Skip to main content

Timeline for Floating node in LTspice

Current License: CC BY-SA 4.0

20 events
when toggle format what by license comment
Jul 3 at 4:53 vote accept Zahra_Alishah
Jul 1 at 7:46 answer added Zahra_Alishah timeline score: 1
Jul 1 at 7:40 comment added Zahra_Alishah You know I have to put the picture of simulation along with my calculation in Project work report. @winny but thank you very much actually I found my mistake, I had to connect the sources of M3 and M4 not their gain. (they are differential pair)
Jul 1 at 7:31 comment added winny Still does not explain why you can't fix it so that the simulation converges. Your alternatives are hand calculation or potentially the alternative solver under Tools -> Control panel -> SPICE.
Jul 1 at 7:23 comment added Zahra_Alishah because the question specifically provided the circuit and asked us to calculate only the value of I3 along with a few other parameters. @winny
Jul 1 at 7:21 history edited Zahra_Alishah CC BY-SA 4.0
add a new picture
Jul 1 at 7:20 comment added winny @Zahra_Alishah Why not?
Jul 1 at 7:18 comment added Zahra_Alishah thanks for your suggestion, but I can not change anything except the value of current source I3! @winny
Jul 1 at 7:16 comment added Zahra_Alishah thanks a bunch for your help, this question is from our university's Electronics 2 course, and our main reference is Professor Behzad Razavi's electronics book. I'd added the picture of original question. @Ste Kulov
Jul 1 at 7:09 comment added winny What happens of you put 10 Mohm from VS3 to ground?
Jul 1 at 7:08 history edited winny CC BY-SA 4.0
added 2 characters in body; edited title
Jul 1 at 6:41 comment added Ste Kulov I'm not sure. I'm not very familiar with this type of input structure. Do you have a reference for it? The problem that the simulator is having is that the node Vs3 is not well-defined because it is floating. If Vs3 is connected to a current source which is connected to VEE=-10V, then there's no way for the simulator to determine the voltage at Vs3. Flagging the current source as load helps because it allows the current source to act as a resistor which prevents the current source and voltage source from fighting each other.
Jul 1 at 6:04 history edited Zahra_Alishah CC BY-SA 4.0
added 4371 characters in body
Jul 1 at 6:01 comment added Zahra_Alishah Surprisingly, when I increase the value of current source I3 to 1mA and enable tick "active load" for current source I1 as you said, it works without any error or even an infinite loop, but why? @Ste Kulov
Jul 1 at 5:58 comment added Ste Kulov Oh ok. I believe then that the “Error” message is not the source of your problem. It’s attempting to do Gmin stepping so it looks like your simulation is failing to converge. If you open your LTspice .asc file in a text editor you can copy the text and paste it into a code block in your question. Then we can “download” your simulation and troubleshoot it further.
Jul 1 at 5:48 comment added Zahra_Alishah Actually, when I ran the simulation, it didn't stop like before, so i used "halt", but in this time it didn't show me anything, so I don't know the error stayed on I1 or moved. @Ste Kulov by using ctrl+L, no error was shown! just said: Direct Newton iteration failed to find .op point. (Use ".option noopiter" to skip.)
Jul 1 at 5:33 comment added Ste Kulov Did the error move to a different current source or did it stay on I1 after you made that change?
Jul 1 at 5:11 comment added Zahra_Alishah thanks for your advice @Ste Kulov, but it didn't work.
Jul 1 at 4:57 comment added Ste Kulov Right-click on I1 and click Advanced. Then under "Parasitic Properties" check the box that says This is an active load and then try again.
Jul 1 at 4:38 history asked Zahra_Alishah CC BY-SA 4.0