LM317 for opamps and HT tube regulation

[rh001]

I suggest simplification...this is what I've done in the past: Assuming the load will be more or less the same and not a lab variable power supply or one whose load changes significantly, drop the transformer secondary+rectifier+first capacitor voltage with power resistors until voltage is just 10 or so volts higher than the needed voltage. Then LM317 will drop another 3 volts giving a 7 volt margin to allow for summertime utility company voltage sag. Each of the power resistors can be and should be followed with another filter cap so ripple voltage at input of the LM317 is low, no more than 2v p-p, but should be easy to get it down to 1v p-p. These voltage dropping resistors might be in the range or 100 to 150 ohms (assuming 2 such resistors) and about 1 watt in which case I use a 3 watt resistor, for instance standard size metal oxide, not hi-performance small 3 watt cuz I want surface area for power dissipation. The resistor to GND on the LM317 adj terminal will be running around 1.8 watts so use a 3 watt there. In pic, the 620 ohm resistor is the hv dropping resistor and the 2 100kohm resistors in parallel are the adj terminal resistor....I needed 50kohm 3 watt but only had 100kohm on hand so that's why two of them. Becasuse there's not much volatege across the LM317 only a modest heat sink was needed. I used an LM317HV though that probably wasn't necessary but the additional cost was negligible so I went for it. Also, note I put two 6.3vac secondary windings in series, then regulated down to something like 8 volts, then dropped with chassis mount power resistors down to 6.3 vdc for filaments. The small transformer is for +48vdc phantom power. (Using an Antek transformer, forget the model num)

 

[Cqwet Dbdfte]

I'll get back with a .sch and BOM in a week or so.

 

[reinw33]

I suggest simplification...this is what I've done in the past: Assuming the load will be more or less the same and not a lab variable power supply or one whose load changes significantly, drop the transformer secondary+rectifier+first capacitor voltage with power resistors until voltage is just 10 or so volts higher than the needed voltage. Then LM317 will drop another 3 volts giving a 7 volt margin to allow for summertime utility company voltage sag. Each of the power resistors can be and should be followed with another filter cap so ripple voltage at input of the LM317 is low, no more than 2v p-p, but should be easy to get it down to 1v p-p. These voltage dropping resistors might be in the range or 100 to 150 ohms (assuming 2 such resistors) and about 1 watt in which case I use a 3 watt resistor, for instance standard size metal oxide, not hi-performance small 3 watt cuz I want surface area for power dissipation. The resistor to GND on the LM317 adj terminal will be running around 1.8 watts so use a 3 watt there. In pic, the 620 ohm resistor is the hv dropping resistor and the 2 100kohm resistors in parallel are the adj terminal resistor....I needed 50kohm 3 watt but only had 100kohm on hand so that's why two of them. Becasuse there's not much volatege across the LM317 only a modest heat sink was needed. I used an LM317HV though that probably wasn't necessary but the additional cost was negligible so I went for it. Also, note I put two 6.3vac secondary windings in series, then regulated down to something like 8 volts, then dropped with chassis mount power resistors down to 6.3 vdc for filaments. The small transformer is for +48vdc phantom power. (Using an Antek transformer, forget the model num)

This is nice.
Not just a knowledge bomb, also a skills bomb.
Stereo channels in a 2RU case with tubes.
All DIY.
This is probably way out of my and many amateurs league.
You have some very good skills and reasoning in electronics sir.
I'm gonna simplify Sir, @rh00.
Keep it simple. Got it.

 

[Matador]

But why does it say R16 "DNP". For what case should it be populated?

R16 (and R17) are used in case you don't need the trimmer adjustability (RV1 and RV2). You can populate a fixed resistor and just rely on the resistor tolerance to set the output voltage.

Note: this is especially true as the tolerance on the internal voltage reference of the LM317 is far worse than that of the resistors. So even if you use expensive 0.1% resistors the output can vary as much as 5% due to the internal VREF. So in short, just a plain resistor is Just Fine (TM).

 

[reinw33]

With a regular pot it's limited to about 350V, but with a fixed voltage there is basically no upper limit as long as the MOSFET can take it. The MOSFET is the part that shoulders the differential voltage and power in a Maida.

Personally I don't find much use for the Maida, when a Zener follower is so much simpler. Zeners are not as accurate of course, and you have to wait for them to settle at running temperature, but valve gear seldom needs that level of voltage accuracy.

Hi @merlin
If simple is what you want. Go for it.
EDIT: Merlin, I was stuck, without a schematic. And yours merlin really did helped me out. Thank you merlin. But I don't go for simple, I go for what sounds best as possible.

I'm not experienced enough in tubes gear to know whether the zener emitter follower performs almost just as good as the maida.
It's dim and dark for me in that regard so to speak.

But I'm almost all ready to go to try out the zener emitter follower of the LM317.

A 200V and a 120V zener in series for reference 320V from a 380V HT at 30mA would exceed the LM317HVT differential voltage by 3,5V or so, to get 300V and "zapidyzap"
So if I drop 50-60V ht with filtering, 1,6k ohms choking, etc etc, leaning on what @rh001 suggested, and see, if sth. like 330V on the Lm317 could get regulated to 300V, it would be less than 1 Watt about 0.85ishW dissipated in the LM317, with a differential of 30V dissipated.

Is it possible to also get a schematic on the zener emitter follower.
I'm not sure I'm capable of doing it without frying some chips @merlin.

 

[reinw33]

R16 (and R17) are used in case you don't need the trimmer adjustability (RV1 and RV2). You can populate a fixed resistor and just rely on the resistor tolerance to set the output voltage.

Note: this is especially true as the tolerance on the internal voltage reference of the LM317 is far worse than that of the resistors. So even if you use expensive 0.1% resistors the output can vary as much as 5% due to the internal VREF. So in short, just a plain resistor is Just Fine (TM).

@Matador
I'm so glad I asked you Matador.
You've been helping so much since I come through the door.
Sometimes you just need help from the Pros.
Thanks, it almost all makes sense by itself.
Thanks to Matador.
NOTED: OK, I will spare the 0,1% resistor for balanced occasions for like P48 on the CT input Tx.

 

[Matador]

Sometimes you just need help from the Pros.

Oh boy...if that's the case, there's lots of other people you can ask.

 

[Cqwet Dbdfte]

I wouldn't push the lm317 beyond 10Watts without heatsink.

A TO-220 without a heatsink will get quite warm with 1W .
10W - 20W not a good idea w/o heatsink.
With Rθ(JA) of 24°C/W x 20 = 580°C it will not last more than the blink of an eye
If you need to drop out 10 - 20W use a chassis mounted resistor. Better yet, use a more suitable transformer if a power supply.

 

[reinw33]

That's insane!
4.16 W of dissipation might cook a coffee.
Will definitely use proper heatsinking for lm317 now.
The transformer output will have some headroom, perhaps 20V-30V-ish for LCR filtering.
The mosfet irf820 supposedly can handle a max. of 80 W power dissipation.
There's definitely some room for tolerances for the output V of transformer.
Thanks for your reply.
Have a good start into the week.