Maximum Voltage of Scope Orange Channel

SimulIDE-R1657_Win32
It seems it is just a display bug.

The maximum voltage of the orange channel (Ch 3) on the expanded scope (left side) is positioned downwards.





For instance, I added the x10 function to let the trace be seen properly; otherwise, the ‘auto’ doesn’t work as expected.

As a test, I also let the yellow channel (CH 1) to display the voltage on the two capacitors (both are connected to ground because this not the final circuit). There was no yellow trace even by using ‘auto’. But the scope cursor read its v(t) values correctly though it is invisible.

The maximum voltage of the orange channel (Ch 3) on the expanded scope (left side) is positioned downwards.

I see... voltage values are in the wrong place.

As a test, I also let the yellow channel (CH 1) to display the voltage on the two capacitors (both are connected to ground because this not the final circuit). There was no yellow trace even by using ‘auto’. But the scope cursor read its v(t) values correctly though it is invisible.

That is because the voltage in that case is below the "Filter" value.
Set the Filter to 0 V to solve it.

arcachofo wrote:

As a test, I also let the yellow channel (CH 1) to display the voltage on the two capacitors (both are connected to ground because this not the final circuit). There was no yellow trace even by using ‘auto’. But the scope cursor read its v(t) values correctly though it is invisible.

That is because the voltage in that case is below the "Filter" value.
Set the Filter to 0 V to solve it.

First, I changed the factor x10 to x1 in the added function to restore the real voltage on the two capacitors. Then, on the expanded scope, I tried to decrease the filter 1V. It becomes zero right away by pressing the down arrow once. And all traces went crazy, sort of quasi-crash.
So, I had to exist the program and re-run it. This time and after I got a zero for the filter, I entered 0.050 for the filter by using the keyboard. The traces were displayed properly. We may conclude that setting the filter to 0V has a bug for the time being.

For instance, changing the formula of the function didn’t let SimulIDE ask to save (or disregard) the changed circuit at exit.

It's not a bug.
If you keep the "Auto" activated it will auto-scale to the detected frequency.
If you set Filter to 0 it can pick very small voltage changes as rising edges.

In general  if you already reached a good scale is better to deactivate Auto.

Setting Filter to 0 might not be the a good option depending on the case, try a value lower than the amplitude of the wave and higher than than the noise.

For instance, changing the formula of the function didn’t let SimulIDE ask to save (or disregard) the changed circuit at exit.

yes, there are many cases like this.

arcachofo wrote:It's not a bug.
If you set Filter to 0 it can pick very small voltage changes as rising edges.

Setting Filter to 0 might not be the a good option depending on the case, try a value lower than the amplitude of the wave and higher than than the noise.

It seems you couldn't reproduce the crash by setting the filter to 0 while running the circuit attached on post #1... right?

On my side, when this crash of the scope starts, the scope's buttons (mainly the 'auto's) don't respond properly. Meanwhile, the traces are displayed at random.

In my humble opinion, the minimum volt of the filter needs to be limited to a value close to zero but not zero. Testing the filter with 0.001V, the traces could be displayed properly, but not with 0.000V.

I could reproduce what you describe, but that is not a crash.

The Oscilloscope is just picking very small voltage spikes and trying to adjust the view to those very high frequencies.
Depending on the computer this can overload the cpu and cause the GUI to become slow.
But the Oscilloscope is doing what is supposed to do.

In this case you know that you can't use Filter = 0, specially in combination with Auto.
And you found that Filter = 0.001 is ok.
But that is specific for this case, in other cases the values could be different.

And also always uncheck Auto once you get the desired view in cases like this where you want to focus in some frequency.

arcachofo wrote:I could reproduce what you describe, but that is not a crash.

You are right, it is not a crash. The scope just cannot restore its synchronization in any way. Therefore, if this happens, to continue the simulation, I have to close the circuit and re-load it.

The scope just cannot restore its synchronization in any way.

You can do it: just click on the "up" button in Filter.

arcachofo wrote:

The scope just cannot restore its synchronization in any way.

You can do it: just click on the "up" button in Filter.

Following you advice, I re-tested the case in which the filter is set, likely by mistake, to 0.000V (its auto lower limit).
By clicking the up key, the filter returned to 0.100V. Then, by clicking on the ‘auto’ buttons, one by one, synchronization returned to the traces which were synchronized previously. And by setting the filter to a value which is a bit higher than 0 as 0.001V, the weak signal re-appeared properly.

Truth be said, if the filter will be set to 0 unintentionally, I will simply close the circuit and re-load it

Anyway, perhaps someday, I will know a circuit (a case) when setting the filter to 0.000V is important or it doesn't disturb the traces on the scope. Because I also believe that you have good reasons to keep the lowest limit at 0.000V

Anyway, perhaps someday, I will know a circuit (a case) when setting the filter to 0.000V is important or it doesn't disturb the traces on the scope.

In most cases setting the Filter to 0 does not "disturb" the traces, for example add a WaveGen component and connect to an Oscilloscope.

Filter = 0 means no filter.
There are many potential reasons for disabling the filter, for example in your circuit it is telling you that there are some small "spikes".

arcachofo wrote:

Anyway, perhaps someday, I will know a circuit (a case) when setting the filter to 0.000V is important or it doesn't disturb the traces on the scope.

Filter = 0 means no filter.
There are many potential reasons for disabling the filter, for example in your circuit it is telling you that there are some small "spikes".

Now, it is crystal clear. Thank you.

In any case I was investigating the issue.
I could find some spikes in channels 1 and 2, those are somehow artifacts of the simulation, caused by the difference between wavegen step and reactive step:

Wavegen is set to 50 Hz, 1000 steps, so it will produce a voltage change every 20 us.
But Reactive step is set to 1 us.
So capacitors have the chance to discharge a little bit between WaveGen steps, creating those spikes:



In most cases this is irrelevant, those "spikes" are around 0.1 pV in 20 us in this case.
But could cause issues like the Oscope going crazy with Filter=0 + Auto.

The issue with the Oscope can be solved just not using Filter=0.
But if needed this could be solved "synchronizing" WaveGen and Reactive steps, for example setting Reactive step to 20 us, those spikes disappear :




But I was scratching my head because I couldn't see any spikes in channel 3, which was causing the problem, but the Oscope is still going crazy with only Ch3 connected.

So I did some debugging and found the "ghost spikes".
And they are indeed ghosts, because the duration of the spikes is 0 s:



This is because there are 2 voltages with the same timestamp.
So the Oscilloscope is working ok, but there is something wrong here. I have to find the origin of that "ghost voltage".

Also found some other minor issue related to "Auto"...