I just asked my students to download the latest KiCad version and realized v7 is quite different from v6. I’m simulating a simple inverting amplifier circuit with the lm741 model. Attached are screenshots of how I configured it, but as you can see in the simulation, the output is not amplified. What am I doing wrong?
I took the liberty to change the category of your post from “Software” to "Schematic / Simulation (Ngspice). I don’t know much about spice simulation myself, but that should help with getting more attention from people who do know such things.
And for the rest…
I can’t see much from a screenshot. Apparently you are a teacher, and then I would at least expect the schematic itself to look proper, without texts overlapping and such. Your -Vcc looks like it’s connected when zoomed in, but it’s difficult to see from a screenshot.
The usual culprits for a failing simulation are something wrong with the model, or an error in pin assignment / remapping. Are there any errors or warnings when you scroll down the text in the spice simulation setup?
And the pin remapping seems to be wrong. Pin 7 (V+, power) of the LM741 is apparently not connected to any pin (net) of the spice model. The pins as used in the spice models are quite often different from the pins as used by KiCad (which are the physical pin numbers of the IC packages) and you have to do the remapping correctly for simulations to work.
Also, if you zip up the KiCad project (without the backups, those are not needed) and upload it here, then that makes it a lot easier for us to have a look at the project.
Thanks, Paul. I have attached a zip file of the project, including the lm741.lib file I’m using.
Apologies for the messy schematic - it was supposed to be a quick test.
I checked the pin assignment, and pin 6 says not connected, but it’s also not giving the option to assign it as an output. I’m familiar with the pin assignment in v6 and before, but I can’t figure it out with v7.
test.zip (123.9 KB)
This is important (taken from lm741.lib):
* connections: non-inverting input
* | inverting input
* | | positive power supply
* | | | negative power supply
* | | | | output
* | | | | |
* | | | | |
.SUBCKT LM741 1 2 99 50 28
You have a symbol with 8 pins, the model has 5 only. In the right column ‘Model Pin’ of the ‘Pin Assignment’ window, change the entries, so that the model pin 1 (ni in) will move to line 3 (Symbol pin 3(+) ) etc.
In the test.zip you have posted, file test.kicad_sch is nearly empty.
The project does contain the .png’s also posted here, but you forgot to save the project.
Personal attacks on forum participants like this are really inappropriate.
Just out of curiosity: Was whatever paulvdh said provably factually wrong, or was the inappropriate part seemingly keeping the teacher to the same standard as the student…?
That was pretty much it.
Unfortunately I am not capable to do such things in a socially appropriate way.
I do not have the time to dig into this deeper, since i live on social welfare, while teachers get paid.
usually, you encounter such problems, if a symbol is not appropriate for a certain spice-model. You get these kind of errors frequently in LTSpice, if you want to plug-in “out-of-the-wildlife” opamp-models with the Standard opamp-symbols of LTSpice.
What you do then, is to fiddle the pins of the Symbol to the pins of the Model - or you are lucky and able to obtain a symbol/model-pair that has been designed specifically for combined use.
What i see in your screenshots is, that V+ is somehow unconnected. But i must admit, that i am not a “professional” KiCad-user. I come from LTSpice and am quite pro in that, but i am afraid, that i am not very helpful with KiCad. I am “learner” here, just as you are.
What i am assuming here, is a mismatch between a new Opamp-Symbol-Type in V7 and a
more or less V6 (or older) Opamp-Spice-Model.
With Spice-Suites you got to get used to the fact, that what you see is NOT what you get.
Symbols (what you see) is not the same as Spice-Models (what you get in the simulation results). You got to have an eye on both entities - basically all the time.
The LM741 is quite an “old man” - why not use recent opamps like the ADA4841 (potted gold)?
€dit 2: The “Tone” … The “Music” in this forum is generally (as anywhere else on the net) a sheer catastrophy for a *TECHNICAL" Forum. As a german i must insist on more behaviour here. Otherwise, the Heads and the Freaks will leave this planet, and move to places where the sensi grows greener.
For one reason, ADA4841 is a quite expensive opamp, students breaking off pins or frying it by inserting it backwards into a breadboard is is all too common, and using a cheap opamp is therefore usually better for use in education.
For another reason, students also have to learn that the ideal opamp does not exist. They need to be confronted with the limitations of opamps, and that is much easier with an lm741.
But I also agree to some point. LM741 is quite obsolete, and it’s even getting much more expensive because of that (uA741 seems to be cheaper). LM358 is still a widely used and extremely cheap opamp. (About 11ct in DIP package (100+) )
Students of the West:
Unite into a front for “WE WANT STATE OF THE ART MATERIAL TO LEARN WITH - NO x386 PROCESSORS!!”
ADI is quite generous (i tested it extensively! ;D)
[insert no coins]
It is the condescending, disrespectful, lecturing and judgemental tone. Nobody needs to be lectured like this. Nobody is superior, but some posters think so, obviously.
I know, and i have loads of understanding for elitarism.
It has reached a level (generally over the net) which is really F*ing ennoying, if you allow me this harsh comment. I am not made of sugar, but… [you know, what comes now…].
Too much of emotion is not goode.
Yes! This was it. I knew it had to do with the pin assignment, but I couldn’t figure out how to make this edit. For some reason, the lib file preview inKiCAd preview didn’t show the first few lines that mention the connections as it did in earlier versions - see the screenshot below.
Thanks for the pointer!
I apologize for the empty zip file. I compressed the folder while I still had the project open and assumed Kicad autosaves like most SW and apps nowadays - my bad. I honestly think this could be a nice feature to add to KiCad.
Thanks for stimulating this idea, @Hagverde !
While changing the chip would mean I have to make major changes in all course documents, I’m curious to learn more.
@paulvdh - I’ll do some reading about the chips you mentioned and come back if I have questions. We also use the AD620, any thoughts about this chip? The gain drifts really badly and we could use a better one. Any recommendations? ( I think this might be better in a different thread though)
Thanks for the support, folks! I was indeed surprised to read @paulvdh 's comment, as I asked way simpler questions and made sillier mistakes on this forum and never felt judged. On the contrary, the professional responses I received helped me learn and improve my KiCad skills.
I started using KiCad for my class a while ago (since v5) and can vouch for the immense improvements. I had to learn it myself and received a lot of help and support from this forum, which I greatly appreciate and send my students to when they need help that goes beyond my KiCad knowledge. The new version’s GUI is very different, and the average user will surely need a minute to adapt.
Also, probably content for another thread:
Aside from the pin assignment, there are two other things I found a bit confusing:
1- the components are annotated automatically in v7. So you don’t need the annotation step anymore, yet the annotation icon is still there.
2- for the sinusoidal Voltage source (VSIN): you don’t have control over the sim.parameters label’s (the purple text) visibility. It either shows all 7 parameters or none. In the past, you could decide which parameters to show/hide. How do you do it in v7.
It opens the annotation dialogue where you can re-annotate, reset the annotation and do other stuff.