measuring closed loop gain/phase

[NZer]

Can anyone give me some assistance in measuring closed loop gain/phase of a closed loop system?  I would like to use the Bode plotter and I don't want to open the loop.  In assessing stability margins, I presumably have to consider both voltage and current gain.  How do I do this?  Any help would be gratefully received.

[lacy]

I am not sure if this will help or even if your version  has them, but there are measuring devices called Static Probes that you can put into the circuit to measure a variety of things like current, voltage and phase. I am not sure about gain, but that may be able to be extrapulated from the other measurements.

As far as the Bode Plotter, you would not technically be breaking the loop by placing it in series because I am thinking it is invisible to the circuit in the simulator. Therefore the series connection would be treated as if were a short to your component that it is tied to. I could be wrong about this, but in my use of the bode plotter I have not had any problem with it interfering with circuit response or operation.

I hope I have helped some.

[NZer]

Cheers Lacy, I appreciate your comments.  I have V10.0.343.  I originally tried what you suggested but I believe the Bode plotter is high impedance between input and output and therefore the feedback current will not pass through it transparently.  I believe what I have to do is measure the loop current gain and loop voltage gain and then post process them to get the overall loop gain G=(Gi*Gv-1)/(Gi+Gv+2).  I think you could measure voltage gain by inserting an AC voltage source in series with the feedback path and then connecting the bode plotter either side of that and similarly, you could measure current gain by injecting a current between ground and your feedback path with an AC current source; with the bode plotter measuring the current either side of the injection point.  I was kind of hoping that someone had done this and there was a simple way of automating the whole lot so you could simply plot loop gain.  Its not an unusual thing to want to know, if you are at all interested in your circuits stability!  It would be good if NI could put together some advanced tutorials.

Thanks once again for your suggestions

[lacy]

One more suggestion that may help (or may not) is to use the Analysis options under the simulate menu. There a a bunch of them from Transient, AC, DC Fourier etc. Maybe one of the can give you what you need. You can do analysis with these using equations so it may be possible to formulate an equation and plug it into one of these and see what happens.

If I sound a little unsure it is because I rarely use these. Most of my applications I can just use the transient mode and the instruments like the voltmeter and O-scope so I am not totally versed as to what they are truly capable of. The only thing I can tell you is to try them and see what you can do with them.

I hope I am helping a litle.

[NZer]

Hey Lacy, thanks for that - you have been most helpful.  It appears as if I can do it all in AC analysis.  However, there is one downside that I'm not sure whether it is possible to overcome.  I have to have two copies of my schematic on the one sheet.  One of them I use with a voltage source in the feedback loop and the other one has a current source feeding the feedback loop.  I can use static probes to measure the voltage and current and all the calculations can be done in AC analysis.  It seems to work ok. 

Do you know how I could use one schematic and say sequence the measurements by switching in and out the voltage and current sources? Is there someway of automating measurement runs where circuit parameters are changed each run?

Thanks for your help

Kind Regards

Ivo

[lacy]

This is probably going to sound like a stupid answer, but it may be what you need. You could use a double pole switch and wire it into the circuit and this may work to switch your voltage and current in one schematic. Since I am going at this blind without seeing how your circuit is configured, this is the best thing I can come up with. There may be a better option, but this is all I can think of right now (my brain is fried from work today so I may not be thinking as clearly as usual.)

As far as the second question that is something I am not sure of. I know that the Analyses under the simulate menu are capable of numerous calculations. The one that comes to mind is the Parameter Sweep Analysis, but I am not sure how to apply it to your application. If you are interested in learning what these can do for you I would suggest that you break out and dust off the User Manual (it is a PDF on the hard drive ). It contains very good information about these analysis and what they are capable of. If I have to use them I still have to bury my head in the Manual as well, because I very rarely have use of them.

[NZer]

Thank Lacy,

Sorry for the delay in acknowledging your post.  Good thoughts.  It may be the double pole switch is the only way.  There seems to be no way of programatically changing the double pole switch.  If you do it manually, it seems that I will then have to save the results of each measurement and then use the post processor to mathematically combine the two results.  Its hard to believe that the clever doods that designed this thing haven't got a better method of doing something so fundamental.  On the surface, it appear as if a combination of sweep parameters running AC analysis while controlling the current and voltage sources, run in stages with the batch processer should do the trick, but it doesn't appear to work the way it is supposed to.  Probably, I am doing something wrong.

Do the designers ever offer any helpful session in this forum?

Thanks

Ivo

[lacy]

I do not know if they are the designers, but there are a few engineers that pop in now and again to answer question as well as what I call the NI Support Team. This Team consists of Nestor, Max, Tien, and sometimes Garret. There are others to, but I am not as familar with their names as these four.

They are very knowledgeable about Multisim and electronics. When the forum users can't seem to come up with an answer they usally pull one out their hats. They are truly magicians (either that or just very, very smart).

If they read this post they may offer better options for you. If you do not get a response from them here you could e-mail your inquiry to http://www.ni.com/support . It may take a while to get a response, but they will look into your problem.

[NZer]

Thanks Lacy.  All the very best to you!