Finding Q point of a Transistor

[gazariff]

Hi there, I'm looking for info on how to find the Q point (linear region) of a transistor.
My college course didn't really teach you how to read data sheets.
I understand you need the collector base revers biased and the base emitter forward bias.
Though, I'm not sure what parameters of the data sheet relate to this.

Cheers guys.

[baklavametal]

only data u need from datasheet is hfe. (you dont even need the datasheet, just a hfe measuring capable multimeter)
you need values of all the resistors used (base resistors wether high or low impedance, collector resistor and emmiter resistor if u have it)
Ubeq (base emitter fw bias) for silicon transitors is usually 0.7, for Ge is about 0.5.
we never used cb reverse bias for Q calculations.
these are the formulas for the common emitter BJT amplifier (like a EHX LPB-1)
first you find the base resistance (if there's only one resistor, then it is it's value, if there's 2, then u use their value in parallel)
then u calculate base current Ibq=(Ucc-Ubeq)/(Rb+(1+hfe)*Re)
Then u calculate collector current (Ic)
Icq=hfe*Ibq
And then u calculate collector-emmiter voltage
Uceq=Ucc-(Rc+Re)*Icq

Ubeq - base emmiter forward voltage - 0.7V for silicon, 0.4-5 for Ge
Rb - base resistance
Rc - collector resistance
Re - emitter resistance
Ucc - power supply voltage

[R.G.]

Hi there, I'm looking for info on how to find the Q point (linear region) of a transistor.
My college course didn't really teach you how to read data sheets.
I understand you need the collector base revers biased and the base emitter forward bias.
Though, I'm not sure what parameters of the data sheet relate to this.

I come from a time when there was a solid course or two on how to bias a transistor, design circuits, and get them to work in the lab.

First: transistors don't have Q (for quiescent) points, circuits do. That's the place a circuit rests naturally when everything is running right. It may or may not be where the transistor is amplifying.

Transistors do have a linear region, a place where if you change the base current/voltage, the collector and emitter currents change mostly proportionately.

Some of this explanation will depend on an X-Y graph; I'm guessing that if you have had courses in EE or similar, you've seen X-Y graphs. If you haven't, yell and I'll either go draw something up (Ugh!) or try a different way.

Imagine an X-Y graph, with the x-axis denominated in volts between the transistor's collector and emitter. The Y axis is the collector current. If we open the base terminal and increase voltage from collector to emitter, the line representing current rises some nano-amperes from truly zero and mostly stays there as we increase voltage, right up to BVceo - the voltage where the collector-base junction breaks down and starts letting current through. This is a line right across the X axis, mostly indistinguishable from the X axis on most current scales.

Now imagine that we pump "a lot" of current into the base, with the collector-emitter voltage held at zero. The collector current is zero, or so nearly so that it doesn't matter. But as we increase the collector-emitter voltage by tenths of a volt, the current rises precipitously. The transistor acts like a mostly saturated silicon junction, perhaps with some small resistance in series with it from leads and such, and rises almost up the Y axis. This is the saturation curve, the biggest current the transistor will conduct when supplied "a lot" of base current.

In general "a lot" means supplying the base with 1/10 or more of the collector current. We can continue raising the collector emitter voltage and the collector current rises precipitously only a fraction of a volt or so from the Y/current axis until we burn out the bonding wires on the transistor chip, having reached the maximum drain current spec on the datasheet.

So  we have a line nearly at the X axis of "not quite on yet", ending with the collector-emitter breakdown voltage, and another line running up the Y axis of increasing current until the transistor reaches its maximum collector current.

Every place in the graph above the "not quited on yet" line and the "saturation current line" (that's what the massive-current line is near the Y axis) is the linear region.

You can construct a circuit that puts the transistor out in there anywhere, and it will amplify somewhat. How to do so is the next blast of typing.

Questions?

[PRR]

> transistors don't have Q (for quiescent) points, circuits do.

+1

> Transistors do have a linear region

And in small audio circuits, nearly any small-signal BJT transistor can easily be made to work (the "linear region" covers about any Q-point you would want). (Tubes and JFETs have narrower linear regions; or perhaps almost as wide but the useful gain falls off.)

> the next blast of typing.

Quivering in anticipation.

> find the Q point (linear region) of a transistor.

The *circuit*.

First define WHAT you want the *circuit* to do.

Input impedance?
Load impedance?
Voltage (or current) gain?
Available power supply?
Required signal level?
Frequency response? Hiss level? Power demand? (You can't ask for infinite zero zero, pick your priorities.)

Skipping to a "real" situation.....

And not defining all our specs up front......

In guitar-cord work, loads are 50K or higher.

Pick your collector resistor 1/2 to 1/10th of the load. 25K, 5K, somewhere in there.

Unless you have severe need for low-low power, pick a number on the low end and stick with it until you find a problem.

I pick 10K. (*)

In pedal-board work, 9V supplies are common, and can work fine. Guitar levels rarely approach 1V peak, so we need 2V supply, plus some for bias, plus some for fudge-factor (our parts are not precision {though circuits can be made precision}). 6V might be workable, but the Pocket Transistor Radio revolutionized music and also how we power our toys.

Set your Q-point so the collector is "about halfway" up on the 9V power.

Already we know the approximate desired collector current.

4.5V drop in 10K is 0.45mA

That "about half" is not exact. Especially since our 9V power is considerably more than we absolutely need. Often all is fine if the collector sits at 1/4 to 3/4 of supply. 1/3 to 2/3 might be a better target, while still leaving some margin of slop. (And 1/3-2/3 of 9V is nice round 3V-6V numbers; (*)always pick numbers you can work *fast* in your head.)

3V drop in 10K is 0.3mA
6V drop in 10K is 0.6mA

Do remember the difference between collector-to-common voltage and voltage-drop in the collector resistor.

So how can you set a precision-tested transistor to flow 0.3mA-0.6mA collector current?

And how can you set a random-grab transistor to flow 0.3mA-0.6mA collector current?

I hauled 842 pounds of flooring upstairs today so no more blast tonight.

[MaxPower]

There's a goldmine of info on these boards (just search for all posts by RG and Paul). RG sometimes goes into mind-blowing engineer-speak and Paul will regale you with anecdotes about flipping tractors or blowing up septic tanks or something. But man, so much great info. It's like a master's course on electronics. Errr, I forgot where I was going with this. I'm still laughing at Quivering in anticipation.

A couple of books I found very helpful:
Transistor Circuit Approximations by Malvino covers a lot of transistor stuff including various biasing methods (common base, common collector, common emitter).

Alternately Electronic Principles by Malvino is pretty good as well. Not as in-depth on transistors but it covers op amps and so on.

[Beo]

Between the theory of R.G.'s post and the practical application of PRR's post, this is a host of experienced and educated guidance... just three posts into this thread and I've learned so much. Unfortunately, as much as I think I understand what they wrote, I don't think I can integrate this info in a truly practical way. I think it takes electrical smarts, and serious experience.

Once I'm on a breadboard with a full circuit... I get crazy thinking what if I swap this 10K resistor with a 100K resistor, without really knowing from a theoretical or experience perspective what will happen. I could spend the time trying to work it out... but I don't have my own level of knowledge integrated so well.

So let me ask R.G. and PRR this question... and anyone else (optional if you choose to answer, I'm just curious)... do you feel your intrinsic gist of circuit design and electrical understanding comes from:
a: a gift at understanding and integrating electronic information (think rain man or good will hunting)
b: years of practical experience where the theory became visceral (like that old engineer geezer at your work who just knows everything)
c: something else or hybrid

[antonis]

Just a helping hand for visualizing the above well said...

[PRR]

> understanding comes from

How did {insert guitar-god here} get so good?

There is some element of "a gift". (One of my gifts is fat blunt numb fingers; guitar-god is not in my future.)

But most guitar-gods have spent YEARS flailing at strings 10 hours a day. Even the young ones; they often have started when very-young.

I probably read and thought about electronics several hours most days for several years before I could write a reasonable spec-list and propose a possible circuit, with good expectation it would work with few final changes. That's no longer (maybe less) than I would expect to work on guitar before going in front of people and playing a diverse play-list without being pelted by tomatoes and dead-cats.

> ...crazy ....what if I swap this 10K with a 100K

Again: your guitar-god has probably done "that" a million times: inserted an unexpected note in a conventional melody and found out what happens.

There are differences music and electronics. Electronics goes by approximate ratios. If you find a 1Meg and a 10K, 1meg and 12K is probably much the same, 1Meg and 22K may put the circuit "dead" (mute). In music a 2:1 ratio is nearly the same as a unison, but a 5.9% difference may be a sour dissonance. The two worlds are different; but knowing that, the process of exploration is about the same.

[plexi12000]

There's a goldmine of info on these boards (just search for all posts by RG and Paul). RG sometimes goes into mind-blowing engineer-speak and Paul will regale you with anecdotes about flipping tractors or blowing up septic tanks or something. But man, so much great info. It's like a master's course on electronics. Errr, I forgot where I was going with this. I'm still laughing at Quivering in anticipation.





hahhha!!!  no doubt, man!  i love when these fellas break lose....my pea brain is ALWAYS blown!!! 

[R.G.]

So let me ask R.G. and PRR this question... and anyone else (optional if you choose to answer, I'm just curious)... do you feel your intrinsic gist of circuit design and electrical understanding comes from:
a: a gift at understanding and integrating electronic information (think rain man or good will hunting)
b: years of practical experience where the theory became visceral (like that old engineer geezer at your work who just knows everything)
c: something else or hybrid

One of my  favorite stories. A group of ugly-American tourists is on a bus tour of English country houses. Upon driving up to an estate house with acres of billiard-table-smooth lawn, one of the tourists asks the tour guide "Hey, this lawn is beautiful! How do you guys get the lawn to look so good? I want to do it to my lawn."

The tour guide, with a dead-straight face says "Well, sir, generally we carefully prepare the ground, tilling it and leveling it, then applying a calculated amount of organic fertilizer. We then seed it, watering it consistently until the grass is about 3" high before beginning first mowings, then when the grass is established, we roll it smooth for 200 years."

[bloxstompboxes]

So let me ask R.G. and PRR this question... and anyone else (optional if you choose to answer, I'm just curious)... do you feel your intrinsic gist of circuit design and electrical understanding comes from:
a: a gift at understanding and integrating electronic information (think rain man or good will hunting)
b: years of practical experience where the theory became visceral (like that old engineer geezer at your work who just knows everything)
c: something else or hybrid

I know it wasn't meant in such a way, but it sort of implys that PRR and/or R.G. are brain damaged in that they can count 294 toothpicks in a pile on the floor, but can't understand that it doesn't matter if you buy your underwear at k-mart or not. lol.

[gazariff]

Cheers guys, I've not logged in, in a few days.

I have some reading to do, when i get a chance.

Is it sad to take a print out of this info on holiday, in case I get a chance to study it?