LTspice tutorial - Modeling vacuum tube triodes

[@brandonavo]

This video is absolutely incredible and a huge resource. Thanks for explaining the circuits in the paper and how they match up to LTspice in a clear manner. Triode stuff is hard to find these days especially explained in such a nice way. Subscribed.

[@astorina]

Thank you for sharing your knowledge and prospections Take care

[@FesZElectronics]

Thank you! You take care too! Cheers!

[@NicoV-yb7gf]

Thank you sir. You saved me A LOT of time here. I just wanted models so I can predict approximate voltage and current levels. I used "multisim" to test their model of a 12ax7 in a JCM800 pre-amp circuit. I got 668 Vp-p on the last gain stage anode/CF grid, while only using 300V B+!! I knew something was off. Probably using that model that doesn't cater for positive grid voltages, giving me -400v odd swing below ground!! Now I will try ltspice with your model. Hopefully I will get to more accurate signal voltage levels.

[@FesZElectronics]

I did not use MultiSim, but I guess its also a spice language circuit simulator. Maybe you can import the model without having to switch circuit simulators. I know getting used to new software is difficult...

[@adrianrevill7686]

Thanks for the video, I worked through how this modelling worked a few months ago and then put the project aside. This was a perfect refresher.

[@keithgreiner2686]

This is an amazing presentation and is just exactly what I am looking for. I’ve been studying it. I have a few questions. First, I am trying to replicate the example LTspice models. I am not getting the same results as shown on the screen. I think there are several issues. First, when I stop the presentation and try to read the characters on the screen, some are fuzzy and difficult to get right. Would it be possible to post high-definition images of the screens so I can be sure of the values I am typing in? Second, after I do get the programs running, I am not getting the same graphs. With the .tran 1 statement, all I get for anode voltage and current is a straight line. So I am making a mistake but I don’t know what it is. It may be that I have some incorrect statements. But it also might be that I’m not following the exact steps used to create the graphs (ie: the graph at 4:23). Can you please explain the specific steps used to create the graphs? Where did you place the cursor? Did you select voltage or current at that location? Third, in LTspice, I am not finding an StR85/10 tube. Is there a substitution that would be in LTspice? Is there a Zener diode that woujld work in this instance and still return similar results? Thank you very much.

[@thevicol]

Excellent explanation. But when we get into the details a few aspects are worth noting. 2:59 The diode and resistor is a bit oversimplified model of the grid current. It does not take into account the influence of the anode voltage. The improved Norman Koren' s model does a good job of representing the triode stage's behaviour in what we would call 'the more linear regions of operation'. It also takes into account the influence of the tube socket and leads capacitances on the frequency characteristics. But if we are interested in analysing non-linearities in terms of the distortion characteristics, it is worth to use a model created by Ayumi Nakabayashi. At least when it comes to triodes in pre-amp circuits. Power amplifiers are a separate story and here all accuracy limitations of tetrode and pentode models come out quite noticeably. There is quite an extensive discussion about tube simulation on the diyaudio forum. There were also quite interesting attempts to improve the precision of mathematical representations. It is also interesting, that on some schematics the given filament voltage is slightly lower than recommended in the datasheets. These are not big differences, but they can affect the operation of some stages. The only model that takes into account the effect of heating is the one created by Duncan Munro, although the precision of this model is not better than that of N.K. or A.N. models.

[@FesZElectronics]

Hello Vicol! I aggrege the NK is not the best model, but it was the simplest complicated model I could find when I was doing the video. There will always be a more complicated and more accurate model... That being said, I did not know about the A. N. model and all the work behind it - which is quite substantial and seems really well made. I guess the big issue is that its all in Japanese, there seems to be no English version on Mr. Ayumi's website. I will try to look int it though, thank you for pointing it out!

[@thevicol]

Yeah, Japanese :) Lucky that there are translators, because without them I wouldn't know where the 'download' button is, haha. But seriously, it's a very interesting topic and it looks like it's still being developed. I mentioned the discussion on diyaudio forum intentionally, because it's a huge source of useful links and unconventional ideas, which would be hard to find just by searching the internet. I think the topic of simulation and explaining what exactly happens inside the spice models is very important and I hope you will continue it. LTspice is a powerful tool and although there are programs with a more user friendly interface, the amount of possibilities is still hard to beat. I have been using this program for a couple of years and I still keep learning new tricks even from your videos :)

[@yveslesage8525]

Just a single word... Brilliant ! Thanks Mr Fesz

[@FesZElectronics]

I'm happy you enjoyed it! Thank you!

[@p_mouse8676]

Absolutely excellent video! In fact I was just about to model a few myself, but most tutorials are just not very clear imo. Thanks for clarifying a few thing! And yes that overlay trick works great, I use it all the time! 😎👍🏻👍🏻👍🏻

[@FesZElectronics]

May I ask what you are working on? Usually there are not a lot of circuits where tubes make sense, but I'm always looking for new and interesting circuits.

[@p_mouse8676]

@@FesZElectronics In general as a hobby, I like to build, repair and convert old radio's. Mostly because I find it very relaxing compared to a lot of PCB design I do with DSP's/Class-D power amps. A lot of stuff can just be calculated on a piece of paper or just as you go. (instead of sitting hours on end staring at a screen). Also some guitar amplifiers and audio amplifiers. It's just fun and simple (relaxing) technology to work with and speaks a lot more than just a blob on a PCB in some cases. Although I have also combined new- and old-technology (DSP with tubes etc). I don't have a preference, everything has its charm so to speak.

[@WalidIssa]

Great and Clever

[@bcsongor]

"Tubes are not efficient, but they are pretty.".. also they sound pretty good in amps :)

[@pahaahv]

When measurement is bad but it sounds good you are measuring the wrong thing.

[@AUDIOETRADIOATUBES]

Fantastic job

[@PetreRodan]

Very nicely done!

[@z3d12]

nice trick overlapping the plot from the datasheet with the one made in excel

[@KamleshGadhvana]

nice one please do simulation on audio equalizers with circuit

[@AndrewCaldarone]

How do you add the custom diode model to the diode on the schematic? I can't type in the "Diode" box so I can't change "D" to "DGK"

[@francescodilorenzo5475]

Very nice video, I'd like to know if you can give me some adevice to simulate a circuit with 4cx250, or I can use an equivalent vacuum tube, becasue I'm not an expert of Vacuum tube. Thanks

[@francescocanora860]

bravo! Is possible tutorial for pentode ? thanks

[@sim1_7]

Thank you!! Really great tutorial, thank you very much. So, can i replace the Ia/Va graphic with others of different tubes datasheets and try to find parameters for them(using your spreadsheet)?

[@FesZElectronics]

Of course, that was the whole point of the spread sheet; you can adapt it for any tube; its just that finding the right parameters is a bit difficult...

[@sim1_7]

@@FesZElectronics yes, in fact i tried after this comment ,and I have to say that after different tries I came up with a good trade off!! I downloaded an image from Google of a tube( V plate versus Current Plate) and loaded that into your spreadsheet. Changed all Vg values( to match those in the graphic) and started tweaking the parameters. A little bit tedious, but in the end I found my parameters! Thank you very much for this work.

[@sim1_7]

Do you plan to do this work for pentodes as well?

[@sanjikaneki6226]

how do you determine slew rate of a op amp in LTSpice?

[@andreybrodianski5596]

Hi. I had very accurate working model of vacuum tube 12AX7. After refreshing my PC new installation (with new blue buttons tool bar) doesn’t run old saved files with tubes. Do you know what is the issue?

[@FesZElectronics]

You can check if the model is corectly linked to the simulation - there should be an ".include" or ".model" statement with a link somewhere in the simulation - maybe the link changed; if its something else, you could post the exact error that you are getting.

[@debunker2022]

It is not convenient to adjust the coefficients in the Excel file. There is a simple and convenient program for creating spice models java jar file paint_kit.jar. Why do what has already been done?