MH's ideal coil and voltage question

[picowatt]

This model was an attempt to make sense of the Emf/Cemf feed back loop and D) represents said attempt. Do not be confused by your thought that I don't understand what you are saying because I understand completely, I just don't agree! I knew that my new model was not correct when I created the two Emf sources.

So, in another approach to resolve the matter, I propose we attempt to simulate the inductor that is, not use the internal inductor model(s) in any given simulator, but break it down into functional components as Webby suggests. 

LtSpice is my preference and it's toolbox includes arbitrary behavioral voltage and current sources plus voltage dependent current sources , current dependent voltage sources, etc, so all we need to do is basically establish the model and the math.

I see no reason why this should not be doable and it would be a great teaching tool.

I have attached a schematic that shows the general circuit we wish to simulate and what I believe to be your equivalent model, but I would like you to confirm or indicate any changes that should be made so we can proceed. I am still working on my own equivalent model and will post it as soon as it is complete enough for criticism.

pm

Partzman,

Under the "PW Model", the notation should read Cemf=L*di/dt and that is what you would have to simulate.

For a 5H inductor, you will have to sense the current flowing thru the circuit and continuously adjust the CEMF voltage source so that the rate of change (di) is maintained at .8A/s.   

In other words, you will have to simulate a rate of change dependent negative feedback loop.

I am not sure how you would go about detecting di, but once you have a way to do so, something like "if di<.8A/s decrease CEMF" and "if di>.8A/s increase CEMF" comes to mind. 

Perhaps .99 could offer some suggestions regarding this...

PW

ADDED:

Just thinking out loud...

If you can program an arbitrary current source so that it rises at a rate of .8A/s, it could be used as a reference current, Iref.  The applied EMF and Iref would be started simultaneously.  The current flowing thru the model would be detected, hereinafter, Idet.  You could then use the difference between Iref and Idet to generate an error signal to control the value of the CEMF Vsource using a comparator or the like.  In the real world of negative feedback, damping would also need to be dealt with... 

[poynt99]

Either your sim dose not represent an accurate description of what happens at T=0,or you are not looking closely enough.

At T=0,the current will rise with the voltage until such time as the capacitance of the windings has been charged. depending on capacitance value,the current will !after a short time! return to a 0 value,and then begin it's rise over time.

At T=0,the CEMF dose not equal the EMF,as at T=0, the current tracks vertical with the voltage,and then drops back to a 0 value. The time span is very narrow,but it is at and after T=0.


Brad

No, the sim is indicating the results it should be. I did clearly specify that these results are for an inductor that is free from parasitic capacitance. I guess you missed that.

[minnie]

   I've got a little question, those of you who are experts will probably be able to
give an answer.
    I've found from experience,whatever you're doing, if the formula you're using
isn't suitable your job goes wrong.
   Question "Do people who design circuits use the accepted laws and do they
work out right?"
   Obviously unidealness of components has to be accounted for.
        Thank you John.

[poynt99]

Partzman,

Under the "PW Model", the notation should read Cemf=L*di/dt and that is what you would have to simulate.

For a 5H inductor, you will have to sense the current flowing thru the circuit and continuously adjust the CEMF voltage source so that the rate of change (di) is maintained at .8A/s.   

In other words, you will have to simulate a rate of change dependent negative feedback loop.

I am not sure how you would go about detecting di, but once you have a way to do so, something like "if di<.8A/s decrease CEMF" and "if di>.8A/s increase CEMF" comes to mind. 

Perhaps .99 could offer some suggestions regarding this...

PW

I was thinking of suggesting/doing this yesterday, but changed my mind. I've been down this road before and the work effort is not worth it. It will only be met with skepticism and will be readily dismissed as invalid.

[picowatt]

I was thinking of suggesting/doing this yesterday, but changed my mind. I've been down this road before and the work effort is not worth it. It will only be met with skepticism and will be readily dismissed as invalid.

.99,

I added the following to my post you referenced:

Just thinking out loud...

If you can program an arbitrary current source so that it rises at a rate of .8A/s, it could be used as a reference current, Iref.  The applied EMF and Iref would be started simultaneously.  The current flowing thru the model would be detected, hereinafter, Idet.  You could then use the difference between Iref and Idet to generate an error signal to control the value of the CEMF Vsource using a comparator or the like.  In the real world of negative feedback, damping would also need to be dealt with... 

If Partzman wants to attempt this, do you have any better or simpler suggestions as to how he might to go about it?

PW