New Paramour IIs have arrived...

[Paul Joppa]

OK, we may need to archive this somehow. I can't believe that it's never been compiled before!

Paramount stock plate choke: 24H, 70mA, 330 ohms
Magnequest PC-7 27H, 70mA, 430 ohms

Paramour II (obsolete) stock plate choke: 10H, ca. 60mA as plate choke, 270 ohms
Magnequest BH-6: 40H, 50mA, 540 ohms

Stereomour stock plate choke: 20H, 50mA, 590 ohms (black wire tap)

SEX stock plate choke:30H, 35mA, 750 ohms
Magnequest BH-2: 50H, 40mA, 540 ohms

(Not yet released) SEX-2011 stock plate choke: 40H, 35mA, 830 ohms (red wire tap)


(The parafeed capacitor is estimated from the choke inductance, not the output transformer inductance which is generally so much higher as to have little effect.)

 

[jrebman]

Thanks, Paul.  Very useful information to have.

-- Jim

 

[jrebman]

Decided to hold on to these after all, but will build them as basic 45s and then maybe later think about upgrading to SR-45s.  I will be ordering an exo-45 and bpc-15 40 mA, and will take it from there.  Yes, will have to rework the operating point, but that's fine.

-- Jim

 

[Grainger49]

I'm curious, does the Xl of the choke equal the Xc of the parafeed cap giving a complementary complex impedance?

I will have to store the choke data somewhere.

  .  .  .    (The parafeed capacitor is estimated from the choke inductance, not the output transformer inductance which is generally so much higher as to have little effect.)

 

[Paul Joppa]

I'm curious, does the Xl of the choke equal the Xc of the parafeed cap giving a complementary complex impedance?

I will have to store the choke data somewhere.

  .  .  .    (The parafeed capacitor is estimated from the choke inductance, not the output transformer inductance which is generally so much higher as to have little effect.)

That's part of the story.  :^)  The complementary impedance can only happen at one frequency, and that frequency is not arbitrary.

The first thing I did was to simplify everything to just the source and load resistances, the plate choke inductance, and the parafeed capacitor. The main assumption is that the output transformer inductance is high enough that it appears as a resistance when loaded by the speaker which itself is assumed to have a constant resistance with no reactance. These assumptions make it clear that the results are a starting point, not the final word!

Then I non-dimensionalized the mathematics - normalizing all reactances to the load resistance and the frequency at which the normalized inductive reactance is unity (XL/R = 1). The capacitive reactance is similarly normalized. Then I ran some simulations, looking at the load on the tube as well as the frequency response. That's because if the load on the tube gets much lower than the normal design value, the tube will distort even if the transformer does not. The best compromise in my judgement occurs when XC/R = 0.5, which after undoing the normalization means C = 2*L/(R squared). That's my starting point value though I may adjust it for various practical reasons in specific designs.

 

[Grainger49]

Thanks for walking me through the math.  It has been several decades since I have done any of this.