Extra filtering node

[Leevi]

When using a pre-filtering node either with a choke or resistor is there
any simple guideline for safe sizing the capacitors. The first cap must be according to the specification of the rectifier tube
but how to determine the capacitor of the second node and the choke (inductance) or resistor?
Probably the cap value also correlates with choke or resistor?

/Leevi

[shooter]

the 2nd node is based on the VDC present + the RC time of what you're trying to minimize Fo = 1/2piRC

[sluckey]

The first cap must respect the tube's characteristics. The second cap is typically the same value as the first cap. Higher inductance (or resistance) will result in a smoother B+ at the second node. The choke must be able to handle the current flowing through it. So, if the plate current will flow through the choke, you would choose a much higher current capacity than you would if only screen current were flowing through the choke. I typically see choke values of 3H to 20H in tube amps. Designer gets to choose. Resistor power rating should be double the actual power dissipated by the resistor.

[Leevi]

Let's take an example when using 5Y3 rectifier tube. The first cap value of the pre-filtering node is 22u. I use a 50R resistor between the first and and second node. I want to increase filtering in the second node in order to reduce hum. Is the 50R enough so that I can use a bigger cap like 47u in the second node without stressing the rectifier tube? The second node is B+ node.


/Leevi

[sluckey]

A much better way to reduce ripple at the second node would be to replace that 50Ω with a choke. Since the current rating for a 5Y3 is 125mA, I would choose a 125mA choke. Plenty of suitable choices.

[shooter]

^^^ what he said
Inductors (chokes) oppose changing AC, so whatever's riding on your DC, gets "blocked"
Resistor in conjunction with caps attenuate AC at a given frequency

[Leevi]

Since the current rating for a 5Y3 is 125mA, I would choose a 125mA choke. Plenty of suitable choices.

I agree that the choke reduces ripple better.
So the choke + 47u is safe and doesn't stress the rectifier tube?


If 50 Ohm resistor is used instead is that too hard for the rectifier tube?


/Leevi

[sluckey]

So the choke + 47u is safe and doesn't stress the rectifier tube?

I believe so. Remember, a choke opposes changes in current. This will provide a degree of isolation between the tube and the second cap.

If 50 Ohm resistor is used instead is that too hard for the rectifier tube?

Yes. That 50Ω just makes the 47µF look like it has a bit higher ESR. The tube basically thinks there is 22+47=69µF. Using a much higher resistance isolates the 47µF a lot more, but at the expense of lower B+, which may not be desirable. This makes the choke look much more desirable, especially if output tube plates are connected to the second node.

Do you have a specific project in mind? If so please share so the smart guys can jump in and give specific answers.

[PRR]

> Is the 50R enough so that I can use a bigger cap like 47u in the second node without stressing the rectifier tube?

Worst-case could be 350V across 50r which is 7 Amps peak. A hellofalot for a 0.15A rectifier. The first stage will reduce it, and there's poorly-specified headroom on the rectifier DC rating, but you gotta wonder.

Duncan PSDU is handy for these questions.

You are on the edge. I mistakenly entered 220uFd for C2, and got a warning. Fixed to 22uFd, no warning.

Why can't you use 100 Ohms? 500 Ohms?

[Leevi]

Do you have a specific project in mind? If so please share so the smart guys can jump in and give specific answers.

I'm building a 5F2-A and noticed  the hum after I got the power amp wired. The extra node reduces the hum but a bigger cap than 22u is needed in order to get totally rid of it. The choke is definitely better fix.

Duncan PSDU is handy for these questions.You are on the edge. I mistakenly entered 220uFd for C2, and got a warning. Fixed to 22uFd, no warning.Why can't you use 100 Ohms? 500 Ohms?

I once downloaded the PSDU but did not get it working properly or did not have enough knowledge how to use it;)
A bigger resistor is an option as well.

/Leevi

[Leevi]

I downloaded the new PSDUII and got the following result:
Should that be interpreted that even 50R + 100u is allowed?
150u gives a warning.

/Leevi

[kagliostro]

if I can suggest

Obviously a different MOSFET can improve max voltage allowable

Franco

[Leevi]

Obviously a different MOSFET can improve max voltage allowable

Interesting!


Here is some discussion about this topic.


https://www.diyaudio.com/community/threads/power-supply-mosfet-vs-choke-filter.261117/


/Leevi

[PRR]

> even 50R + 100u is allowed?  150u gives a warning.

Seems OK, but if an increase (2:1 is not a huge increase) raises a warning, I would split the difference. The cap-input filter is a violent thing, and the old guys didn't have fine-scale instrumentation, so we don't really know what these tubes were made to stand.

[Leevi]

if I can suggest

Obviously a different MOSFET can improve max voltage allowable

Franco

Actually I found a similar circuit on my shelf and tried it. Works somehow but causes buzz. I don't think this particular circuit is the solution for my case.
/Leevi

[kagliostro]

That circuit is to be used as a capacitor input so It must be preceded by an e-cap

Is not possibile to use it like a choke input

As it is on the plan I posted it was realized (different Layout) and used in a problematic PSE (2 x GU50) amp with optimal result, no humm at all







Franco

[Leevi]

Hi Franco,

That circuit is to be used as a capacitor input so It must be preceded by an e-cap

In my case there is an e-cap (22u) on both side of the circuit. It reduces the ground hum but probably takes some disturbance outside which causes buzz when increasing volume.


I'm not 100% sure about the schematic of the circuit. I ordered those long time ago from ebay;)


/Leevi

[kagliostro]

The schematic I posted is the last improved version published by Merlin on his books, it decreases the uF value of the cap to 10uF and changes the value of the resistor, result is that H increases from 47H to 155H of the last version (previous DC resistance was 47R now it is 66R)


Franco

[Leevi]

The schematic I posted is the last improved version published by Merlin on his books, it decreases the uF value of the cap to 10uF and changes the value of the resistor, result is that H increases from 47H to 155H of the last version (previous DC resistance was 47R now it is 66R)

Probably my circuit has nothing to do with Merlin's  product.


/Leevi

[kagliostro]

Ciao Risto



Your circuit seems a version of this unit

(on the schematic the protection zener is missing but is present on PCB)









Franco

[Leevi]

My choke circuit has a zener.
The PCB layout seems to be same.

/Leevi

[kagliostro]

The big resistor didn't permit me to see that there are holes for two resistors instead of only one

seems same PCB, may be different component values ?

Merlin changed only component values and increased greatly the performance on his gyrator

and looking to the schematics it looks similar to your's also if not exactly the same





Franco