Dr. David Middlebrook Course !

vatche vorperian course

Please have a look at this:

https://www.ardem.com/index.asp

I think it is interesting. Has anyone already had a look at that material? Is it worth the money?

Re: Interesting Analog stuff

Prof. Middlebrook is a very talented person with much practical experience. I read a book about his new theories written by one of his students. I also have looked at the ICAP material.

The drawback about his material is that it deals with things that rarely occur and can be "mesured" with a spice simulation without knowing his theories. One example is a feedback amplifier where at higher frequencies the feed forward effect of the feedback network affects the amplifier output in undesirable ways. You can precisely calculate this effect with his equations and theories. I have yet designed a transistor amplifier where this effect occurred in the signal frequency range and therefore I was unconcerned that the gain or output impedance out of band was not the same as inband.

Interesting Analog stuff

I appreciate Middlebrook's contribution to measuring return ratio.
---
The other stuff seems tedious.

Re: Interesting Analog stuff

jasmin_123 said:

I appreciate Middlebrook's contribution to measuring return ratio.
---
The other stuff seems tedious.

Click to expand...

And which is the other stuff, that's tedious?

Re: Interesting Analog stuff

Middlebrook's extra element theorem is very interesting.

Re: Interesting Analog stuff

sutapanaki said:

jasmin_123 said:

I appreciate Middlebrook's contribution to measuring return ratio.
---
The other stuff seems tedious.

Click to expand...

And which is the other stuff, that's tedious?

Click to expand...

All the other stuff, except that related to measuring return ratio,
seem me very tedious. I feel no need to use it in analog design.
In most cases, simple intuitive things like superposition,
Miller's theorem, Blackman's formula, etc., are more than enough.
---
Moreover, their attitude to block diagrams seems me unfounded.
---
"Block diagrams are useful tools in linear system theory to help visualize
abstract concepts, but they tend to be very awkward tools in network analysis. For
instance, in an electronic feedback circuit neither the impedance loading nor the
bi-directional transmission of the feedback network are easily captured by the
single-loop feedback block diagram unless the feedback network and the amplifier
circuit are both manipulated and forced to fit the block diagram. The fact is block
diagrams bear little resemblance to circuits and their use in network analysis
mainly results in loss of time and insight." (Vatache Vorperian.)
---
Block diagrams have nothing to do with impedance loading.

In an electronic feedback circuit, bi-directional transmission of the feedback network can be easily captured by the single-loop feedback block diagram with no need to manipulate the feedback network and the amplifier and to force them to fit the block diagram.

Block (functional) diagrams bear much resemblance to circuits functioning and their use for network analysis (understanding) helps a lot.

Re: Interesting Analog stuff

Well, I understand that what you say is primarily a matter of personal opinion as will be what I intend to write now. I have not read in detail the book of Vatche Vorperian mainly because it seems he put too much math in it which obscures the essence of the methods as they apply to more practicle problems. However, I have listened to Dr.Middlebrook's course and I think that tedious is not the right qualification for his work. Actually I find it quite useful and this is not only for the way return ration is proposed to be measured. Someone already meantioned here his extra element theorm. I would add also the 2 extra elemts theorem.
I completely agree with you that intuitive understanding of circuits can not be replaced by any theory no matter how good it is in, producing results. Miller theorem, Blackman's rule, open circuit time constants, are all good and helpful, unfortunately not always. For example, Miller requires that the poles are split apart enough so that the gain is unchanged until the Miller effect starts kicking in. Plus it doesn't account for zeros in the circuit. OCT also does not account for zeros. Both these methods as you know are approximate and helpful for back of the envelope estimates. Blackman's rule is applicaple for systems with feedback.
I think that Middlebrook's way of presenting poles and zeros terms in the transfer function, their graphical interpertation and especially the methods of doing algebra on the graph are even more helpful than the above methods and rules. It is all about reducing the amount of algebra, not getting lost in it and not losing the intuitive understanding of the circuits. The most beneficial thing however, is that applying his methods one gets results that are readily suitable for design, something that he calls low-enthropy representation of the results.
As to block diagrams I think it is true that they don't reflect the real situation. Also more often than not they don't include the loading effects in a feedback circuit. In an arbitrary circuit the loading at high frequencies can come from many different places. And this can be a tricky thing when it comes to the stability of a loop. We all know the standard block diagram for a negative feedback system which is 1:1 applicable to a non-inverting opamp. However, it is not at all applicable to a inverting one, which is also very widely used configuration. As for me, I would rather use Middlebrook's feedback analysis than all four tyeps of analyses for the four types of feedback configuration as described in Sedra-Smith, Grey&Meyer etc.
And what if the circuit is such that it doesn't fit in any of those four configurations?
It is rather easy to account for all those things using the double-injection methods, plus you also takes into account the feed-forward transmission of the circuit.
I have used many of these methods in my work along with more traditional ones. I found out that in certain situations it is easier and faster to use traditional methods like KCL, KVL, Miller or Blackman's to solve a problem. But in more of the practicle situations you can not substitute for the ease of using Middlebrook's methods.

Re: Interesting Analog stuff

Hi, sutapanaki,

I like your reasoning, it sounds very professional. However, I have not yet seen a problem (except measuring return ratio) that cannot be solved in a simple way without Middlebrook theory.
---
Regarding block diagrams, I cannot agree with you. Block diagrams, or functional diagrams, are no more than a graphical representation of functions; none requests from a function to fit or reflect a circuit.
---
Regarding the "four types of feedback approach" (open-loop gain approach),
its drawbacks are well known, and in most cases it is very successfully replaced with the original Bode return-ratio approach.

By the way, there is no problem to take into account the feed-forward transmission of the circuit and the reverse gain of the amplifier in the open-loop gain approach.

I see no principal difference between the Bode-Blackman and Middlebrook approaches to feedback.

By the way, Middlebrook does use approximations and block diagrams, and his block diagrams do not always seem me well fitting the essence of feedback.
---
>We all know the standard block diagram for a negative feedback system which is 1:1
>applicable to a non-inverting opamp. However, it is not at all applicable to a inverting
>one…

Is not it enough to replace the "+, -" summing point with a "+,+" one to make the
simplified block diagram of feedback applicable to a non-inverting amp?
===
>But in more of the practicle situations you can not substitute for the ease of using Middlebrook's methods.

I would be interested to try and solve a practical situation of the above kind.
Could you please suggest me one?