# Energy storage within two-port networks.

Discussion in 'General Electronics Chat' started by WBahn, Feb 17, 2014.

1. ### WBahn Thread Starter Moderator

Mar 31, 2012
20,224
5,748
Here's the deal. The textbook I am using (Nilsson and Riedel, Electric Circuits, 9th Ed) has a chapter regarding two-port networks and there are a couple of things that don't make a lot of sense to me. In particular, they state that the first restriction on a block that is to be modeled as a two-port network is that it can have no energy stored within it (note that this is in addition to the restriction that it can't have any independent sources, so that isn't what they are referring to).

They don't clarify it any more than that (that I can see). In particular, they don't explain the reason for ANY of the restrictions, which is frustrating. But this restriction would seem to indicate that there can't be any reactive elements with a two-port network, yet that would not only seem to defeat much of the utility, but it is also counter to many of the networks that they have in the problems for the chapter. So just what are they talking about?

I have to admit that I have done very little directly with two-port networks (though certainly many of the things I have done over the years would probably fall into this category if looked at the right way). In looking over information online, I can't find any place else that places this same restriction on the circuits in such a network, at least not in so many words. So I'd like to get some of your thoughts on the matter.

Thanks.

2. ### Tesla23 Senior Member

May 10, 2009
340
71
It seems like a strange statement.

When studying passive two ports it is usual to require that there are no internal sources of power, this allows you to assert that the power leaving the two port won't exceed the incident power, which leads to useful properties of the various circuit matrices. I suspect this is what they mean. No reactive power would mean studying resistive circuits only - is that what they are doing?

The best thing to do is to scan the chapter and see where they use this assumption, if it is well written this should be clear.

3. ### WBahn Thread Starter Moderator

Mar 31, 2012
20,224
5,748
Funny you should suggest that. I have access to the electronic version and I searched the entire chapter for the word "energy" and got exactly two hits:

In the chapter overview, where it says, "First, there can be no energy stored within the circuit."

In the chapter summary, where it says, "The model is limited to circuits in which no energy is stored inside the circuit between the ports."

So I can't use context of use to figure out what they heck they mean!

I don't know that I can claim that this book is "well written". There are lots of worse texts out there and this one has some redeeming features, but on balance I think I dislike it more than I like it.

4. ### nsaspook AAC Fanatic!

Aug 27, 2009
3,613
3,746
Do you think they are implying no 'initial' energy storage and the network reactive elements are at a zero energy state before any external sources are connected.

5. ### WBahn Thread Starter Moderator

Mar 31, 2012
20,224
5,748
I don't think so. It sure doesn't read like it. Plus, in the earlier chapter where they introduce the concept of a "transfer function" they are pretty explicit about stating that very condition.

I'm coming to the conclusion that they are simply wrong. There are a couple other places in the text where that is the case, so it doesn't come as a complete surprise.

6. ### GopherT AAC Fanatic!

Nov 23, 2012
7,249
5,964
If it is so unclear, send her an email and ask.

http://www.marquette.edu/engineering/electrical_computer/facstaff_riedel.shtml

7. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
784
Perhaps that no independent energy sources may 'reside' in any 2-port.

8. ### studiot AAC Fanatic!

Nov 9, 2007
5,005
519
This one should please Tesla since Gatland is from downunder.

Electronic Applications of Two-Port Networks
Gatland

Morton

Active Network Theory
Haykim

Higher Electrical Engineering
Shepherd, Morton and Spense

Extract from the last mentioned

So I guess that Nilsson is referring only to passive two port networks.
I do not have your book, in the one by Nilsson I have (Electric Circuits) his very short chapter does seem only to discuss passive networks and places the sources at the terminals. However he introduces some active two port networks in the questions at the end.

Go figure.

9. ### WBahn Thread Starter Moderator

Mar 31, 2012
20,224
5,748
Thanks for the link. I couldn't find any contact information on the publisher's website and didn't try any harder.

I've sent her an e-mail about this and also about some other questions regarding their material on the initial- and final-value theorems of the Laplace transform. I'll let you know what (assuming if) I hear back.

10. ### WBahn Thread Starter Moderator

Mar 31, 2012
20,224
5,748
So, I heard back from the author. nsaspook nailed it -- they meant that the initial energy needed to be zero. This makes sense if you have the restriction that there can't be any independent sources since any initial energy stored in a reactive component will effectively be transformed into an independent source when you take it into the s domain.

She also agreed that their wording on the constraints on f(t) and F(s) relative to the initial- and final-value theorems for the Laplace transform was confusing/misleading and will be revising that.

tshuck likes this.