SPICE built in primitives in an AMS simulation

There's no reason why you can't use a vcvs in an AMS simulation (I just tried it and it works fine). It sounds as if you have got some strange binding in the Hierarchy Editor for the vcvs - what do you have the vcvs set to have it's "view to use" (either in the cell table or the instance tree in the hierarchy editor)? Also, just in case it matters, what do you have your Simulation->Netlist and Run options set to in ADE (i.e which netlister; is it UNL, OSS or CellView-Based?)

Regards,

Andrew.

I set the view to use explicitly as spectre in the instance tree view but the editor did find the spectre view for all the elements from analoglib. For the netlist and run options I have selected AMS UNL netlister with xrun ( I thought that was the recommended option to choose). Right now I am having a different error. I am not able to plot voltage signals in a transient analysis and it says " it doesn't evaluate to an object that can be plotted, like a waveform or a parametric wave".  Below is a the schematic capture. The switch and the decoder are both verilog ams instances..

I opened the psf in the results browser and am trying to see if any of the signals are saved but nothing seems to be plotted vs. time. For example the vpulse instance above( connected to ben net)  has a voltage signal but when I open it up, it plots 5V against some axis labelled Design_Point. 

I know I am setting something wrong in my simulation that it gives an error that doesn't allow to plot signals. 

Some questions:

1. Which IC subversion are you using (Help->About in any of the windows will tell you)?
2. Which simulator version (the Incisive/Xcelium version - I assume it must be Xcelium since you mention run) - this will be at the top of the xrun.log file
3. Did you ask to save the signals - the default in AMS is not to save all voltages
4. Can you post a picture of the results browser and the plot you're getting? I'd like to see what the signals look like in the results browser

Andrew.

1) I am using 6.1.7

2) I am using the Xcelium version (side question has the incisive platform been renamed or is this an upgraded version but standalone?) 64bit.

3) I have checked the boxes in my output setups  window in assembler and explorer to save and plot the signals I selected in the config view. Is there another form where I should check a box to save the signals?

4) below is the signal for the ben signals which should be a a pulse going from 0 to 5 with delay time of 100n, rise and fall time as 10n, pulse width 1s and period of 2s. The simulation is run for 500ns. 

I selected the tranOp folder in the psf directory and then selected instanced V49 which is the pulse wave source above. There is I, PWR, V signal but no ben net signal. 

A few points (corresponding to your points):

1. This is the major version rather than the subversion - it should IC6.1.7-64B.500.NN or something like that
2. This also is not the version number - that's just telling you that it's the 64bit executable. XCELIUM is the next generation digital simulation engine which is the successor to INCISIVE
3. Maybe - depends on which boxes you're talking about
4. You're looking at the transient operating point results - which is just the operating point values at the end of the simulation. You should be looking in the tran results to see the transient waveforms. The tranOp will just be a single value per simulation run.

Regards,

Andrew.

1)oh ok, it is 6.1.7-64b.500.20

2) xcelium 17.04-s006

3) These are the output signals selected in ade explorer from outputs>to be plotted> select on design and then selecting the nets/signals I want to plot. I checked both the plot and save boxes.

4) the tran folder seems to be empty so I must be not saving the signals

I think this would be best solved by contacting customer support. I think we'll need to have a web screen-sharing session to see what's going on. Can you do that? Trying to guess possible failure mechanisms is pretty hard...

Andrew.

Hey Andrew,

Just to let you know that I fixed the issue. I changed the names of the input pins/nets and removed the underscores. When I simulated again using ams I could now see waveforms ( for even the ben signal connected to the vpulse). Don't know why the naming change made it identify all the signals in the setup as ones which can be plotted. Thanks for your help.

BTW is there a good RAK for Simvision?  I know there is one for VIVA. 

I don't know of a specific RAK on SimVision. It's mentioned in a few (you could search the RAKs for the keyword "simvision" on support.cadence.com - e.g. https://support.cadence.com/apex/articleattachmentportal?id=a1Od00000066P4bEAE&pageName=ArticleContent&attachId=069d0000003IHEBAA4&sq=005d0000001T3WNAA0_201872519558746 )

Andrew.