Try a 2nd-order filter, by cascading another RC after R25/C8. It can give a decent improvement for the cost of only two components.
Yes, It's an emitter follower so the ripple at the base is replicated at the output. 820 ohms and 22u? That doesn't do a lot at mains frequency. 100mH is not to much either. As noted, 75ohms, + the 820 is about 900, with 22uF.
The Fc is about 7Hz, so at 50 or 60Hz you can get about 20dB, tops.
Can we see the calculations, we may overlook something.
Jan
The Fc is about 7Hz, so at 50 or 60Hz you can get about 20dB, tops.
Can we see the calculations, we may overlook something.
Jan
BTW the PS is for a preamp, shouldnt be more than 30mA current draw.
certainly will.
There are several pc prgrams to play around power supplies with no much knowledge. Why not to play with one of them? (IMO they are of little importance for me).
Without knowing the cap value that's just pie in the sky!
Still waiting for the OP's calculation.
Jan
That’s a good point, other than cascading an additional stage, a single capacitor would not be very effective, it would have to be rather large I would think, at least 1000uf and I’m using 500V types. With the current setup, running it through an LC filter attenuation calculator I came up with -61dB, that’s if We assume that the 100mH choke is effectively seeing 22,000uf of capacitance.
Last edited:
But there is no 22000uF I can see.
You probably think about the cap 'seen' from the output, which determines the output voltage droop due to load currents.
But here we talk about attenuating the ripple which comes in from the rectifier/cap etc.
There is no multiplication. The best situation would be if the pass transistor or FET has infinite input (base/gate) impedance.
Then the ripple attenuation at that base/gate is just the R, L and C in the circuit.
Since this is an emitter/source follower, whatever is at the base/gate will be at the emitter/source.
Jan
You probably think about the cap 'seen' from the output, which determines the output voltage droop due to load currents.
But here we talk about attenuating the ripple which comes in from the rectifier/cap etc.
There is no multiplication. The best situation would be if the pass transistor or FET has infinite input (base/gate) impedance.
Then the ripple attenuation at that base/gate is just the R, L and C in the circuit.
Since this is an emitter/source follower, whatever is at the base/gate will be at the emitter/source.
Jan
My inspiration for pre-reg filtering comes from the power supply of a Counterpoint SA-2 Pre-pre, I’ve thought of using the Zener Follower/Multiplier combination. This particular preamp specs at -90dB S/N, Im not sure how this is possible using an AC heater supply for the pass/error tubes, great care would have to be taken on the heater wiring.
Attachments
Ahh yes I see, I would need the resistor-choke in series, with the choke/capacitor at the base junction? I’m going to have to breadboard this, I’m not sure how effective the choke is in this case. The standard capacitance amplifier setup is just as effective without the choke possibly.
You will have to excuse some of my posts, which are nothing more than "thinking outloud". Sometimes that can be dangerous!
With the best of my intentions, I suggest to read books, magazines and the like prior to start your designs. Power supply voltage stabilization/regulation is saturated of good data, examples, theory and practice. One of the earlier articles about it is dated in 1939 from Hunt and Hickman. In those 80 years, much water has passed under the bridge and (again) full of good things to read and learn.
Kind regards.
Kind regards.
Really? You want to seriously study that and spend many hours doing so?
You picked probably the least accesible text.
Not to brag, but start with the superreg article in the diyaudiostore.
That is written for people who are relatively new to regulators, to get up to speed easily and quickly, math-free.
Jan
You picked probably the least accesible text.
Not to brag, but start with the superreg article in the diyaudiostore.
That is written for people who are relatively new to regulators, to get up to speed easily and quickly, math-free.
Jan