| Electronics > Projects, Designs, and Technical Stuff |
| How can I obtain amplifier circuit loop gain? |
| << < (6/6) |
| Wimberleytech:
--- Quote from: Cerebus on January 19, 2020, 04:47:27 pm --- --- Quote from: Wimberleytech on January 19, 2020, 03:44:41 pm ---Anyway, back to the technical stuff. Why do you say Vf/Vin does not represent the loop gain? --- End quote --- What is Vf? No, seriously; it's just been introduced out of thin air. If Vf is, as a I said a naïve assumption would be, the voltage at the negative input, then Vf/Vin can't be the loop gain. Let's take the classic feedback equation Vout = Vin A / 1 + Aβ Where Vout and Vin are obvios, A is the open loop gain and β is the feedback fraction. The loop gain (by definition) is Aβ, Vf (as assumed) is βVout <Problems fighting with the editor and greek letters, wait until this has disappeared before assuming I'm finished> --- End quote --- I agree with your analysis given your assumptions. But I interpreted it as the output of Aβ. This forum is due for a LaTex or some other formula entry capability (including greek characters), I think. |
| Cerebus:
--- Quote from: Wimberleytech on January 19, 2020, 03:44:41 pm ---Frankly, I thought you were a bit pejorative in your response, so I was just playing along. --- End quote --- No it was just a straight question. It was, at first inspection, not the answer to "How can I obtain amplifier circuit loop gain?" (this is complicated by the fact that several commenters appear not to know what "loop gain" is) so I scanned back, couldn't see a question (implied or direct) that it appeared to fit as an answer and so asked what question he thought he was answering - I actually thought I'd missed something. Pejorative would have been handed out directly, I'm not known for my subtlety. |
| Cerebus:
--- Quote from: Wimberleytech on January 19, 2020, 05:10:33 pm ---This forum is due for a LaTex or some other formula entry capability (including greek characters), I think. --- End quote --- The forum does actually support \$\LaTeX\$ via Mathjax but that is complicated (for me) by three things * Remembering how to invoke LaTeX on here * Remembering, from my infrequent usage of TeX/LaTeX how to express what I want in the same, and * MathJax have deprecated their CDN (that this forum links to) and eventually it will break thus rendering LaTeX indecipherable by us mere morals, or rather by not rendering LaTeX A search of the forum yielded this, which you can probably use as a starting point: --- Quote from: emece67 on September 05, 2019, 09:32:45 am ---For those unaware of its use, MathJax allows one to put math content coded ala LaTeX way. Thus --- Code: ---\$\lim_{x\to\infty}\left(1+{1\over x}\right)^x=e\$ --- End code --- renders as \$\lim_{x\to\infty}\left(1+{1\over x}\right)^x=e\$ and --- Code: ---\[N = \lfloor{1\over h}\cosh^{-1}\left({1\over \pi}\ln\left(2\cdot10^d\min\left(1, {b - a\over2}\right)\right)\right)\rfloor\] --- End code --- as: \[N = \lfloor{1\over h}\cosh^{-1}\left({1\over \pi}\ln\left(2\cdot10^d\min\left(1, {b - a\over2}\right)\right)\right)\rfloor\] --- End quote --- So, if I've got this right, I said: \[V_{out} = {V_{in} A \over {1 + A\beta}}\] And I forgot: 4. The output comes out too bloody small by default. |
| bson:
--- Quote from: Lucky-Luka on January 19, 2020, 09:01:36 am ---Apart from the fact that I've already received and answer, how can I obtain Vf? I don't think that such an answer adds something to the thread. Maybe you can prove me wrong. --- End quote --- The feedback voltage Vf is the voltage at the base of Q9. Vin the voltage at the base of Q10. The easiest way to measure them in LTspice is to add labels for those nodes; press F5 and type "Vf" in the box, then press ENTER. This gives you a net label to place: place it on top of the wire that leads to the base of Q9. Repeat for Q10 but call it "Vin" instead. Then run the simulation so you have a plot window. In the plot window, press the A key, and add a trace for "V(Vf)/V(Vin)". This will plot the loop gain. I mostly just assumed you know how to use LTspice already, and are familiar with basic amplifier analysis (since you asked for an analytical quantity). |
| Lucky-Luka:
The issue is not related to LTSpice... The fact is that... your method doesn't give the loop gain... At least... from my calculations... Vf=Vout*beta; Vout=[A/(1+A*beta)]*Vin; Vf/Vin is just about 1... Did I miss something? |
| Navigation |
| Message Index |
| Previous page |