[Ngspice-users] doAnalyses: Too many iterations without convergence

[BERTRAND J. <joe...@sy...>]

	Hello,

	I'm trying to simulate a circuit that contains two pentodes EL509. As
simulation results are totally different than real measures, I have
tried to test all models.

	I have written a simple script that computes Kellog network and JJEL509
model I use seems to be correct for anode voltage, grid current, but
screen current is false. Thus, I have replaced this model by another one :

.subckt 6KG6 1 2 3 4 ; A S G K
*= EL509
* Calculate contribution to cathode current
*
Eat     at      0       VALUE={0.636*ATAN(V(1,4)/20)}
Eme     me      0       VALUE={PWR(LIMIT(V(1,4),0,10000),1.5)/17}
Emu     mu      0
VALUE={PWRS(V(3,4),1-(LIMIT(-V(3,4),5,9999)-5)/1150)}
Egs     gs      0       VALUE={LIMIT(V(2,4)/19+V(mu)/4.3+V(1,4)/800,0,1E6)}
Egs2    gs2     0       VALUE={PWRS(V(gs),1.5)}
Ecath   cc      0       VALUE={LIMIT(V(gs2)*V(at),0,V(me))}
Elim    el      0       VALUE={LIMIT(V(gs2)*V(at)-V(cc),0,99999)}
*
* Calculate anode current
*
Ga      1       4       VALUE={6E-2*V(cc)}
*
* Calculate screen current
*
Escrn   sc      0       VALUE={V(gs2)*(1.1-V(at))+1.2*V(el)}
Gs      2       4       VALUE={1.2E-2*V(sc)*LIMIT(V(2,4),0,10)/10}
*
* Grid current (approximation - does not model low va/vs)
*
Gg      3       4       VALUE={PWR(LIMIT(V(3,4)+1,0,1E6),1.5)*50E-6}
*
* Capacitances
* NOTE: Change Cg1 from 37p to 25p for Svetlana EL509
*
Cg1     3       4       37p
Cak     1       4       18.5p
Cg1a    3       1       2.5p
.ends

	Of course, I have added :

.func LIMIT(x,a,b) {min(max(x, a), b)}
.func if(x,y,z) {ternary_fcn(x, y, z)}
.FUNC PWR(x,a) {abs(x) ** a}
.FUNC PWRS(x,a) {sgn(x) * PWR(x,a)}

	I have computed Kellog network and this model is better than old one.
Thus, I have replaced JJEL509 model in my circuit by 6KG6. And I obtain :

...
Note: Starting dynamic gmin stepping
Trying gmin =   1.0000E-03 Warning: Further gmin increment
Trying gmin =   5.6234E-03 Warning: Further gmin increment
Trying gmin =   8.6596E-03 Warning: Further gmin increment
Trying gmin =   9.6466E-03 Warning: Further gmin increment
Trying gmin =   9.9105E-03 Warning: Further gmin increment
Trying gmin =   9.9775E-03 Warning: Further gmin increment
Trying gmin =   9.9944E-03 Warning: Further gmin increment
Trying gmin =   9.9986E-03 Warning: Further gmin increment
Trying gmin =   9.9996E-03 Warning: Last gmin step failed
Warning: Dynamic gmin stepping failed
Note: Starting source stepping
Supplies reduced to   0.0000%
Trying gmin =   1.0000E-02 Warning: gmin step failed
Warning: source stepping failed

Transient solution failed -

doAnalyses: Too many iterations without convergence

run simulation(s) aborted

	I have tried to add some options (uic, gminsteps, gmin, altinit...)
without any success. Here is my analyze options :

.tran 1u 100m
.options method=gear reltol=1m minbreak=200ps
.control
    run
    write results.raw
    quit
.endc

	I can send my netlist but it was generated by Kicad.

	Best regards,

	JB