I remember seeing some schematics for "two wire" active electronic chokes that can be "dropped right into" the choke position of a pi filter. But its an all electronic choke that runs in series like a regular choke. Tent labs makes a commercial one, so something similar to that I can make myself.
Well, I can't for the life of me find those schematics again. Does anyone have such a circuit they can share? 50 to 100 ma 150-400 V that can simulate 10H electronically is what I'm hoping to find. Thx.
Well, I can't for the life of me find those schematics again. Does anyone have such a circuit they can share? 50 to 100 ma 150-400 V that can simulate 10H electronically is what I'm hoping to find. Thx.
pretty sure merlin describes a circuit in his pre-amps book, but I've mispalced my copy and so can't confirm!
Any active choke circuit that I have ever used required three connections, input, output, and common or ground. I'm sure that something could be designed using opamps and a pass fet (or BJT) that was self powered from and internal voltage drop, but I tend to prefer the simplicity of a capacitance multiplier with or without zener diode stabilization. Active circuits have internal feedback loops that can misbehave if subjected to a widely varying load like that seen in a class ABH push pull amp.
Yes this is along the lines of what I wanted to play around with. The schematic I was hunting for used an LR8 regulator and was published by a ham guy for the purpose of restoring old radios as a drop-in for the choke. But I think this one is similar enough! Thanks. Maybe I'll run across that LR8 one too.
Attachments
For a Newby trying to understand this two-wire active choke and how it works, I have a couple questions, can we do a quick group-analysis of how this circuit works?...
Is the drop across R3 being used to provide the power for the circuit since there is no common?
Is C1 there to inject the ripple into the gate?
Is R4 simply a grid stopper?
R1 and R2 form a divider what is its purpose?
The Zener I assume is bias the grid at 12V right?
The FET is the pass device cancelling the ripple right? How is it doing that where is the opposing phase coming from to cancel the ripple?
Am I way off here?
Is the drop across R3 being used to provide the power for the circuit since there is no common?
Is C1 there to inject the ripple into the gate?
Is R4 simply a grid stopper?
R1 and R2 form a divider what is its purpose?
The Zener I assume is bias the grid at 12V right?
The FET is the pass device cancelling the ripple right? How is it doing that where is the opposing phase coming from to cancel the ripple?
Am I way off here?
Attachments
an ideal inductor would have V = L × dI/dt where V is the voltage across the inductor, L is the inductance, and dI/dt is the rate of change of the current I in the inductor.
So taking the integral (antiderivative) of both sides, you get : integral (V) = L × I
So to approximate this, the circuit uses R1 and R2 and C1 to form a low pass filter, which is an approximation of an integrator, to integrate the voltage across the circuit and apply it to the gate (which you called the grid) of the mosfet. The mosfet and R3 approximately form a voltage controlled current source whose transconductance is set by R3 and the transconductance of the mosfet. (If it is a large power fet, its transconductance will be large enough to ignore), and whose control voltage is the voltage across C1.
So it gives a current of approximately the integral of the voltage across the circuit scaled by an approximate inductance equal to R3 multiplied by the time constant of the RC filter consisting of C1 and the parallel combination of R1 and R2.
The zener is there to limit the voltage applied to the gate to a safe value ( it should have a voltage lower than the gate oxide breakdown voltage of the fet) and the R in the RC filter is split into R1and R2 so that the voltage can go down as well as up in case the thing driving the circuit has asymmetrical output impedance (like a rectifier), in other words so that the C can be discharged as well as charged, in case the driving circuit can only provide one-way current, as would be the case in a power supply. Depending on the application, one might include a reverse discharge diode across the whole circuit to provide a well-defined discharge path for downstream capacitors when the device is turned off. (Though the body diode of the fet probably takes care of that.)
So taking the integral (antiderivative) of both sides, you get : integral (V) = L × I
So to approximate this, the circuit uses R1 and R2 and C1 to form a low pass filter, which is an approximation of an integrator, to integrate the voltage across the circuit and apply it to the gate (which you called the grid) of the mosfet. The mosfet and R3 approximately form a voltage controlled current source whose transconductance is set by R3 and the transconductance of the mosfet. (If it is a large power fet, its transconductance will be large enough to ignore), and whose control voltage is the voltage across C1.
So it gives a current of approximately the integral of the voltage across the circuit scaled by an approximate inductance equal to R3 multiplied by the time constant of the RC filter consisting of C1 and the parallel combination of R1 and R2.
The zener is there to limit the voltage applied to the gate to a safe value ( it should have a voltage lower than the gate oxide breakdown voltage of the fet) and the R in the RC filter is split into R1and R2 so that the voltage can go down as well as up in case the thing driving the circuit has asymmetrical output impedance (like a rectifier), in other words so that the C can be discharged as well as charged, in case the driving circuit can only provide one-way current, as would be the case in a power supply. Depending on the application, one might include a reverse discharge diode across the whole circuit to provide a well-defined discharge path for downstream capacitors when the device is turned off. (Though the body diode of the fet probably takes care of that.)
Last edited:
to answer your initial questions:
Is the drop across R3 being used to provide the power for the circuit since there is no common?
---> no, the mosfet is also dropping voltage
Is C1 there to inject the ripple into the gate?
---> no, it is part of a filter to keep ripple from the gate of the mosfet
Is R4 simply a grid stopper?
---> well, yes, its a gate stopper
R1 and R2 form a divider what is its purpose?
---> R1, R2, R3 together bias the mosfet's grid; they are also part of the ripple filter to the gate
The Zener I assume is bias the grid at 12V right?
---> no, in normal operation it does nothing; during power-up and -down it protects the gate from over-voltage
The FET is the pass device cancelling the ripple right? How is it doing that where is the opposing phase coming from to cancel the ripple?
---> the fet is the pass device but it cancels ripple because of the RC filter at its gate it has a much higher impedance for AC currents than for DC
---> the mosfet acts like a source follower; its source follows the gate; if we keep AC from the gate it cannot show up on the source
---> well, a little bit shows up because the mosfet's gain (gm) is not infinite
---> similar to a real inductor in a CLC filter (but contrary to an inductor it cannot store energy ...)
Better to understand when you complete the circuit with caps and load;
In this simulated example the input is 400V with 10V p-p ripple, the output shows 18mV ripple at 100mA of load current.
Attachments
Last edited: