What is the purpose of this 3rd transistor ?

[Beamin]

What does the transistor do on the top of this schematic? Why does it have so many parts around it, and is C750B an polarized cap? I have never seen a cap/ground schematic diagram like that. Its a back light dimming circuit to a yeasu radio where the input is off a software controlled board.
I figured out the middle transistor is the actual amplifier for the LEDs voltage and bottom left trans acts as a switch/dimmer controlling the power trans. I labeled the wires that are off the diagram but also there is a link to it in ze… er...ah... TRXBench er...ah...video (yes its slow to watch him explain zeez things).













https://youtu.be/epGTkALH5as?t=1340

[Nerull]

That's a zener controlled transistor series voltage regulator. The zener maintains the base at a specific voltage, and emitter follows at a diode drop less.

[Nusa]

and is C750B an polarized cap? I have never seen a cap/ground schematic diagram like that.

That's C7508, not C750B (notice C7507 label on adjacent cap). And yes, it's an electrolytic cap. There are many different capacitor symbols internationally. That one originates from Japan, I believe.

The *** presumably refers to a footnote somewhere about what value should be used.

[Beamin]

That's a zener controlled transistor series voltage regulator.

I figured it adjusted the voltage but how does that work? Does it make sort of a PWM circuit using the caps for timing the trans on and off or work off the 0.6 volt drop?

[Nerull]

That's a zener controlled transistor series voltage regulator.

I figured it adjusted the voltage but how does that work? Does it make sort of a PWM circuit using the caps for timing the trans on and off or work off the 0.6 volt drop?

The transistor acts as a variable resistor that maintains a desired voltage drop for any given output current. I edited a diagram into my earlier post which might help.

[Doctorandus_P]

To easily understand it, first takeout the resistor.
What's left is a simple resistor with a 12V zener diode.

Then put the transistor back in.
The base of the transistor is connected to the zener, and will be at 12V.
Which means that as soon as the Emitter goes above (approx) 11.4V, the base current of the transistor will cut of, and it wil not deliver any current to its output anymore.
If the Emitter of the transistor gets (much) below 11.4V, then it will also pull the base of the transistor to below 12V, and all the current the resistor can deliver will flow through the base of the transistor.

Result is that the circuit surronding that transistor is an (approx) 11.4V voltage regulator.

[Nerull]

The circuit in the OP uses an 8.2V zener, so would have an approximate output voltage of 7.6V.

[AG6QR]

You can't draw much current out of a zener shunt regulator.  It's just a voltage divider with the bottom resistor replaced by a zener.  As you draw current, it flows through the (top) resistor.  But if the voltage drop across that resistor exceeds the difference between the supply voltage and the zener voltage, the zener can no longer maintain regulation.

The pass transistor allows you to draw a lot more current while the voltage remains regulated.  The zener shunt only supplies the base of the transistor, while most of the current comes from the supply rail through collector and emitter.

[xavier60]

Q7502 and Q7503 form a DC amplifier with negative feedback resistors that set the gain at about 3.5.
I don't think that this stage is fed directly with PWM pulse, but rather PWM filtered to a DC control signal.
The zenner controlled transistor regulator is to supply reduced voltage to Q7503 to reduce heat dissipation.

[Beamin]

The circuit in the OP uses an 8.2V zener, so would have an approximate output voltage of 7.6V.

But the input is 12v so how would it drop to 7.6 that would be a lot of heat given 9 LEDs for lighting. Does this circuit give a constant voltage drop or does it oscillate with a RC timing circuit formed on the emitter by C7501 and R7501and the two caps on the output/collector (C7507 C7508) smooth it out? Or is this putting the transistor in a semi on state (linear vs saturation?)

[Nusa]

The circuit in the OP uses an 8.2V zener, so would have an approximate output voltage of 7.6V.

But the input is 12v so how would it drop to 7.6 that would be a lot of heat given 9 LEDs for lighting. Does this circuit give a constant voltage drop or does it oscillate with a RC timing circuit formed on the emitter by C7501 and R7501and the two caps on the output/collector (C7507 C7508) smooth it out? Or is this putting the transistor in a semi on state (linear vs saturation?)

Six LED's, not nine. Those other 3 components are resistors.

If you assume the circuit was designed to provide 20mA to each LED, what is the total current draw when the LED's are lit? At that current, how much power would be lost to heat to drop the voltage that much? Is that actually a large enough number to qualify as the same "a lot" you were thinking of before doing this exercise?

[Nerull]

The circuit in the OP uses an 8.2V zener, so would have an approximate output voltage of 7.6V.

But the input is 12v so how would it drop to 7.6 that would be a lot of heat given 9 LEDs for lighting. Does this circuit give a constant voltage drop or does it oscillate with a RC timing circuit formed on the emitter by C7501 and R7501and the two caps on the output/collector (C7507 C7508) smooth it out? Or is this putting the transistor in a semi on state (linear vs saturation?)

No, it doesn't' oscillate. Its a trivially simple series regulator. You are vastly overthinking how it functions.

[james_s]

Maybe it would help to draw it differently, first draw the zener and resistor between Vcc and Gnd, that will give you a reference voltage. Then draw an emitter follower vertically between Vcc and Gnd, then tie the output of the zener reference to the base of the transistor.

[Beamin]

Maybe it would help to draw it differently, first draw the zener and resistor between Vcc and Gnd, that will give you a reference voltage. Then draw an emitter follower vertically between Vcc and Gnd, then tie the output of the zener reference to the base of the transistor.

Is this the same principle? Is this whats inside an op amp to make it always try to equal out?

[xavier60]

This page has good progressive explanations starting with a simple zenner shunt regulator and its limitations, http://www.ko4bb.com/e102/e102-4.php
You need to also understand basic electronics.

[james_s]

An op amp starts with a differential pair, look up the internal schematic of the 741, it's readily available and shows what's inside it.

[Beamin]

An op amp starts with a differential pair, look up the internal schematic of the 741, it's readily available and shows what's inside it.

You could build this using transistors right? Since I don't see any of those multiple gate transistor or other weird things they can do in IC's. There was a vacuum tube logic gate on mr carlsons lab I wonder if you could make this with tubes too?

[Nerull]

opamps predate ICs, so you could certainly build them with tubes.

[Beamin]

opamps predate ICs, so you could certainly build them with tubes.

That was the thing in Mr. Carlson's videos. Imagine reading what you can do with an op amp in an advertisement? Funny to see something so basic advertised as benefits over the coopetition with their non op amp circuits. So to make the equivalent of a 16 pin DIP IC op amp you would need four of those with the power draw of eight tubes. 

[james_s]

You could build this using transistors right? Since I don't see any of those multiple gate transistor or other weird things they can do in IC's. There was a vacuum tube logic gate on mr carlsons lab I wonder if you could make this with tubes too?

https://shop.evilmadscientist.com/productsmenu/762

[Beamin]

You could build this using transistors right? Since I don't see any of those multiple gate transistor or other weird things they can do in IC's. There was a vacuum tube logic gate on mr carlsons lab I wonder if you could make this with tubes too?

https://shop.evilmadscientist.com/productsmenu/762

I was thinking about selling a whole series of such things but seems like someone already came up with it. It like good ideas; every time you think you have one, it turns out there is some obscure patent for it already and no one is capitalizing on it. If I was born 100 years sooner I would have more inventions then Edison.

[james_s]

Don't be too sure about that, you have the advantage today of hindsight, and access to vast amounts of information at your fingertips. A lot of inventions that are obvious and easy today were not so obvious or easy in the time and place they were developed with the resources that were available.

[Beamin]

Don't be too sure about that, you have the advantage today of hindsight, and access to vast amounts of information at your fingertips. A lot of inventions that are obvious and easy today were not so obvious or easy in the time and place they were developed with the resources that were available.

I found a drawing of mine I made in the third grade: it was a schematic and diagram of a car. It showed drive by wire with the steering wheel hooked up to a bunch of resistors with the -/\/\/- symbol around a steering wheel to electric motors on the wheels. I remember drawing it. When I was in 3rd grade fly by wire wasn't even a thing.

[Nusa]

Don't be too sure about that, you have the advantage today of hindsight, and access to vast amounts of information at your fingertips. A lot of inventions that are obvious and easy today were not so obvious or easy in the time and place they were developed with the resources that were available.

I found a drawing of mine I made in the third grade: it was a schematic and diagram of a car. It showed drive by wire with the steering wheel hooked up to a bunch of resistors with the -/\/\/- symbol around a steering wheel to electric motors on the wheels. I remember drawing it. When I was in 3rd grade fly by wire wasn't even a thing.

Sure there was, unless you're older than I am. You just weren't personally aware of it. The Concorde, first flown in 1969, was fly-by-wire. Going further back, there's the LLRV&LLTV (lunar lander trainers) flown in 1964 and the Avro Canada CF-105 Arrow in 1958. Or if you want something still flying, there's the Airbus 320 in 1987 and the F-16 in 1974.

[james_s]

In cars Saab did a complete drive by wire implementation in an experimental car in 1992, it was not one of their best ideas but it was electrical, using a motor to steer the wheels.

https://www.saabplanet.com/saab-9000-drive-by-wire-1992

This was after decades of experience with fly by wire in Saab aircraft. The concept is most certainly not new.