Simple high voltage regulator suggestions

[zintolo]

I use this circuit for years, it works fine with a good range of voltages and current.

Thank you waltube!

 

[zintolo]

I think it's maybe handy to explain a bit more context?

Sure, I have planned to build three SE amp:

• worst one is an ebay-bought pcb that I got from a friend, with output transformers included;
• mid one is very simple a guitar amp;
• best one is an UNSET that I've briefly tested with a HV PSU and its toroidal transformers.

Because all we know at this point is that you want some kind of regulator for a SE amplifier?

Yes, correct.

Are you specifically looking for ultra low noise and PSRR?
Or does it just has to suppress the ripple adequately?

The latter!

When just resistors are being used it also won't have a fixed voltage. Meaning it will follow its input voltage.

It is clear, the mosfet follows the voltage at its base, being referenced to a voltage divider, it will be a percentage of the infeed voltage.

 

[jan.didden]

Thank you Jan, this is is what I've got from the simulation of the circuit on the first post:
View attachment 1316416

Change the vertical scale to linear.

Jan

 

[zintolo]

Here it is!

0,01 Ohm @100 Hz
0,2 Ohm @2 kHz
2 Ohm @20 kHz

 

[zintolo]

Is there any advantage in having the phase stable at 90° in the full audio bandwidth?
With R4 at 11,4 kOhm, phase at 20 Hz is 89,993°, at 20 kHz is 90°. Impedance is unchanged.

 

[zintolo]

I will start with a capacitance multiplier as suggested, and report the results.

 

[b_force]

Those simulations don't look quite right.

Usually you put a AC source in series with the DC source.

 

[zintolo]

You can get same results using this 10V @100Hz sine with 450V DC Offset:

 

[b_force]

I know, but that's not what I mean 🙂

 

[zintolo]

😀 ...and would you mind to explain me please?

 

[b_force]

😀 ...and would you mind to explain me please?

Well apparently it also works this way lol 😀

That either used to be a bug in LTSpice or in another one of those SPICE programs (multisim maybe)
Can't remember, have been using this stuff for so long 😀 😀

Btw, you could go for a RCRC filter on the base or gate.
That way we can get a little more ripple reduction like this;

(this is a slightly different circuit, but the same thing applies)

 

[jackinnj]

With closed-loop regulation, using an error amplifier, keeping to a plain resistive output is difficult (the more so at high voltage); and phase has to be manipulated to preserve the phase margin, or prevent it crashing through crossover at >20dB/decade. This is the reason that industrial chips like the LM317, LT108x (etc) present an inductive looking output phase that collides with the output capacitors to produce a lump in the impedance at some audio frequency, if we don't take care.

If there is a "lump" in the output impedance, it's a sign of potential instability. For LM317 easily remedied by adjusting the ESR of the output capacitor. For LDO regulators it's a bit more complex.

 

[jan.didden]

Here it is!
View attachment 1316483

View attachment 1316484

0,01 Ohm @100 Hz
0,2 Ohm @2 kHz
2 Ohm @20 kHz

How do you get at the ohm values?

Jan

 

[zintolo]

How do you get at the ohm values?

By using the formula you gave me on the first reply to the topic:

Make a current source loading the output, say 50mA rms, and do a .ac with it from 20Hz to 20kHz, and look at the signal that this produces at the output.
For instance, if you see 10mV signal at the output when loaded by 50mA rms, that is an output impedance of 10mV/50mA = 0.2ohms.

 

[jan.didden]

👍

 

[b_force]

Here it is!
View attachment 1316483

View attachment 1316484

0,01 Ohm @100 Hz
0,2 Ohm @2 kHz
2 Ohm @20 kHz

You are aware that you can change the graphs by right clicking on them?

In that case you can put the formula right in there and actually plot the impedance.

 

[zintolo]

you can put the formula right in there and actually plot the impedance

I didn't think about that. Thanks!

Is there any advantage in having the phase stable at 90° in the full audio bandwidth?

I try to reply to myself: stability of the voltage regulator during/after transitories?

 

[zintolo]

The ripple can be very low; fractions of a mV are not difficult, even at 100mA loads. And since there is no longer any need to consider a vacuum rectifier, large capacitors can be accommodated in the Raw DC, reducing it further.

I've found that for tube amps one source follower is more than "proper", i.e. adequate. If you feed the gate by a filtered voltage, you may call it "a capacitance multiplier". If you feed it by a regulated reference voltage, you may call it "a voltage regulator".

Following your suggestions, this could be a simple regulator (without any feedback) for B+ around 400V and g2 around 200V.
Soft-start (with its high RC-constant, so low low-pass frequency) should improve ripple.
I'm still missing the latch in case the mosfet shorts.

Am I thinking right that a fixed reference for B+ is preferrable for fixed bias SE amps, while a simple voltage divider without zener can be "compensated" by cathode bias, where the resistor at the cathode of the power tube compensates the variations in voltage?

 

[jan.didden]

while a simple voltage divider without zener can be "compensated" by cathode bias, where the resistor at the cathode of the power tube compensates the variations in voltage?

It's called 'negative feedback' ;-)

Jan

 

[Ite]

to measure the output impedance over frequency in LTspice:

1) Put a current source (set to "AC 1") from the output node to gnd.
(this must be the only "AC 1" source in the schematic)
2) run an AC sweep
3) click on the output node, this will plot output-voltage.
4) In the plot-window left-click on the left axis, and set it to "Linear"
5) In the plot-window left-click on V(output),and change to V(output)/1A
done!