Having problems with DC SWEEP SOLVED

Has anybody gotten the Simulation/DC Sweep to work?

My circuit is a very simple voltage divider. https://www.circuitlab.com/circuit/5qmyxd/example_voltage_divider/

Symptoms: The Parameter label is red, with the tooltip saying "Parameter does not exist" [the field is set to "supply.v", "supply" is a node in my circuit]

If I try to run the DC Sweep, I get "DC Sweep Error, no such subcircuit"

My environment is: Google Chrome V 17, Windows XP SP 3.

Regardless of my execution problems, I love the concept! I wish this was around 30 years ago! No problems with the UI.

by ericdavies
March 05, 2012

You need to provide a voltage source connected to your labeled node "supply", then sweep the value of that source. As is, there is no voltage source to sweep.

CircuitLab seems to allow a loose syntax where a node labeled "5V" would actually be connected implicitly to a 5V source, but a text label like you used will be interpreted as a node to which you have assigned an arbitrary node name.

by CarlSawtell
March 05, 2012

Hi @ricdavies,

@CarlSawtell is correct, you need to drop in a Voltage Source element, and then sweep across the V parameter of the voltage source.

Just a small correction, if you want an implicit voltage source you need to do +5V (or -5V), the solver only interprets it as a voltage source if it has a plus or minus in front of it.

by hevans
March 06, 2012

There is the fixed circuit. The DC analysis shows exactly 500mV.

by rkuris
March 06, 2012

Hi Gents,

Thank you for your solutions. I had tried a voltage source at one point, but the real problem I'd run into then was specifying the name of the voltage source. I'd tried V1.v whereas it should have been V1.DCOffset.

Once I cottoned onto that and reversed to a voltage source, the DC Sweep worked fine and I was able to apply it to an experiment in constant current sources I'd been curious about.

by ericdavies
March 06, 2012

Dear all,

It seems that there is a problem when setting the parameter of a DC sweep simulation using Google Chrome: you cannot change it and the following message appears when running the simulation:

DC Sweep Error ParameterSet missin key ""

I changed to Firefox and everything worked fine.

Regards

by ivasan
March 07, 2012

Hi @ivasan,

Can you please post a link to a circuit that is giving you problems in Chrome. It will help me get to the root of this issue.

by hevans
March 07, 2012

Hi @hevans

I have been playing a little more and I finally managed to run a DC sweep in Google Chrome.

My initial problem was that I could not edit the field "Parameter", but now I can. I don't know what I was doing wrong.

Thanks for your interest :)

by ivasan
March 08, 2012

That's strange. If you manage to recreate the issue, do let me know.

by hevans
March 08, 2012

I looked at the difference between Chrome and Firefox (on Linux in my case). This field has a placeholder that shows up in Chrome and not in Firefox that might be what is confusing. It is a grayed text (' V1.V ') to give a usage hint. You can not select, delete or place your cursor within it because the field is actually empty. Just type over it.

by ssross
March 10, 2012

I was having the same problem in Chrome (also on Linux), but occasionally, I was able to select and change the text if I clicked on the field enough times. Dragging across the ghosted out 'V1.V' seemed to help as well, rather than just clicking in the field, although again, sometimes I could edit and sometimes not.

by joelmathewson
March 10, 2012

I too am having this same problem in Chrome, but on Windows. I see the same "V1.V", which is grayed out and unchangeable. My Chrome is Version 23.0.1271.64 m. I'll be glad to try any other tests to help figure this out.

by FazJaxton
November 25, 2012

Could you make your example unlisted or public. and post a link here?

by signality
November 26, 2012

Can you run this?

If yes, then do Simulate > DC Sweep and look at the left hand panel.

Note that the first swept parameter is V1.1 That is: the parameter V (voltage) of source V1 is to be swept.

The second swept parameter is R1.R That is: the parameter R (resistance) of resistor R1 is to be swept.

Click in a parameter box and enter the first letter of the device that you want to sweep to see a list of devices with that prefix and the parameters for each that are available.

Note that these parameter boxes are case sensitive.

This all works for me on Win XP Pro 32bit with Chrome Version 23.0.1271.64 m.

Can you do that on your setup?

by signality
November 26, 2012

I have the same problem with that sketch. The parameter box accepts focus (cursor appears in the box and the box has a blue border), but I cannot select the text or type new text. The text (reading "V1.V") is a lighter gray than the other fields, all of which can be changed.

by FazJaxton
November 26, 2012

I should also mention that this is Windows Vista 64-bit.

by FazJaxton
November 26, 2012

i) Please make your example unlisted or public and post a link here so that your circuit can be sanity checked in case it's some sort of bug in the way the editor has built the circuit;

ii) Try a different browser?

See: https://www.circuitlab.com/docs/faq/#q_specific_web_browser

iii) post a formal bug report:

https://www.circuitlab.com/forums/support/topic/8s9n9hav/how-to-use-the-circuitlab-support-forum/

by signality
November 27, 2012

i) Done:

https://www.circuitlab.com/circuit/bc98qh/555-voltage-controlled-pwm/

Since your sample circuit has the same problem in my browser but works on yours, I think mine will probably work on yours. I was trying to sweep the "Modulation" node (and trying different types of sources to see if that made a difference).

ii) I'll try firefox when I get home tonight

iii) I'll also do this this evening.

Thanks for your help! This is a fun site.

by FazJaxton
November 27, 2012

@FazJaxton: Not sure if I understood completely, but the “Modulation” node in your circuit can’t be sweep-ed.

Try to type the first letter (here “V” or “R”or “C”) into that sweep parameter box and CL should make some suggestions what could be sweep-ed in this context.

Nodes (to me) are essential but difficult to understand, as they are both, input and output, however, not to be used as “ … write to node ‘Modulation’ ”. You can read them, though, in your “expressions”.

One more thing: Take care when using such a “ideal voltage source” (V1) to supply your “real” circuits in the Time Domain Simulation” - - until you know why …

For the use of nodes and power supply see:

Read the complete project information from here: 555 pwm for fan speed control

Regards, Sancho

by Sancho_P
November 27, 2012

If I open your circuit and do:

Simulate > Time Domain

then tick the Sweep Parameter: and then type 'V' in the text box to the right of Parameter: (highlighted in red) I get a drop down list:

V1.V

V3.Amplitude

V3.DCOffset

V3.Frequency

V3.Phase

From your description, you can't get that far?

Sanity check:

You aren't trying to type something like 'Modulation' into that text box are you?

You can only enter something from that drop down list. Your starting point should always be the upper case letter of a source or a component. You can't enter net labels.

Typing 'Modulation' or just 'M' in the text box does nothing.

by signality
November 27, 2012

I'm sorry, I get it now. There is nothing wrong with the simulator. I was confused by the parameter box. Once I start typing these options do appear.

I had tried clicking in the box to move the cursor, using arrow keys to move the cursor, dragging to highlight, backspacing to remove text, and delete to remove text. This, in combination with grayed out text, led me to believe it was an unchangeable field. (I don't know why I didn't try to type characters). All of these things work on the other text fields in the same pane. This was completely my fault, but as user feedback, this box has a lot of the elements of a traditional text field that is uneditable. You might want to do something a little different to indicate to users that they should type to select a parameter name.

Also, thanks for the simulation help. I was actually trying to sweep the voltage source V3 that was driving the node labeled modulation. I'll read your example to understand more about ideal voltage sources.

Again, thanks for all the help.

by FazJaxton
November 27, 2012

A bit more info on Thevenin vs Norton equivalent sources here:

https://www.circuitlab.com/circuit/g5k727/behavioural-opamp-02/

and about ramping up and introducing glitches into sources:

https://www.circuitlab.com/circuit/b24363/ramped-and-glitched-signals-and-supplies-01/

by signality
November 28, 2012

Post a Reply

Please sign in or create an account to comment.