Check me.please. is it correct to say that the required OPT primary inductance for a particular low frequency response can be somewhat less for and output stage run in Triode mode since the lower Rp in parallel with the reflected impedance lowers f3 for a given primary inductance?
Rp||Rl
Rp for Pentode is 5X=10X Rl. Makes little difference.
Rp for happy triode loading is usually 1/2 or 1/3 of Rl. Makes big difference.
Radiotron manual, 3rd edition, chapters 3 and 26.
Rp for Pentode is 5X=10X Rl. Makes little difference.
Rp for happy triode loading is usually 1/2 or 1/3 of Rl. Makes big difference.
Radiotron manual, 3rd edition, chapters 3 and 26.
Last edited:
High inductance is more important for power delivery than frequency response. Relatively low inductance can give satisfactory FR.
Inductance should be specified according to power delivery. If one is going to build a SE amp for example and wants full power down to 30Hz then he can start from the fact that Bdc must be bigger than Bac. Re-formulating the inequality one gets L > Vrms/(4.44*f*Idc) (where Vrms is RMS primary voltage developed into the nominal load, 4.44 is a form factor for sinusoidal regime, f is the low frequency of interest and Idc the DC plate current).
Once the inequality is satisfied that power will be achieved however it is still not enough because there is distortion to consider. One might want the ratio (2*pi*f*L)/Req at least equal to 8 or more. The equivalent resistance Req is the parallel resistance above also including the copper resistance of the transformer which can be neglected for pentodes but not always for triode. For PP OT's is not difficult to achieve even with bare pentodes, for SE it's relatively easy only for low plate resistance triodes, assuming one doesn't use local fb to drop Rp. For higher Rp devices some compromise has to be done accepting some more distortion below 50Hz with ratios (2*pi*f*L)/Req of 4-to-6. However, IF the reference frequency for full power will be in the region of 30Hz for SE, even in latter case of lower (2*pi*f*L)/Req ratio FR will always be excellent....
The reference frequency might be lower at 20-25 Hz for PP if the output power is generally below 50-60W without getting enormous transformers.
Inductance should be specified according to power delivery. If one is going to build a SE amp for example and wants full power down to 30Hz then he can start from the fact that Bdc must be bigger than Bac. Re-formulating the inequality one gets L > Vrms/(4.44*f*Idc) (where Vrms is RMS primary voltage developed into the nominal load, 4.44 is a form factor for sinusoidal regime, f is the low frequency of interest and Idc the DC plate current).
Once the inequality is satisfied that power will be achieved however it is still not enough because there is distortion to consider. One might want the ratio (2*pi*f*L)/Req at least equal to 8 or more. The equivalent resistance Req is the parallel resistance above also including the copper resistance of the transformer which can be neglected for pentodes but not always for triode. For PP OT's is not difficult to achieve even with bare pentodes, for SE it's relatively easy only for low plate resistance triodes, assuming one doesn't use local fb to drop Rp. For higher Rp devices some compromise has to be done accepting some more distortion below 50Hz with ratios (2*pi*f*L)/Req of 4-to-6. However, IF the reference frequency for full power will be in the region of 30Hz for SE, even in latter case of lower (2*pi*f*L)/Req ratio FR will always be excellent....
The reference frequency might be lower at 20-25 Hz for PP if the output power is generally below 50-60W without getting enormous transformers.
Last edited:
- Status
- Not open for further replies.