How do you use the bode plotter differentially?

[didymus7]

I've run into this problem on and off since Multisim 6 and I'm beginning to think that I might not be doing it right.

Attached is my circuit, note that the input is two sine wave voltage sources which simulate the real-world input.  If I attach the input of the Bode Plotter to one or the other of the inputs and the Bode Plotter Output to the output of U3A, the phase looks okay but the magnitube is weird, it looks like a bandpass filter.  If I try and use the Bode Plotter with a differential input, the + Input to one generator, the - input to the other, I get phase but no magnitube plot at all.  It's not even off-screen, it's just not there.

Previously the workaround was to start with a single source, use that as the Bode Plotter input, then use circuitry to generate both inputs.

Am I doing something wrong?  Is there some dead chicken I'm supposed to wave over my head? 😉

Thanks...Steve

[_user32]

Please no dead chickens this is a family friendly forums page.  🙂

I have not had time to look at your circuit, but here is what happens with the bode plotter...

The bode instrument plotter controls 1 sinusoidal input, linearizing the circuit and performs an AC analysis.  I would think you want to connect up a single AC voltage source differentially to the input (+ to the input where you have the first input and - to the second input) and test it in this fashion.   Hook up the input of the bode plotter to the +/- terminals on the voltage source.

Let me know if this helps, I may have some time to look more in detail tomorrow on this issue.

Regards,

Pat

[didymus7]

Pat,

That's pretty much what I try to do and I don't get the proper response.  It seems like I have to either ground or leave open the minus input of the Bode Plotter.  In this case also, even if I use just one input, the result is still not proper.  It acts like their are low frequency poles, which, obviously, there aren't. 

For example, if I connect (in this circuit) V1 to the + input and V2 to the - input, the bode plot begins about 100 db down, peaks up  and then rolls off.  I actually get virtually the same if I only connect V1 to the + input and leave the - open.

Steve

[didymus7]

Sorry, let me correct myself.  If I use only V1 or only V2, I get an odd bandpass result beginning about -112 db, ramping up to a peak about 50khz, then rolling off.  If I try to use V1 (or V2) to the positive input and V2 (or V1) to the negative input, I get no magnitude track.  It just never appears.  And, no , it's not even off-screen.

Thanks....Steve

[didymus7]

Well, time to clear the weeds here again.

The question still remains.  The bode plotter includes + and - terminals on both input and output.  How does one use them?  In a previous message I attached a circuit where two things happens, if one of the differential inputs are used and run to the + input of the bode plotter, the result is a big error, despite the circuit, it shows a bandpass, rather than a lowpass and the gain is way too low.  If you try to put one input to the + and one to the - of the bode plotter, you get NO amplitude display, the phase comes out, but no trace on amplitude.

Why did everyone run away and hide from this question????

[_user32]

didymus7,

I tried to run your circuit in transient and with the bode plot but I got a few errors.   It appears that in the original circuit, U4A output is disconnected and you are looking at the bode plot of only the last U1 stage?

Can you send the original circuit where you are trying to running a complete system differential mode bode frequency plot?   I took out the second stage (since this is where I saw some of the errors) and ran a differential mode bode plot on stage 1 only and here is what I got for AC response.

When I attempted to wire up the second stage back in I was getting some railing in transient mode (which would prevent the bode plot from deriving an appropriate AC response) so I think I wasn't hooking it up like it was originally intended or there is some other circuit error going on...

Regards,

Pat

[didymus7]

Sorry for the delay, just getting time to get back to this.

Unfortunately, you can also get partial (and good) results from the output of U3A to the end of the circuit.  The total combination is what is making Multisim barf.  BTW, the simulation is a circuit that had been operating/sold for nearly 35 years (by my company).  Just pointing to the circuit and saying 'Bad, bad', is not going to do anything here.  The total simulation does work fine with the O-scope, why not the Bode plotter?  I have had other circuits where I tried to use the Bode plotter as if those really were differential inputs and it does not work.  I would think that the - inputs would not always have to be put to ground, but, from my experience with Multisim, that's the case.  Unless, of course, I'm missing some guidelines ;-).

Any help would be appreciated, but for now I really have to work around this.

Thanks...Steve