Vacuum tubes SPICE models, another way

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Did you see that paper: Modeling and Simulation of the Power Amplifier for the RHIC 28 MHz Accelerating Cavity.pdf by S. Zheng and John Keane.

They are starting from a modified Leach model and get an outstanding tetrode model. As pointed out in this paper, it is based on the constant current family of curves. It is the main problem with this model, one that want to make one must begun with the extrapolation of the constant current curves from the constant voltage curves. When this is done, the paper explain how to determine the parameters for a given tube.

I begun to make one such model for the 807, but a failling harddrive bring my work with it, so I am back to the beginning. I find that software that is of great help: Engauge Digituzer, but even so it is very time consumming to extract the datas form series of curves to make new ones.

I choosed the 807 because I use that tube and I have extensive documentation on it, including Ig2 curves which are impossible to measure for an amateur with a little budget. Anyway, I think such a model can give a much better approximation than the other models when the tube is in saturation (which is normal behaviour in a guitar amp). But the only mean to be 100% sure of that is to make one such model and test it.

What do you think about this model based on the constant current family of curves?
 
Last edited:
Ex-Moderator
Joined 2011
Perhaps you can re-post the question on this thread, I read that paper awhile back, since it's for RF application, I didn't see the relevance for audio... As for using constant current curves, that's pretty easy to do, but how do you deal with the knees which are challenging to model properly for the pentodes/teterodes?
 
It is sure than to draw the constant current curves is pretty easy to do, time consumming, but easy. Also, in their model, they have parameters for when Ua is close to Ug2 and for Ua < Ug2. And the curves they give look very similar to the curves from the datasheet. It is why at the first place I want to try this model.

Also,my main concern with tubes in audio is not Hifi but hi-end guitar amplifiers, and to modelize such amps, you have to take in account the saturation and the knees of the used tubes, both things this model seam to modelize pretty well. And yes, this paper is about RF, but many tetrod and pentods can be used both in AF and high frequency. Higher the frequency, more the parasitic caps will influence the tube, which force to use other montages, but the SPICE model remain the same independently of the frequency.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.