Automatic control theory: rules of block diagram reduction.

Thread Starter

Cyrenaica —

Joined Jun 7, 2015
17
Hi to everyone!
I need small help about "how to make structural reforms correctly".
Here I have the initial circuit:IMG1.jpg.
Here I have my own conclusions about this:IMG2.jpg.
Am I correct?
If I am wrong, then any ideas about this?
Please, reply as fast as possible.
Thousand of excuses for broken English.
 

t_n_k

Joined Mar 6, 2009
5,455
The correct English terminology applied to this process is "block diagram reduction".
Your initial reduction is confusing (and possibly incorrectly executed) and you would be well advised to show the reductions in simpler /smaller steps and re-draw the reduced block diagram configuration after each step.
While others may be able, I'm afraid I personally couldn't assess your work without you following my above suggestions.
 
Last edited:

MrAl

Joined Jun 17, 2014
11,474
Hi,

I have to agree that seeing the individual steps would make it easier for us to check it, and we would also be able to see how you went about solving this.

From a quick look, it appears that you would start with the inner loops and work your way out, reducing loops as you go. If you would like to self-check your own work (which would also help you in the future) you can look up Masons Signal Flow Gain Formula, as that will give you a faster way to prove that your own reduction worked out correctly. it's probably better to do it both ways anyway and you should get the same result from both methods or else you made a mistake in one or both methods. If you do get a difference, you should do it over again, being more careful.
 

Thread Starter

Cyrenaica —

Joined Jun 7, 2015
17
>>"block diagram reduction"
thanks, i will note
>>initial reduction is confusing
well , I agreed now :(
>>re-draw the reduced block diagram configuration after each step
that would be too long
>>Masons Signal Flow Gain Formula
thanks ! will read about it

so I think this version is correct:

your opinions ?
 

t_n_k

Joined Mar 6, 2009
5,455
Yes solution looks good but check your notation for WIV ...

You appear to have

WIV=W1*W5/(1-WIII*W1*W5)

Should this not be

WIV=WI*W5/(1-WIII*WI*W5)
 
Top