• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Vacuum Tube SPICE Models

Question 1

I looked, though not finding, a source for potentiometers. I followed the instructions, but...
IMG_2523.jpeg


Here are the original formulas, suspect where the fault is as all the nodes are connected in the above test circuit.
IMG_2525.jpeg
 
SPICE models are just text files. As such, these files often have a .txt extension, although you'll see others like .sub for subcircuit, .lib for a library file, etc. The forum doesn't allow uploading files with certain extensions so it's common for a lot of model files uploaded here to carry the .txt extension.

To attach these files to your schematic you need to "include" them in your .asc file. This is just a matter of adding a statement like ".inc model.txt" somewhere on the schematic. Then store this model file in the same folder as your .asc file so that LTspice can find it. There are other locations that will work but if you are just starting out with LTspice it might be best to keep things simple at first.

I hope this helps a little.
 
  • Like
Reactions: 1 user
To attach these files to your schematic you need to "include" them in your .asc file.
Can they be in the folder with the schematic of the circuit? I am unable to open the tube directory as a .mdb file extension.

This is just a matter of adding a statement like ".inc model.txt"
You mean a Spice directive?

I hope this helps a little.
Still have the bug even with tubes that work fine in the schematics José (@jcalvarez ) generously gave me, so something other then the individual tube (successfully tube rolled in other schematics).
 
Can they be in the folder with the schematic of the circuit? I am unable to open the tube directory as a .mdb file extension.
Yes, I believe I said that in my post.
You mean a Spice directive?
Yes, an .inc statement is a SPICE directive.
Still have the bug even with tubes that work fine in the schematics José (@jcalvarez ) generously gave me, so something other then the individual tube (successfully tube rolled in other schematics).
I don't understand the question. If you have an appropriate tube SPICE model which is in the same folder as the schematic file, and have "included" it in the schematic, LTspice will be able to use it.

To offer any more help, we will need to see both your .asc file and your tube models, as Sorento pointed out.
 
  • Like
Reactions: 1 user
I don't understand the question. If you have an appropriate tube SPICE model which is in the same folder as the schematic file, and have "included" it in the schematic, LTspice will be able to use it.
Unless the whole darn program has a bug with running any tube.

I apparently forgot the picture, though the same error as above.
IMG_2535.jpg


I could change tubes, with the .inc statement, until I was blue in the face and get the same error.

To offer any more help, we will need to see both your .asc file and your tube models, as Sorento pointed out.
I posted the .asc file, though why ask me for something that is not possible? The tube models are not an accepted extension.
 
Disabled Account
Joined 2013
I have edited asc so it runs without errors. In order for receiver to work you need to mutual coupled inductors using K statement (K1 L1 L2 0.9) with a suitable coupling factor. So that will make it regenerative that radiate at certain frequency you can actual pickup by radio. If you can point to website that explain how it works we may help with you more.

Found it: https://www.cool386.com/12v_6BL8/12v_6BL8.html
 

Attachments

  • 1_tube_regenerative_tuner_6BL8_v2.asc
    6 KB · Views: 53
Last edited:
  • Like
Reactions: 1 user
I have edited asc so it runs without errors.
Interestingly it corrupted in the file transfer, though still runs.
IMG_2539.jpg


I would love to know what "floating" means, please. We have had this before and puzzling as the wires are all connected, no floating in my understanding of the word.

In order for receiver to work you need to mutual coupled inductors using K statement (K1 L1 L2 0.9) with a suitable coupling factor. So that will make it regenerative that radiate at certain frequency you can actual pickup by radio. If you can point to website that explain how it works we may help with you more.
Well, I was simply using logic not finding an answer. Now I am kind of understanding, though feeling an inability to be independent. That is fine, I am doing this for fun. Simply modify for other radio schematics.

Incorrect, it is the 160 volt version: https://www.cool386.com/6BL8_srrx/6BL8_srrx.html.

Last night I wanted to see if with my new knowledge I could get something to work. First run the output tube (one half of an ECL86 which is equivalent to a EL84) was not working right, the ECC83 half was not getting correct voltages, turns out petting the dog was distracting and didn't put a capacitor and resistor in the ground leg. :LOL: Felt amazing when that produced a run with voltage! Went to bed with a smile on my face.

I am going to hold off on sharing it until make the tube and regeneration corrections. Plus, feel maybe more to learn from the one tube. Learning is why I created it, after all.

What I love to learn is to build a transformer, seems the instructions I found aren't computing in my brain. I love how José creates them as blocks, unlike any instructions out there, in part because look neater and more importantly, a better model.

I have this feeling the 3μV output needs to go into a transformer to step it up.
 
This model is traced by me using CurveCaptor, Koren8 model, pretty close, maybe you like to use it, there is more than one way to get the model, included also the marker and data file for reference.

* ec88 LTSpice model
.subckt ec88 P G K
Bp P K I=(0.1271584816m)uramp(V(P,K)*ln(1.0+(-0.08785188791)+exp((2.922603012)+(2.922603012)((70.68877455)+(-2844.697934m)V(G,K))V(G,K)/sqrt((66.12972009)**2+(V(P,K)-(34.81182848))**2)))/(2.922603012))(1.087629465)
Cgk G K 3.8p ; added
Cpk P K 0.055p ; added
Cgp G P 1.7p ; added
Rpk P K 1G ; to avoid floating nodes
d3 G K dx1
.model dx1 d(is=1n rs=2k cjo=1pf N=1.5 tt=1n)
.ends ec88
This seems to be working, no errors generated. There is also a model done in 2013, post number 5 which didn't use.
 
As to the model created last night, spent a little time trying to get grain and output watts to work, realized beyond my ability at this time, though did achieve some learning. Other than that, the model does seem to be working, what do you all think?

Note the rheostat for the reaction is modeled as a resistor as at the time didn't have a potentiometer component. Additionally, the ELC86 is modeled as two tubes which supposedly are equivalent: https://www.diyaudio.com/community/threads/ecl82-or-ecl86-amplifier-decision.168695/.

Clearly I am currently inept with transformers, so unable to compare to the one tube regenerative, for now.

Here is the website and no, I don't speak French (few words of manners I somehow learned): https://sites.google.com/site/france5ndl/recepteur-a-reaction.

Over all, I have great satisfaction and smile ear to ear because able to draft a schematic, put in a tube from this archive, and achieve some results, which was part of my intent. Feel could do another, especially better once know how to do transformers and script.
 

Attachments

  • 2_tube_regenerative_tuner_EC88_ECL86_v3.txt
    5.6 KB · Views: 38
Member
Joined 2021
Paid Member
What I love to learn is to build a transformer, seems the instructions I found aren't computing in my brain. I love how José creates them as blocks, unlike any instructions out there, in part because look neater and more importantly, a better model.
I'm glad you enjoy the transformer models, but I did not create them. Robert McLean created an extremely useful spreadsheet that allows you to enter certain transformer parameters and will generate a model. I find the models a good match with a couple of Hammond transformers I've used. All credit to him.
https://www.diyaudio.com/community/threads/spice-transformer-model-spreadsheet.181578/
 
  • Like
Reactions: 1 user
I swear I hit post... :LOL:

I'm glad you enjoy the transformer models, but I did not create them. Robert McLean created an extremely useful spreadsheet that allows you to enter certain transformer parameters and will generate a model. I find the models a good match with a couple of Hammond transformers I've used. All credit to him.
https://www.diyaudio.com/community/threads/spice-transformer-model-spreadsheet.181578/
Oops, sorry about that. Thank you for the correction and help.

What do you do if do not have the information to put in? Only now I see simply states it was from an old phonograph. Still helpful as the thread does include modeling a basic step up transformer to explore.
 
Why are 1N4007 diodes not in the list under diodes? Are they not common?

Additionally, tried to find the file to add the 1N4007, however, both hidden in the interface and in file explorer.

Also, tried inserting a diode, labeling it with the model script, added the model script, however generates an error code saying can't find "D." Tried several times to edit the script and make it happy to no avail.

The reason I was doing it this way as the .asy file of the 1N4007 degenerated.
IMG_2579.jpg


And not showing up when selected.

Since present a significant challenge, is there a direct equivalent to use in LTSpice to achieve an accurate model?

In case matters, attaching the .asc file along with the schematic. This is a schematic that was provided and modified to a Morse SE.
morse-6bm8-se.gif


I warn those who attempt to run the file (it is only included to show the diode situation) there are a few errors.
IMG_2584.jpg