Spice model for GU50 (LT-Spice)

[voltagezone]

I am looking for a spice model for the GU50 for LT-Spice.
Please help.

 

[euro21]

http://ltwiki.org/files/LTspiceIV/lib/sub/Valves.Lib

 

[voltagezone]

Thank you. In this model G3 is missing. So basically a tetrode is simulated and not a pentode. Does the model work anyway?
.SUBCKT GU50 1 2 3 4 ; P G2 G1 K (Pentode) 21-Nov-2001 X1 1 2 3 4 Tetro MU=73.53 EX=1.150 KG1=194.7 KP=13.28 KG2=4500 + KVB=30.3 VCT=0.00 RGI=2k CCG=14P CPG1=0.1P CCP=9.2P .ENDS

 

[cogsncogs]

Yes. The model assumes G3 is at 0V with respect to the cathode. Or I should say G3 is tied to the cathode or at 0V.

 

[voltagezone]

I got it working. Results seem good.
Thank you.

 

[jazbo8]

It works but far from accurate, here is a comparison of the model vs. the datasheet.

An externally hosted image should be here but it was not working when we last tested it.

 

[RajkoM]

It works but far from accurate, here is a comparison of the model vs. the datasheet.

An externally hosted image should be here but it was not working when we last tested it.

Who cares. GU50 pentode curves are inconsistent anyway.
What to choose?

 

[jazbo8]

Point taken, since in this case we do not know which datasheet the model was based on... BTW, where did the image on the left came from?

 

[RajkoM]

... BTW, where did the image on the left came from?

If I remember well, from Wavebourn.

 

[jazbo8]

If I remember well, from Wavebourn.

Cool, thanks, found it here.

After some sleuthing, I think the SPICE model posted above came from Eugene Karpov with some slight modifications to the parameters.

 

[ja2dhc]

I think the SPICE model posted above came from Eugene Karpov with some slight modifications to the parameters.

I have down-loaded Eugene Karpov's model and drawn the
Ep-Ip curve of the triode connection by TINA simulator.

 

[jazbo8]

I think Mr. Karpov built his model not based on datasheet but by tracing an actual tube, so the model would need to be adjusted to match the datasheet. I am also interested to see some actual traces of the Chinese FU50 tubes to see if they are even close to the old LS50 and GU50. Then I will build the pentode model using Ayumi's pctube.

 

[RajkoM]

Triode strapped GU50 curve is much easier than pentode.
Anatoli presented excellent triode curves from tracer.

 

[jazbo8]

Perhaps I left out something, to build the pentode model using Ayumi's method, we need to have the triode strapped (not pentode) curves, would you happen to know which tube Ale used to make the above trace? If not, I will check with him.

 

[RajkoM]

Perhaps I left out something, to build the pentode model using Ayumi's method, we need to have the triode strapped (not pentode) curves, would you happen to know which tube Ale used to make the above trace? If not, I will check with him.

Chek with Ale. But I think this curves from tracer presented Anatoli aka Wavebourn.

 

[smoking-amp]

Some more GU50 triode curves. Modified pentode too.

http://www.diyaudio.com/forums/tubes-valves/160240-suppresor-grid-used-feedback-3.html#post2071537

 

[RajkoM]

Some more GU50 triode curves. Modified pentode too.

http://www.diyaudio.com/forums/tubes-valves/160240-suppresor-grid-used-feedback-3.html#post2071537

Nice, thanks.

 

[jazbo8]

Some more GU50 triode curves. Modified pentode too.

http://www.diyaudio.com/forums/tubes-valves/160240-suppresor-grid-used-feedback-3.html#post2071537

Hey, how did I miss those? Which make was the GU50? Just curious if those curves match up with any of the commonly available datasheets...

 

[RajkoM]

... model using Ayumi's method.....

jazbo8,
based on my several simulation/finished project, I can say that Ayumi`s pentode models faultlessly good and precise.
I hope you share with us your model based on Ayumis method.
Thank you.

 

[smoking-amp]

"Hey, how did I miss those? Which make was the GU50? Just curious if those curves match up with any of the commonly available datasheets... "

They are labeled ry-50 and have two sets of three curved lines that merge to form an upside down tree with a short arrow above pointing up. Wavebourn sent them over to curve trace. Also has a diamond shape mark with OTH 34 inside, probably some test code.