LTspice

For a resistance value, use a variable name enclosed in curly braces (to solve the variable before each run), e.g. {RV}. Then add a parameter step command to the schematic, e.g. .step param RV 1 10k 1k using the Edit, Spice directive menus.

To make a potentiometer, define the values the two halves of the pot (on each side of the wiper) with a position formula in curly braces), e.g. {1+Rtotal*position} and {1+Rtotal*(1-position)}.

Then add a parameter to define Rtotal, .param Rtotal=10k and a step command for the position, .step param position 0 1 .1

The extra ohm in the formulas prevents the section resistance from reaching zero ohms which blows the math up, and sort of represents the wiper resistance.

If you want to run single position runs, add a position parameter definition .param position=.5 and comment out the step position by right clicking on the command and clicking the comment button.

Later you can comment out the position parameter and uncomment the position step command, if needed.

Interesting you should ask. I was 100% certain that there was a potentiometer. But, using the latest version, like you, I could not find it.

Tam

Posted many moons ago...

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

Nope! From hearing Mike Engelhardt (the author) answer the question, I'm pretty sure it's that way on purpose -- he really does prefer two resistors, with one set to, e.g., {10k*x} and the other set to {10k*(1-x)}. (And then you use a .param statement to have LTspice step through the various values of x you're interested in.)

I'm told that someone has posted (presumably on the Yahoo! group) a potentiometer just bundling up the two resistors into a single package (as Jim's .pdf demonstrates) in case you want something that looks "nicer."

It's interesting that Mike seems to feel so strongly about pots one way whereas John Warner of SIMetrix fame seems to like pots so much he added the ability to use the cursor keys to "turn" the pot and automatically re-run simulation each time!

---Joel

I don't suppose that Linear Tech puts many pots on their chips.

I'm still toying with a way to delay a signal 0 to 5 ns with one turn of a pot, without wrecking the bandwidth and eye diagram. Maybe it can't be reasonably done.

John

Does the pot adjust the clock frequency of a delay line?

No, that would be easy, but we want picosecond jitter and at least, say, 50 ps settability.

I have seen LC delay lines that used varicaps instead of fixed caps, but that's really nasty. Base delay is big, delta-delay is small, impedance changes with delay, and tc's are rotten. The HP 8133A pulse generator does that, and it's a design worh studying, for its performance and sheer monstrosity.

John

"Joel Kolstad" schrieb im Newsbeitrag news: snipped-for-privacy@corp.supernews.com...

Hello,

I have posted a few months ago a second type of pot into the Files section of the Yahoo group. The first pot has been controlled with a parameter while the latest example has a voltage controlled input.

Membership is free!

Late at night, by candle light, John Larkin penned this immortal opus:

- YD.