Multisim 14 - Transient Analysis bombs out after some amount of time

[KennyL]

Hi all! For multisim 14, I notice that the "transient analysis" using the function generator (in squarewave out mode and triangular output mode) provides results up to a certain amount of time, but the results eventually just cut-out (goes to zero)....eg such as after 0.1 seconds.

I've attached a screenshot of my basic (mosfet) circuit schematic, and a multisim14 project file, and a screenshot of my simulation graph file (which shows that the simulation just drops and flatlines at zero after a relatively short amount of time).

I notice that the simulations will cut-out (bomb out) when the function generator is in squarewave output mode and in triangular wave output mode.... while no issue occurs when the generator is in sinewave output mode.

Could someone let me know if I can do something to fix up this issue?

Thanks in advance!

[KennyL]

UPDATE: It looks like the cut-out is also unfortunately occurring with the function generator set to sinewave output as well. This occurs for the 'transient analysis'.

Any help or recommendations to fix this is greatly appreciated. Thanks in advance again!

[DiegoH214]

Hello KennyL

It seems you used a 5s for Maximum time. Isn´t that too large? As well, the transient is for current. And when using default values the response seems better. See attached. 

what were you expecting to get for the transient analysis?

Please take a quick look to this document.

http://www.ni.com/tutorial/12774/en/

Diego H 

National Instruments

[KennyL]

Hi Diego! Thanks for your kind reply!

I did use a 5 second maximum time. However, the problem I'm having is that the output signal - in this case i(L1) - just drops out (goes to zero) after about 100 millisecond.

It is a transient analysis, but I'm expecting that the 50 Hertz current should not just disappear at 100 millisecond. I'm expecting that current signal (having a 50 Hz frequency) should just continue in time, and not just vanish at 100 millisecond.

My aim is just to plot the current, which I expect to remain visible (not vanish) up to a time of say 5 seconds.

I've attached a screenshot of my plot of current, and I've indicated (within the screenshot) the problem that I'm getting, which is the unexpected vanishing of the current at about 100 millisecond.

Regards,

Kenny

[DiegoH214]

Hello KennyL

I tried replicating the issue with different circuits and ir worked correctly (I ran it for 10s), have you tried changing the ICs with others just to tested?

Diego H

National Instruments.

[KennyL]

Hi Diego! Thanks for trying to help me out here.

I made some progress here. I found that my original NI Multisim 14 schematic bombed out at about 100 millisecond.

But, when I deleted (removed) the tektronix oscilloscope from the schematic (virtual circuit), the simulation bombed out at approximately 1.4 seconds. That is, the simulated waveform I(L1) dropped (permanently) to zero at 1.4 second, which is at least better than 100 millisecond. But the issue was still there. The simulation was still bombing out.

But, when I went back to the original schematic again, and then bypassed (shorted) the 1 Ohm resistor (that was originally in series with inductor L1), the transient analysis simulation worked properly (ie. behaved normally up to the user-specified simulation time).

So, it appears that the software can't handle the 1 Ohm resistor in series with the inductor (for this simulation).

I've attached the .ms14 file that works, which represents the schematic with the 1 Ohm resistor removed (and the 1 Ohm resistor is replaced with just a wire).

Diego ..... so your Multisim 14 doesn't bomb out when you tried my .ms14 file from my original post (with the 1 Ohm resistor included)?

[DiegoH214]

KennyL

Yes. when I tried your schematic it presented the same behavior. But in other circuits it worked for the time we wanted.

That´s why I suggested changing some components. Sometimes transistors or these kind of components "blow out" during simulation. 

The behavior you presented makes sense. 

Glad to know you were able to find the source of the issue.

Please see attached:

http://digital.ni.com/public.nsf/allkb/A65253BD02F585CD86257A6B0049945D

Diego H

National Instruments.

[KennyL]

Hi Diego! Thanks for confirming that you did a simulation based on my original schematic, and that you saw the same behaviour (issue).

The issue occurs when I the resistor is present - in series with the inductor. The issue disappears when the resistor is replaced with a short circuit.

Would that be an indicator of a bug or glitch in the NI Multisim 14 simulator software, that needs to be fixed?

You kindly provided a weblink that relates to 'simulation error'. However, in my simulation, there is no pop-up message that indicates a simulation error. Instead, the results just drop to zero (unexpectedly), so it points towards a software glitch - maybe. Could the NI software team look into this?

Thanks again Diego!

[DiegoH214 · Accepted solution]

KennyL

Actually, it does not seem like a bug. But a normal Multisim behavior. Depending on how you select a component, we have two diferent scenarios, both are programmed in the SPICE model. 

If we use a normal Multisim native component, then the behavior of "blowing out" which is the one we were noticing is expected. 

If you select a component from a manufacturer. The SPICE model probably is going to behave differently, ignoring conditions that in real life will not work (like having 1000 A running through a resistor)

I have exchanged the resistor on your circuit and the transient worked correctly. So, I would advise this approach.

What do you think?

Diego H

National Instruments

[KennyL]

Thanks very much Diego!

I understand what you mean now. I took a look at the resistor type that I had originally used, and it turned out to be from:

group BASIC -> Rated_Virtual -> Resistor_Rated

But now, after selecting a resistor from a different area, namely.....

group BASIC -> RESISTOR ...... everything is now working nicely. So, the resistor from Rated_Virtual can have issues in the transient analysis due to the virtual power handling (Watts) value being too low. Increasing the power handling value (eg, making the Maximum Rated Power value equal to say 1000 Watt, just as an example) resolves the drop-out issue. I didn't realise that I had selected the Rated_Virtual component before. I was oblivous to it. I only just understand what happened now. Although, I didn't get any pop-up message to say that my 1 Ohm resistor 'virtually blew up' or 'virtually burned out' hahaha. If this virtual burn-out is really what happened, then Multisim is very impressive - although, a pop-up message to warn about this would have sorted it out quickly.

Thanks for your time to help me out with this Diego. Very much appreciated. I'll make sure to use the general Multisim resistors from now on.