Additional power supply question for SE amps
Is there an advantage to using choke input on a single ended amp? When is it advantageous to use choke input?
My understanding is that choke input is constant current supply and capacitor input is constant voltage supply?
Thanks
Is there an advantage to using choke input on a single ended amp? When is it advantageous to use choke input?
My understanding is that choke input is constant current supply and capacitor input is constant voltage supply?
Thanks
ToddD,
Yes, I use choke input power supplies on my SE amps when I can.
There are lots of issues, one of them is to reduce transient ground loops, and resulting hum.
Generalizations:
A capacitor input supply has fairly poor voltage regulation with varying load.
(Especially if you use a fairly low capacitance as the first cap). Especially for some rectifier tubes, and even for solid stage rectifiers, the input capacitor has a fairly low value.
That is easier on the rectifier tube, and also easier on the power transformer primary (lower transient currents).
An example of a fairly complete cap input filter is CRCRC or CLCRC.
With solid state rectification, the output approaches 1.414 times the rms from the primary (whole primary for a bridge; half primary for center tapped full wave). Tube rectifiers drop the voltage additionally especially when the load comes up to operating current.
After the tubes that load the supply, the voltage drops according to the current, and the resistors of the filter (and DCR of the choke if there is a choke).
The choke input supply is very easy on the rectifier tube, and on the transformer secondary, choke input gives average current,versus the transient current of a cap input supply. Think of Isquared x R. The average current squared is much much lower than the transient current squared.
A choke input supply does have fairly good voltage regulation with varying load.
An example of a fairly complete choke input filter is LCRCRC.
With solid state rectifiers, correct choke, and reasonable load, the voltage is 0.9 times the rms of the secondary, or the half of the center tapped secondary with circuits as above. A tube rectifier adds voltage drop, but not as much as it would with the high transient currents of a cap input supply.
But the choke input supply must have a very good choke that is made for swinging, and it must meet the minimum critical inductance, Lcrit.
Lcrit = 350/Load in milliamps.
Example, 50mA load: 350/50 = 7 Henry.
Caution: before the output tubes warm up and draw current, the voltage will go to 1.414 of the rms secondary or half the center tapped secondary.
A Bleeder resistor will provide some current, and reduce that peak voltage that occurs at the output tubes warming up.
In all cases, both cap input and choke input power supplies need bleeder resistor(s).
Safety First!
One of my low power push pull class AB power amplifiers has about 8mA in the bleeder, about 7ma in the phase splitter driver tube (when it warms up), and about 48mA total from the output tubes when they warm up. That is 63mA.
350/63mA = 5.6 Henry
But I use a 20 Henry choke there. I would not use less than 10 Henry there, I like to have more than Lcrit.
I use solid state diodes from a center tapped full wave secondary.
The supply voltage does peak at power-up, but I use electrolytics that have voltage ratings quite a bit higher than the peak voltage at power-up.
This supply gives me very good voltage regulation when the tubes warm up, and as the music causes the current to vary widely from quiescent/quiet passage minimum current, to very loud passages (class AB), maximum current.
The power transformer runs Really Cool, and the power factor is 0.94, qualifies for those modern green regulations in terms of load on the power mains.
Chokes:
Add Cost
Spray magnetic fields (can cause interference - hum)
Heavy
Take real estate
Needs a higher voltage secondary
Cap input:
High transient current (can cause large ground loops that cause interference - hum)
Can cause heating of the secondary windings.
And I prefer to use choke input power supplies when I can.
That push pull amp with choke input B+ I mentioned above has less than 100uV hum and noise (0.0000001V), and no global negative feedback and no hum bucking.
Yes, I use choke input power supplies on my SE amps when I can.
There are lots of issues, one of them is to reduce transient ground loops, and resulting hum.
Generalizations:
A capacitor input supply has fairly poor voltage regulation with varying load.
(Especially if you use a fairly low capacitance as the first cap). Especially for some rectifier tubes, and even for solid stage rectifiers, the input capacitor has a fairly low value.
That is easier on the rectifier tube, and also easier on the power transformer primary (lower transient currents).
An example of a fairly complete cap input filter is CRCRC or CLCRC.
With solid state rectification, the output approaches 1.414 times the rms from the primary (whole primary for a bridge; half primary for center tapped full wave). Tube rectifiers drop the voltage additionally especially when the load comes up to operating current.
After the tubes that load the supply, the voltage drops according to the current, and the resistors of the filter (and DCR of the choke if there is a choke).
The choke input supply is very easy on the rectifier tube, and on the transformer secondary, choke input gives average current,versus the transient current of a cap input supply. Think of Isquared x R. The average current squared is much much lower than the transient current squared.
A choke input supply does have fairly good voltage regulation with varying load.
An example of a fairly complete choke input filter is LCRCRC.
With solid state rectifiers, correct choke, and reasonable load, the voltage is 0.9 times the rms of the secondary, or the half of the center tapped secondary with circuits as above. A tube rectifier adds voltage drop, but not as much as it would with the high transient currents of a cap input supply.
But the choke input supply must have a very good choke that is made for swinging, and it must meet the minimum critical inductance, Lcrit.
Lcrit = 350/Load in milliamps.
Example, 50mA load: 350/50 = 7 Henry.
Caution: before the output tubes warm up and draw current, the voltage will go to 1.414 of the rms secondary or half the center tapped secondary.
A Bleeder resistor will provide some current, and reduce that peak voltage that occurs at the output tubes warming up.
In all cases, both cap input and choke input power supplies need bleeder resistor(s).
Safety First!
One of my low power push pull class AB power amplifiers has about 8mA in the bleeder, about 7ma in the phase splitter driver tube (when it warms up), and about 48mA total from the output tubes when they warm up. That is 63mA.
350/63mA = 5.6 Henry
But I use a 20 Henry choke there. I would not use less than 10 Henry there, I like to have more than Lcrit.
I use solid state diodes from a center tapped full wave secondary.
The supply voltage does peak at power-up, but I use electrolytics that have voltage ratings quite a bit higher than the peak voltage at power-up.
This supply gives me very good voltage regulation when the tubes warm up, and as the music causes the current to vary widely from quiescent/quiet passage minimum current, to very loud passages (class AB), maximum current.
The power transformer runs Really Cool, and the power factor is 0.94, qualifies for those modern green regulations in terms of load on the power mains.
Chokes:
Add Cost
Spray magnetic fields (can cause interference - hum)
Heavy
Take real estate
Needs a higher voltage secondary
Cap input:
High transient current (can cause large ground loops that cause interference - hum)
Can cause heating of the secondary windings.
And I prefer to use choke input power supplies when I can.
That push pull amp with choke input B+ I mentioned above has less than 100uV hum and noise (0.0000001V), and no global negative feedback and no hum bucking.
Last edited: