like switches.

Pat Barry
06-04-2013, 1:43 PM
A fundamental attribute of a transistor is its beta or current gain. This in itself implies amplification just by definition. There is a current amplification and that amplification is limited or adjusted by other circuit components. The end result is usually measured as some output voltage amplification, due to the current amplification. The other circuit components can be carefully selected to result in unity gain or complete saturation of the transistor. Anyway, I don't think a EE should be in fact concerned about whether or not the uneducated choose to use the terms short circuit or amplifier inappropriately. The educated person can either correct the error or nod and carry on without it bothering them.

Greg Portland
06-04-2013, 5:37 PM
Not a switch...not an amplifier....it's not like two diodes back to back (who comes up with this nonsense???).
On silicon, a diode is created with a P-N junction. Two of these junctions in series can make a P-N-P or N-P-N junction which is a BJT (bipolar junction transistor). That's where the "transistor = back to back diodes" comes from.

John Coloccia
06-04-2013, 6:38 PM
On silicon, a diode is created with a P-N junction. Two of these junctions in series can make a P-N-P or N-P-N junction which is a BJT (bipolar junction transistor). That's where the "transistor = back to back diodes" comes from.

Well, I know where it comes from. In fact, in one of my current designs I "misuse" the body diode of a FET as an actual diode. It hardly helps explain how a transistor functions, though. Look, I didn't have to choose transistors as my pet peeve. I could have chosen quantum entanglement...or maybe even just the fluff explanation of electron spin. Or I could have chosen any number of things with corresponding bad analogies. I define a "bad" analogy as one that falls apart when you actually try to build something practical. It's false knowledge because you're armed with words that are in one sense vaguely correct, but in a practical sense useless.

Myk Rian
06-04-2013, 7:04 PM
What ticks me off is people spelling it BANDSAW, instead of BAND SAW.
I lose sleep thinking of ways to get back at them. :D

John Coloccia
06-04-2013, 7:08 PM
What ticks me off is people spelling it BANDSAW, instead of BAND SAW.
I lose sleep thinking of ways to get back at them. :D

Oh, don't get me started. I'll trade my Bang Saw for your Radio Arm Saw.

Chris Padilla
06-04-2013, 8:18 PM
Is it a joinTer or a joiNer?? :D

John Coloccia
06-04-2013, 8:21 PM
Is it a joinTer or a joiNer?? :D

I don't care. As long as it has a helicoil head, I'm buying.

Stephen Cherry
06-04-2013, 8:33 PM
On a positive note (get it), it's great that people are that interested in these sort of distinctions. I recently found the probe to my O scope, so maybe I'll get the energy to play around with some of these doo dads.

John Coloccia
06-04-2013, 8:46 PM
I'm looking around for one now, actually. I was spoiled working in labs for 20 years where I got anything I wanted. It's harder when you work for yourself. Honestly, I don't really NEED a scope at the moment since my ears are the final arbiter, but I'm very curious just exactly what my circuits are doing, for future reference. Possibly for future digital projects. There's still a big gulf between pure analog processing and digital processing for the work that I'm currently doing. There is still a very poor understanding of exactly how to design audio circuits....non trivial audio circuits at any rate....to achieve the desired response to the human listener. It's very complex and non-intuitive.

Stephen Cherry
06-04-2013, 8:59 PM
John- I'm a recovering ee, certainly not an audio or circuit expert, but in my opinion I would definitely want an oscilloscope, and a function generator and a decent volt meter. It would be nice to see what a nice sounding amplifier looks like, and how it reacts to different speaker loads and sources. Fortunately, I don't thing that audio takes very much of oscilloscope. Maybe even a craigslist item.

John Coloccia
06-04-2013, 9:18 PM
John- I'm a recovering ee, certainly not an audio or circuit expert, but in my opinion I would definitely want an oscilloscope, and a function generator and a decent volt meter. It would be nice to see what a nice sounding amplifier looks like, and how it reacts to different speaker loads and sources. Fortunately, I don't thing that audio takes very much of oscilloscope. Maybe even a craigslist item.

I keep missing the Craigslist ones! I call and they're always gone.

There's something of a trick with guitar circuits....pickups, at any rate. They're big coils wound around magnets. It's a bit tricky to maintain the....oh...."feel"....of the guitar once you pass it through a buffer, like an op-amp for example. Anything you put in the circuit affects the frequency response of the pickups...so you figure that you should go to something like a high-impedence input but that isn't quite right either because it interferes with the response you get from an actual tube amplifier. It's very touchy and no one's really cracked the code yet. It's a very complex interaction that's going on there because you're essentially dealing with systems that have a very very very complex frequency response....dependent on amplitude also. Very tricky stuff. I've worked on detector circuits that are far beyond anything you could buy commercially, but at least you can plug and chug through those. There's no way I know to do what I'm currently doing other than blind experimentation...try this, try that....see what works. It's a bit unnerving, actually, but a lot more fun!

Pat Barry
06-05-2013, 2:02 PM
Its been a long time since my buddies used to rock but they would only accept the true Marshall sound with Stratocaster, Telecaster, Flying Vee or Les Paul. Nothing else sounded right so I know what you mean but I never thought this applied to the pick-up circuit, just the amp so they could get the distortion they wanted to hear. If in fact its a very complex frequency response problem then I doubt a simple scope is going to do much for you. You will need a frequency analyzer, sampling scope, frequency generator, anechoic chamber and lots of other stuff. Do you buy the pre-amp circuit with the pick-ups or roll your own?

David Weaver
06-05-2013, 2:56 PM
I keep missing the Craigslist ones! I call and they're always gone.

There's something of a trick with guitar circuits....pickups, at any rate. They're big coils wound around magnets. It's a bit tricky to maintain the....oh...."feel"....of the guitar once you pass it through a buffer, like an op-amp for example. Anything you put in the circuit affects the frequency response of the pickups...so you figure that you should go to something like a high-impedence input but that isn't quite right either because it interferes with the response you get from an actual tube amplifier. It's very touchy and no one's really cracked the code yet. It's a very complex interaction that's going on there because you're essentially dealing with systems that have a very very very complex frequency response....dependent on amplitude also. Very tricky stuff. I've worked on detector circuits that are far beyond anything you could buy commercially, but at least you can plug and chug through those. There's no way I know to do what I'm currently doing other than blind experimentation...try this, try that....see what works. It's a bit unnerving, actually, but a lot more fun!

John...what are you attempting to get, super warm live hard driven tube sound out of solid state? Or are you attempting to create electronics that can be in a loop or between a guitar and amplifier without screwing up the ability to still get the raw pure sound?

Stephen Cherry
06-05-2013, 4:01 PM
Its been a long time since my buddies used to rock but they would only accept the true Marshall sound with Stratocaster, Telecaster, Flying Vee or Les Paul. Nothing else sounded right so I know what you mean but I never thought this applied to the pick-up circuit, just the amp so they could get the distortion they wanted to hear. If in fact its a very complex frequency response problem then I doubt a simple scope is going to do much for you. You will need a frequency analyzer, sampling scope, frequency generator, anechoic chamber and lots of other stuff. Do you buy the pre-amp circuit with the pick-ups or roll your own?

This is true- if you want to get highly analytical, you may need to buy some gizmos. BUT, even just looking at the transient response to a step input can give some feel for what the characteristics of a device are.

John Coloccia
06-05-2013, 4:52 PM
Its been a long time since my buddies used to rock but they would only accept the true Marshall sound with Stratocaster, Telecaster, Flying Vee or Les Paul. Nothing else sounded right so I know what you mean but I never thought this applied to the pick-up circuit, just the amp so they could get the distortion they wanted to hear. If in fact its a very complex frequency response problem then I doubt a simple scope is going to do much for you. You will need a frequency analyzer, sampling scope, frequency generator, anechoic chamber and lots of other stuff. Do you buy the pre-amp circuit with the pick-ups or roll your own?

I'm currently building some pedals, actually. I have an overdrive coming out in a couple of weeks. The interaction with things like guitar cables and things like that changes the resonance frequency of the pickups. Funny how pickups work, actually...there's a pull and tug between the pickup's inductance and capacitance which contributes to it's resonant peak...and add any capacitance anywhere in there and you will move it. That's why longer cables, especially cheaper cables with higher capacitance, will muddy up your sound. You're lowering the frequency response of the pickup.

I'm playing around with some DSPs, but everything slated for release at the moment is mostly discrete components with the occasional dual or quad op-amp chip, mostly for convenience, when I can get away with it.

David: everything I do is pure bypass, so there's never anything in the way of the sound when it's off. There are advantages and disadvantages to that. Most setups will benefit from a buffer SOMEWHERE in the chain. A typical setup could easily have 30' of cable. 10' or 15' from guitar to pedal board, and another 10' or 15' from the board to the amp. That can really muddy things up unless you grab the signal somewhere and buffer it. Once it's buffered, you can drive it with very little high end loss for quite some distance. Of course, there's the famous coiled guitar chords that were popular in the 60s. Sometimes, you use one of those because you actually WANT to get that mud. Hendrix was famous for doing that in the studio.

The point of bringing in a scope is to build up a library of response vs my impression of the resulting tone. Over time, that will become my Rosetta Stone for dialing in exactly what I want far faster than making educated guesses and experimenting. All the interesting stuff is happening around audio frequencies though, so pretty much any scope will do. I can buy a new Rigol, 2 channel 100MHz scope for $400, believe it or not. Incredibly cheap.

David Weaver
06-05-2013, 5:11 PM
You mean dial in what you want to create more effects without screwing around to get there?

I'm out of the loop on scopes. The last time I saw a scope was a McIntosh from an enthusiast about two decades ago, but I understand that the high priced allure of a lot of vintage mac components is defunct because the internet made it easy to buy and sell them.

I'm still in dummy land as far as "OK, I like that tone, I don't know what makes it that way" or "I don't like it...ew".

Most tones and most effects fit in the second category.

John Coloccia
06-26-2013, 4:06 AM
Stay tuned. I picked up a scope a couple of weeks ago. I had to debug some funkiness on the prototype PCB boards. As sometimes happens, things don't translate perfectly well when you try to cram a circuit into a tiny little space, and you have to get in there to see where it's going wrong. Fortunately, after much head scratching, I have that all sorted out now. Anyhow, I'll try to post some pictures tomorrow of what this stuff looks like. If you've never seen it before, it will suddenly bring to life why we use the terms "cllipping" and "distortion".