
In preparation for part two of our circuit simulator series, here’s a quick puzzle for you. Click away and see an instant summary of the answers so far.
Check back in few days for:
Everything you always wanted to know about SPICE*
*But were afraid to ask.
OK A few days are up. Here’s the answer at Circuit Simulation Part Two – How Various Components and Analyses are Handled
PS: In case you missed them, the puzzles and solution for events 1 and 2 are here
![[del.icio.us]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/delicious.png)
![[Digg]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/digg.png)
![[Facebook]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/facebook.png)
![[Google]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/google.png)
![[LinkedIn]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/linkedin.png)
![[Twitter]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/twitter.png)
![[Email]](http://signal-integrity.tm.agilent.com/wp-content/plugins/bookmarkify/email.png)

There actually are two correct answers!
Hint: V=I*R and I=G*V
You got me there, Larry! I meant there to be only one right answer. What I should have said is “Non-linear resistor for the case that its current is a well behaved function of its voltage.” Complicated stuff, this circuit simulation
Pure nodal approach would fail with ideal voltage source for sure, and sometimes with nonlinear resistor, if its currect cannot be expressed via its terminal voltages. BTW, strictly speaking, short circuit can also be considered as special case of the non-linear resistor, why not? Same, as voltage source with V=0
This puzzle reminds a test question for college students in circuit theory
Hi Vladimir, Thanks but you’re not supposed to give the answer away!
– Colin
Non-linear resistor, am i right?
Hi Benny, Sorry! Pure nodal analysis can be used for non-linear resistors (assuming the typical case of one with current that is a well behaved function of its voltage).
I vote voltage source. The number of votes for open-circuit is surprising. For DC simulation an open circuit node would require just one resistive connection. For transient, I think an open circuit node would just need one connection. If the simulator has a gmin setting, than I don’t see a problem at all.
Hi Juan, I think you might be on to something
In nodal analysis, you can deal with any open circuit between any two nodes for free! (Pattern of four zeros stamped into the conductivity matrix: Any node with no connections at all can be dropped from the analysis as an unphysical “don’t know” floaters)
Good point. I was thinking having a no connection node would result in a singular matrix.
Who am I to contradict Larry Nagel?
Voltage source is certainly one answer. That is one reason why Larry chose modified nodal analysis for Spice.
Thanks Larry, your invention was worth half a career for me.
Probably it’s kind of word game. The correct answer is BSIM “model”, since the model itself cannot be modeled using nodal simulator; the model is developed using analytical and other approaches.
Hi Pavel, no I didn’t mean it to be a word game. So to clarify, for this puzzle “model, simulate, run” are interchangable verbs and “model, component, branch equations” are interchangable nouns. And model creation is a separate activity outside of the scope of this puzzle.
hth
– Colin
I think then it’s ideal voltage source (if you meant ideal) – becuase the nodal approach only allows the current-defined branches and the node voltages to be included into the solution vector. In case of ideal voltage source the current is not defined for its branch and for the modeling the ideal voltage source using nodal approach we have to include shunt resistor in schematic.
Circuit Simulation - Part Two - How Various Components and Analyses Are Handled // Apr 20, 2009 at 7:52 pm
[...] Puzzle Event 3 – Nodal Circuit Simulator [...]