EEVblog #516 - LTSPICE Tutorial - DC Operating Point Analysis

[Ronald1962]

Hi folks,

I do have some trouble in displaying the voltage values at the nodes of the circuit.

It always shows three question marks, but not the values as Dave it did in the video. One time he also
got the question marks and solved this problem, but I couldn't find out how.

Could anybody help?

Thanks a lot in advance!

Ronald

[Ronald1962]

Solved the problem by myself!

You have to keep the mouse on the line while pressing the mouse button...

[wblackledg]

Anyway, you haven't put your home country in your profile, so I don't know whether advising you on how to simulate a crystal radio in SPICE would be considered aiding and abetting criminal blasphemy.

Well, I'm an American, if that's what you mean? 

c4757p ,  seriously, I would love to see a crystal radio / germanium diode simulation using LTSPICE.

thanks.

[c4757p]

Looks like if you poke around on Google for "germanium diode spice model", you do get some results. I tried germanium transistors a while ago and couldn't come up with anything resembling an accurate simulation. But try them and see if they work.

Or, just build the circuit...

[wblackledg]

Looks like if you poke around on Google for "germanium diode spice model", you do get some results. I tried germanium transistors a while ago and couldn't come up with anything resembling an accurate simulation. But try them and see if they work.

Or, just build the circuit...

ok, thanks.

[vk6zgo]

As a "greybeard"(Old Fart) what freaks me out is that the source is always displayed using a DC generator symbol,so the audio signal source in Dave's amplifier is drawn the same way.

Strangely,the negative supply has its negative side earthed,& a note underneath saying it is a negative voltage,
(in other words you have to multiply it by-1).

The base bias source in another part of the circuit is the right way around,however!

I suppose it makes the programming easier,having only one type of source,but it is quite annoying to those of us who have read  standard schematics on paper for many years.
 

[c4757p]

You don't have to do it that way, if you're concerned about the diagram's clarity. There's an alternative symbol for an AC source, and you can of course change V2 to a 10V source and flip it around. You can also draw your own symbols if you're really finicky; they'll work just as well as the built-ins.

[Zepnat]

Great stuff this, I actually downloaded spice after watching this video, I had heard of it before but didn't know what it was. Soon got a simple circuit going. The detail Dave put in these videos is incredible and is much appreciated.

[AF6LJ]

Good Stuff, I am looking forward to the next in the series. 

[ohmineer]

Nice tutorial, thanks for it.

I hope you continue with the rest of the series soon!

[DutchGert]

Nice tutorial, thanks for it.

I hope you continue with the rest of the series soon!

Yes, hoping for that as well!

[ssassen]

Alright, can't seem to figure this one out, not helped by the fact my Google-Fu is not what it used to be due to excessive intake of sugary treats (it is Christmas afterall). This ties in what Dave posted in his DC Operating Point Analysis video, so please excuse me for bumping a >120 days old topic.

I have a schematic drafted where I want to label a number of nodes with their DC voltages in a transient analysis.

I do this by simply left clicking on the nodes in question, et voila, the DC voltage shows up. However, it shows up formatted with 5 decimal places, which is a little too much. I'd like to round that off to one or two decimal places.

If I however right click the node and add the expression 'round(V(a1))' to round off the voltage at node 'a1' it rounds off to no decimal numbers, which in my case means that 1.38798V is rounded to 1V, that's not what I'm after, I'd like to see either 1.4V or 1.39V.

Any of you fine gentlemen know how to accomplish this undoubtedly trivial feat? Thanks and Merry Christmas!

[ssassen]

Alright, a bit cumbersome, but round(V(a2)*0.01k)/0.01k seems to work, change 0.01k/0.01k at your leisure, for example to display 925mV, this needs to read 1k/1k.

[Ian.M]

Try adding:

Code: [Select]

* restrict precision for .op data labels
.func sf(x,n)  {round(x*10**(n-1-floor(log10(abs(x)))))/10**(n-1-floor(log10(abs(x))))}
to your plot.defs file (in the same folder as the LTspice executable), then you can restrict the .op data label precision by using the expression sf($,3) in place of the default $ for the label.  Of course you can replace the 3 with any other integer for the precision in the range 1 to 8, or replace the $ with any expression calculable from ONLY node voltages and component currents.

Unfortunately the .op data labels are provided by the DC initial operating point solution, which isn't available if you use the .tran uic option and is meaningless if you use the .tran startup option.   Its regrettable that there is no way of setting them to display the results at an arbitrary time t or to display the result of .measure commands.

N.B.  The .op data labels and the waveform viewer can't 'see' any functions or parameters defined on (or included by) the schematic.  If it isn't either built-in or in plot.defs you cant use it for data display.