Spice and distortion

[Michael A]

For relatively simple triode circuits, how close do you find measured distortion to compare modeled distortion in spice. In magnitude and order?

I'm using LTSpice. It seems the distortion in the function generator is so high as to make it useless so I have been feeding a wave file of a basically distortionless 1kHz sine to make my measurements. I have only recently started using spice so I haven't yet had a chance to compare something I have modeled to something I have actually built.

Thanks,
Michael

 

[bcarso]

My program has some pretty simple ideal Child-Langmiur models, but there are some posted ones that could be imported that are supposedly more accurate.

Haven't been able to do a close comparison of the sims vs. the circuits though, although things haven't been too far off. I'll find out soon though as a system with some 6H30's is nearing completion. And the predicted distortion is quite low, almost too good to believe. To some extent it depends on matching though, and I haven't tried to introduce mismatch into the simulations.

I think one of the significant things that's neglected in models are the dynamics of the space charge around the cathode, which will get important at large signals.

 

[NewYorkDave]

Brad, you use Circuitmaker, right? The only 3F4 tube models I've found that are trustworthy enough to be usable are those written by a fellow on Usenet named Robert Casey. Look him up. I use his 12AV7 model all the time and so far, it's been remarkably accurate.

*Vacuum Tube Triode (Audio freq.) pkg:VT-9 (A:1,2,3)(B:6,7,8)
.SUBCKT X12Av7A A G K

* ANODE MODEL

BLIM LI 0 V=(URAMP(V(A)-V(K))^ 1 )* 0.0037

BGG GG 0 V=V(G)-V(K)--0.5
*below is variation of grid slopes
BRP1 RP1 0 V=URAMP(-V(GG)* 0.07069 )

BRP2 RP2 0 V=V(RP1)-URAMP(V(RP1)-0.999)

BRPF RP 0 V=(1-V(RP2)^ 1 )+URAMP(V(GG))* 0.1

BGR GR 0 V=URAMP(V(GG))-URAMP(-(V(GG)*(1+V(GG)* 0.012937 )))

BEM EM 0 V=URAMP(V(A)-V(K)+V(GR)* 28.093 )

BEP EP 0 V=(V(EM)^ 1.4 )*V(RP)* 0.00001963

BEL1 EL1 0 V=URAMP(V(EP))

BEL EL 0 V=V(EL1)-URAMP(V(EL1)-V(LI))

BLD LD 0 V=URAMP(V(EP)-V(LI))

BAK A K I=V(EL)

* GRID MODEL

BGF GF 0 V=(URAMP(V(G)-V(K)--0.5 )^1.5)* 0.00012

BG G K I=V(GF)+V(LD)

* CAPS

CAK A K 0.00000000000024

CGK G K 0.00000000000323

CGA G A 0.0000000000019

.ENDS X12Av7A

And here's a 12AY7 model of his.

*Vacuum Tube Triode (Audio freq.) pkg:VT-9 (A:1,2,3)(B:6,7,8)
.SUBCKT X12Ay7A A G K

* ANODE MODEL

BLIM LI 0 V=(URAMP(V(A)-V(K))^ 1 )* 0.0037

BGG GG 0 V=V(G)-V(K)--0.5
*below is variation of grid slopes
BRP1 RP1 0 V=URAMP(-V(GG)* 0.009869 )

BRP2 RP2 0 V=V(RP1)-URAMP(V(RP1)-0.999)

BRPF RP 0 V=(1-V(RP2)^ 1 )+URAMP(V(GG))* 0.1

BGR GR 0 V=URAMP(V(GG))-URAMP(-(V(GG)*(1+V(GG)* 0.012937 )))
*below is grid sensitivity
BEM EM 0 V=URAMP(V(A)-V(K)+V(GR)* 45.093 )
*below ^is grid linearity,*plate current at 0Vg
BEP EP 0 V=(V(EM)^ 1.3 )*V(RP)* 0.000007503

BEL1 EL1 0 V=URAMP(V(EP))

BEL EL 0 V=V(EL1)-URAMP(V(EL1)-V(LI))

BLD LD 0 V=URAMP(V(EP)-V(LI))

BAK A K I=V(EL)

* GRID MODEL

BGF GF 0 V=(URAMP(V(G)-V(K)--0.5 )^1.5)* 0.00012

BG G K I=V(GF)+V(LD)

* CAPS

CAK A K 0.0000000000006

CGK G K 0.0000000000013

CGA G A 0.0000000000013

.ENDS X12Ay7A

 

[bcarso]

That does look a bit more detailed than the circuitmaker ones. I imported someone's model for the 6H30---I forget whose. The canned ones with the program were very limited---I think just AX7, AU7, and SN7.

 

[Larrchild]

bcarso, Take the more refined model and run the 2-tone IMD meltdown test on it now in Circuitmaker.=)

While I do the real one.

 

[Michael A]

I've been using mostly models generated by tracing the curves with Andrei Frolov's program CurveCaptor. I think they have all been based on an 8 parameter Koren model. It will output the spice in different dialects so I guess I will have to compare 3f4 with LTSpice and see if I can convert the models NYDave posted. Seems to be working well, just wondering about the reality of the distortion levels.

Thanks,
Michael

 

[bcarso]

[quote author="Larrchild"]bcarso, Take the more refined model and run the 2-tone IMD meltdown test on it now in Circuitmaker.=)

While I do the real one.[/quote]

I'm in heavy sand mode right now, but all in good time, as the Witch says about returning the little dog in The Wizard of Oz...