Analogue Engineers in the house?

[The_Audio_Lord]

Ok then. Let’s move on.

I am very much interested what a sound engineer has to contribute. So what is it you like to talk about then? (aside from your trolling inputs/jokes, which backfired as you saw).

Yeah, it indeed backfired

I've talk with a lot engineers during the last year's and none of then seems to know what define the "small signal" concept in analog electronics.
Everytime that I say the small signal stands for classification of signal under the order of a couple Vth ( signal amplitude related to the thermal voltage) they do not accept it very well, mainly because it involves the acceptance that Embers moll equation yells that when the Vth ( reads N Vth quantity) is bigger then the Vbe, the IC current tends to infinity. And now, this consequence implies that the BJT is a voltage control device.
The treatment of a BJT as a voltage controlled device implies that we do consider always the input impedance and the Vth as the main parameter to design with it.
( of course, Vth/Zi + Vbe), otherwise the BJT would not operate in the linear region.

Comments on this please.


Please, any doubt do.not hesitate to ask, because English is not my first language... By far....

We have two themes here...
Small signal and BJT... Let's focus on the small signal concept first.
Cheers

 

[AnalogSteph]

I've talk with a lot engineers during the last year's and none of then seems to know what define the "small signal" concept in analog electronics.

Do you mean electrical engineers or recording engineers (the latter generally haven't had an actual engineering degree in decades, like past the 1960s or so)?

I'm not exactly sure what you're trying to get at. A system's small-signal model is a linearization around a specific operating point, making use of the fact that things in the physical world tend to be free from the kind of eccentricities that would trouble a mathematician. How far it holds is going to depend on how crummy your system is alongside your tolerance for inaccuracy / distortion - if in doubt, record output vs. input and pick a threshold of deviation.

Whether a BJT is to be considered voltage-controlled or current-controlled is one of those old debates that may never be fully settled. But does it even really matter? As they say, all models are wrong but some models are useful. Each of them is a tradeoff between accuracy and computational complexity. These days, you'd be silly not to make use of SPICE simulation anyway, which covers everything up to Gummel-Poon though you still need decent models for your components. It gets interesting when that is hitting its limits (e.g. junction capacitance modulation tends not to be accurate, so distortion in BJT circuits tends to underread, while MOSFET performance may come out as overly pessimistic).

 

[The_Audio_Lord]

Do you mean electrical engineers or recording engineers (the latter generally haven't had an actual engineering degree in decades, like past the 1960s or so)?

I'm not exactly sure what you're trying to get at. A system's small-signal model is a linearization around a specific operating point, making use of the fact that things in the physical world tend to be free from the kind of eccentricities that would trouble a mathematician. How far it holds is going to depend on how crummy your system is alongside your tolerance for inaccuracy / distortion - if in doubt, record output vs. input and pick a threshold of deviation.

Whether a BJT is to be considered voltage-controlled or current-controlled is one of those old debates that may never be fully settled. But does it even really matter? As they say, all models are wrong but some models are useful. Each of them is a tradeoff between accuracy and computational complexity. These days, you'd be silly not to make use of SPICE simulation anyway, which covers everything up to Gummel-Poon though you still need decent models for your components. It gets interesting when that is hitting its limits (e.g. junction capacitance modulation tends not to be accurate, so distortion in BJT circuits tends to underread, while MOSFET performance may come out as overly pessimistic).

Electrical engineers.

 

[sergeauckland]

Electrical engineers.

They generally know little about electronics, much more about electrical generation and distribution. Electronic engineers who have specialised in audio engineering are the ones who know about how audio circuits behave. There are lots of specialisations in electronics, just as in medicine and law, so it matters who you're talking to and about.

S

 

[DonH56]

Yeah, it indeed backfired

I've talk with a lot engineers during the last year's and none of then seems to know what define the "small signal" concept in analog electronics.
Everytime that I say the small signal stands for classification of signal under the order of a couple Vth ( signal amplitude related to the thermal voltage) they do not accept it very well, mainly because it involves the acceptance that Embers moll equation yells that when the Vth ( reads N Vth quantity) is bigger then the Vbe, the IC current tends to infinity. And now, this consequence implies that the BJT is a voltage control device.
The treatment of a BJT as a voltage controlled device implies that we do consider always the input impedance and the Vth as the main parameter to design with it.
( of course, Vth/Zi + Vbe), otherwise the BJT would not operate in the linear region.

Comments on this please.


Please, any doubt do.not hesitate to ask, because English is not my first language... By far....

We have two themes here...
Small signal and BJT... Let's focus on the small signal concept first.
Cheers

Vth = threshold voltage for FETs, must note application -- For BJTs, VT is the thermal voltage as I learned it (VT = kT/q)

Ebers-Moll BJT model, not "Embers", largely superseded by Gummel-Poon and later VBIC and quasi-static models for BJTs. Dr. Ebers died a little before I was born (way too young). Dr. Moll I actually met once or twice, not that either of use would remember, and of course he is no longer with us either. I met Dr. Gummel when working with Tektronix (the company I was working for was using their foundry).

BJTs to me are/were transconductance (technically transresistance) devices though it depends upon the application and model used (some focus on current, others on voltage).

The region of small-signal operation depends upon the device and the circuit. Not sure why we would discuss/debate on ASR; I don't think ASR is the place to discuss device physics.

 

[The_Audio_Lord]

Do you mean electrical engineers or recording engineers (the latter generally haven't had an actual engineering degree in decades, like past the 1960s or so)?

I'm not exactly sure what you're trying to get at. A system's small-signal model is a linearization around a specific operating point, making use of the fact that things in the physical world tend to be free from the kind of eccentricities that would trouble a mathematician. How far it holds is going to depend on how crummy your system is alongside your tolerance for inaccuracy / distortion - if in doubt, record output vs. input and pick a threshold of deviation.

Whether a BJT is to be considered voltage-controlled or current-controlled is one of those old debates that may never be fully settled. But does it even really matter? As they say, all models are wrong but some models are useful. Each of them is a tradeoff between accuracy and computational complexity. These days, you'd be silly not to make use of SPICE simulation anyway, which covers everything up to Gummel-Poon though you still need decent models for your components. It gets interesting when that is hitting its limits (e.g. junction capacitance modulation tends not to be accurate, so distortion in BJT circuits tends to underread, while MOSFET performance may come out as overly pessimistic

Vth = threshold voltage for FETs, must note application -- For BJTs, VT is the thermal voltage as I learned it (VT = kT/q)

Ebers-Moll BJT model, not "Embers", largely superseded by Gummel-Poon and later VBIC and quasi-static models for BJTs. Dr. Ebers died a little before I was born (way too young). Dr. Moll I actually met once or twice, not that either of use would remember, and of course he is no longer with us either. I met Dr. Gummel when working with Tektronix (the company I was working for was using their foundry).

BJTs to me are/were transconductance (technically transresistance) devices though it depends upon the application and model used (some focus on current, others on voltage).

The region of small-signal operation depends upon the device and the circuit. Not sure why we would discuss/debate on ASR; I don't think ASR is the place to discuss device physics.

Yes.... "Ebers", this happens all the time. I'm a foreigner in a non English country, so imagine writing with three languages text correctors on the phone.

 

[MakeMineVinyl]

Where is the engineers ?
People who actually are designing audio circuitry?

I'm an engineer who designs audio circuitry and there are many more here on ASR. However at least speaking for myself, I'm too busy actually designing audio circuitry to debate things like negative feedback for no apparent purpose.

 

[Piere]

Are serious engineers seriously debating feedback here? Any higher educated technical engineer did study control theory during his/her study. Although most have had serious difficulties with it, it is a basic skill as a technical design engineer, mechanics as well as electricity/electronics.

 

[tucura]

I do analog circuit design and I am involved now with thermal stability problems of class A/B power amplifiers using mosfet output devices. Is this forum still active?
Thx

 

[fpitas]

Probably using the curving transfer function as a limiter?

I suck @ tube circuits but looks pretty open loop to me

Looks like that center primary winding on the output transformer goes back to the cathodes of the drivers.

 

[fpitas]

I do analog circuit design and I am involved now with thermal stability problems of class A/B power amplifiers using mosfet output devices. Is this forum still active?
Thx

Nope. No one here but us chickens

 

[Killingbeans]

I do analog circuit design and I am involved now with thermal stability problems of class A/B power amplifiers using mosfet output devices. Is this forum still active?
Thx

Make a new thread in the DIY forum?: https://www.audiosciencereview.com/forum/index.php?forums/diy-audio-forum.39/

I bet there's tons of ways to add circuitry for temperature compensation.

 

[tucura]

Make a new thread in the DIY forum?: https://www.audiosciencereview.com/forum/index.php?forums/diy-audio-forum.39/

I bet there's tons of ways to add circuitry for temperature compensation.

Thank you for your suggestion but I would like to discuss an existing solution I am trying.

 

[egellings]

it has been a while I worked on pro audio gear but I can categorically tell you that I don’t believe feedback to be a weakness. Like everything else it can be used wrong, but without feedback you can’t process audio signal and stay within universally acceptable distortion levels.

Misapplied feedback would be a weakness; done properly, it is a strength.

 

[sarumbear]

Misapplied feedback would be a weakness; done properly, it is a strength.

Misapplied anything is bad. Doesn’t the “miss” part of the word explains that?

 

[fpitas]

Gotta wonder if this was simply a troll thread.

 

[egellings]

Nope. No one here but us chickens

Bedawk!

 

[fpitas]

Bedawk!

That wasn't chicken dung that OP dropped in his first post lol