Making a filter that remains the same when loaded

  • Thread starter Thread starter Ry122
  • Start date Start date
  • Tags Tags
    Filter
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
Sign up Log in
76 replies · 6K views
LvW said:
May I ask (for clarification purposes only): ...better than what?

Better than other configurations. No emitter resistor is somewhat unstable. A simple resistor costs gain. Bypassing it makes the gain dependent on the ß. But bypassing most of it gives the best of all worlds.

Remember there's a cost to each added component. I've seen professional designs with a simple base bias resistor (plus the collector resistor). These have lots of drawbacks, but they are cheap. Sometimes cheap matters more than other considerations.

But for learning purposes, extra parts are free and doing the calculations is good practice.
 
Engineering news on Phys.org
Jeff Rosenbury said:
Better than other configurations. No emitter resistor is somewhat unstable. A simple resistor costs gain. Bypassing it makes the gain dependent on the ß.
....
.
Yes - that`s what I also have tried to explain in my posts#50,#52 and#54.
However, I doubt if Ry122`s five.transistors circuit (5 emitterstages in series) is really "better than other configurations".
Example: Where is the bandpass circuitry?
 
LvW said:
Yes - that`s what I also have tried to explain in my posts#50,#52 and#54.
However, I doubt if Ry122`s five.transistors circuit (5 emitterstages in series) is really "better than other configurations".
Example: Where is the bandpass circuitry?
It's not a contest. He is a student needing encouragement.

His coupling capacitors will act as high pass filters. He had some low pass filters at the beginning. As I said, I didn't run the numbers and have no idea if they are the right values. I suspect I could find lots more wrong with it as well. (Why put the filter where the signal strength is lowest, for example?) But this is a teaching moment. Besides, most of the "wrong" things aren't wrong except in specific contexts. In many (most?) designs the filter is fine at the beginning of the circuit. But there are some cases where that would bite him.

If he wanted a working circuit, he would have used the op-amps. He wants to learn. I'm all for that.

I do have to agree that if I were his boss and he handed me this five transistor stage design for a two pole, 26 dB gain filter, I'd want an explanation.
 
Jeff Rosenbury said:
It's not a contest. He is a student needing encouragement.
His coupling capacitors will act as high pass filters. He had some low pass filters at the beginning. As I said, I didn't run the numbers and have no idea if they are the right values. I suspect I could find lots more wrong with it as well. (Why put the filter where the signal strength is lowest, for example?) But this is a teaching moment. .

OK - I got it and I agree with you. It is - perhaps - a teaching event.
And from my own teaching experinece I know that, of course, there are always different teaching approaches.
However, in particular, because of this background I have recommended to the questioner a SYSTEMATIC approach: Starting with the desired filtering action (bandpass) - and shifting the design of the gain stages to the end of the design process. For my opinion, this is the best way to solve the problem.
To be honest - do you really think it would be possible to design the various coupling capacitors of the five stages so that the meet the highpass requirements?
More than that, also the by pass capacitors in the emitter legs exhibit a highpass function!
I really have severe doubts that this approach is a good one.

In short: I think, the best/most simple/most systematic approach would be a series combination of
* A passive 4-pole RC bandpass - decoupled with an emitter follower, followed by
* a two-transistor gain stage (emitter follower and common-emittere stage).
 
I agree that his approach isn't the one I would use. But we have pointed that out to him repeatedly.

He doesn't seem to want to learn proper design methodology. Nor should he unless he's planning on working as an engineer. Only he knows what he wants to learn from this. He has repeatedly stated he wanted to use common emitter BJT transistors. I have to accept that.

Since this is a Physic Forums, I assume he's a physicist and knows what he wants. So I'll try to help him as I can.

Scientists don't like learning engineering. It's a point of pride with some. It seems to interfere with their ability to ignore reality. Had Einstein been an engineer, he probably would have learned Newtonian mechanics so well he never would have found relativity.

Of course I wouldn't want to drive over a bridge built by a scientist. To each cat his own rat.
 
LvW said:
Because you have ac-shorted the emitter resistance in both stages the input resustance at the base node as well as the signal gain depends considerably on transistor parameters. Bad design.

Contrary to what you said, I found this at allaboutcircuits"The fact that we have to introduce negative feedback into a common-emitter amplifier to avoid thermal runaway is an unsatisfying solution. Is it possibe to avoid thermal runaway without having to suppress the amplifier's inherently high voltage gain? A best-of-both-worlds solution to this dilemma is available to us if we closely examine the problem: the voltage gain that we have to minimize in order to avoid thermal runaway is the DC voltage gain, not the AC voltage gain. After all, it isn't the AC input signal that fuels thermal runaway: its the DC bias voltage required for a certain class of operation: that quiescent DC signal that we use to “trick” the transistor (fundamentally a DC device) into amplifying an AC signal. We can suppress DC voltage gain in a common-emitter amplifier circuit without suppressing AC voltage gain if we figure out a way to make the negative feedback only function with DC. That is, if we only feed back an inverted DC signal from output to input, but not an inverted AC signal.

The Rfeedback emitter resistor provides negative feedback by dropping a voltage proportional to load current. In other words, negative feedback is accomplished by inserting an impedance into the emitter current path. If we want to feed back DC but not AC, we need an impedance that is high for DC but low for AC. What kind of circuit presents a high impedance to DC but a low impedance to AC? A high-pass filter, of course!"

That high pass filter mentioned above is the bypass capacitor posted in mebigguy's image

Or were you referring to the AC gain being dependent on the beta value and not that thermal runaway is still occurring?
 
He is referring to the AC gain depending on the transistor characteristics. Either you can live with that or not, depending on your application's requirements.

One solution is to get as much gain as you can in a multistage amplifier and then stabilize with negative feedback. This both controls the gain AND reduces distortion. It really is required in high quality audio circuits using discrete devices.

Look at google for "discrete negative feedback audio amplifier design"

http://www.allaboutcircuits.com/vol_6/chpt_5/13.html
http://circuit-diagram.hqew.net/Feedback-Amplifier-Using-Transistors_2749.html

http://www.pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Self_Douglas/Small_Signal_Audio_Design.pdf Do you understand open loop gain in an opamp? Why it is important, and how it decreases distortion? You can accomplish some of that in a well designed multi-stage negative feedback amplifier.
 
Last edited by a moderator:
Ry122 said:
Contrary to what you said, I found this at allaboutcircuits
Ry122 - where is the contradiction? I spoke about the DEPENDENCE of the signal gain on transistor parameters - not about the VALUE of the gain.
Of course, the gain drops without the capacitor because you now have not only dc but also ac feedback. But this is a desired effect!
All good amplifier stages work with signal feedback - BJT, FET, Opamp.
(remember: Opamps do not work at all without signal feedback).
 
To put it another way, maximum gain is dependent on the ß of the transistor. But since each transistor has a different ß it is hard to know just what that maximum gain will be.

Adding a small AC emitter resistance lowers the gain to a fixed and therefore predictable level.

If your goal is maximum gain, hand pack your transistors and ditch the resistor.

All of these variations have uses in differing applications. The key is to choose the one you want for your particular application. Knowing how to do that is why EEs get paid the big bucks. After all, anyone can look up a few schematics online.
 
Jeff Rosenbury said:
To put it another way, maximum gain is dependent on the ß of the transistor. But since each transistor has a different ß it is hard to know just what that maximum gain will be.
However, I must state - for the sake of accuracy - that two gain stages with different ß values and the same dc quiescent current Ic have the same signal gain.
This gain does NOT depend on ß but on the transconductance gm only (that is the slope of the Ic=f(Vbe) curve).
What differs is the signal input resistance at the base node only (smaller base current for higher ß values).
(It is a common misconception that higher current gain would give also higher voltage gain; perhaps, this false conclusion is caused by the - false - assumption that the BJT would be a current-controlled device).
 
hopefully there's some other benefit in addition to it making the gain stage's gain being less dependent on the inherent beta value of each transistor, because hand selecting a couple of transistors so they have the same beta value is extremely doable for me.
 
Ry122 said:
hopefully there's some other benefit in addition to it making the gain stage's gain being less dependent on the inherent beta value of each transistor, because hand selecting a couple of transistors so they have the same beta value is extremely doable for me.

Ry122 - in my post#25 I have listed all the benefits in case of ac feedback.
 
only 1 of those is applicable though right, since all the others can be obtained by only having an Re for DC
 
Ry122 said:
only 1 of those is applicable though right, since all the others can be obtained by only having an Re for DC
Sorry - but that`s not true. If Re is bypassed using a capacitor you only have one single effect: Stabilization of the dc opereating point.
All other benefits connected with ac feedback are lost.