Bleeders values and ratings

[mahleriana]

Hello,

Are there calculation rules for bleeder value (ohm) and especially rating (watts)? I intend to use allen bradley 2w resistors as this part now has a strong influence on sound since i have upgraded all other parts. The bleeders are now Mills 12w, but i have found very good results when ALL resistors across the amp are the same type.

Also, is there a better place to put it ?
I have a psu with C1 - L - C2 - R - C3 (B+ output stage) - R - C4 (B+ input stage). Bleeders are across C2 and C4.

thanks in advance
eric

[SY]

If the bleeder has any effect on the sound, you have some serious circuit work to do.

In any event, use T = RC to calculate the interacion of the capacitance, the bleeder value, and the desired discharge time. This will give you the time when about 60% of the voltage has been drained. For safety's sake, you'll want to wait about ten time constants before poking your fingers around. Power dissipated is the usual V²/R. Set the resistor power rating at least three times higher.

[mahleriana]

In fact as i went upgrade my amp i just find that now ANY slight change has a HUGE effect on sound (most important i found being grid leak resistor : it changes the whole character of the amp).
It is certainly true because the amp is single ended, every part being actually "in the signal path".
PSU = 100% Obbligato oil capacitors
Resistors = all allen bradley except bleeders (mills)
signal cap : jensen NOS PiO
all wires are silver (except earth sides)
no grid stoppers (you just hear them too much)
Zero bypass capacitors (so higher primary Z on OPT and lower power)
Fot the moment it is a 5V4 - 6SN7 - EL84 triode design.
I'm sure there is no problem in my amp.

I tried there a few different wirewounds (mills, welwyn, white coffins) and cheap carbon resistors (paralleled to increase power wattage).
What i hear when changing bleeders is the same kind of sound difference if i had made a change somewhere else in the amp (i.e. cathode) with the same resistor type, with perhaps a slighter effect.
But the change only occurs now the amp was strongly upgraded towards transparency.
I will post another thread concerning my opinion about resistor brands later...problem is that you are never sure it is their sound or a sound they let you hear from somewere else...Kiwame was a very strong disappointment.

If i understand well :
ex : 100k bleeder under 350v charged cap is = 350*350/100k = 1.225 > 3w resistor but T=7s, so a longer discharge time. 33k bleeder means 12w but T=2s. My amp is effectively discharging in 10-15 sec now (using 33k)
I will then parallel several Allen bradley to increase power rating and lower value and time.
thanks
eric

[PreCD]

Originally Posted by SY

If the bleeder has any effect on the sound, you have some serious circuit work to do.

In any event, use T = RC to calculate the interacion of the capacitance, the bleeder value, and the desired discharge time. This will give you the time when about 60% of the voltage has been drained. For safety's sake, you'll want to wait about ten time constants before poking your fingers around. Power dissipated is the usual V²/R. Set the resistor power rating at least three times higher.

Revisiting an old link here. I am in the process of replacing the filter capacitors in my GFA-555 amplifier (not a tube amp) and am also going to replace the factory bleeder resistors which reside on each filter cap. Is it necessary to use a resistor such as a Mills or can you get by with a less expensive resistor? Recommendations welcome.

[M Gregg]

Bottom line is:


Once you have worked out the power rating and voltage rating see what choices you have in resistor...It might be limited.

Heat is going to cause noise in any component...so higher wattage of the same preferred type is going to be better..(probably bigger)

Again this is a personal choice what sound do you like....it's no good saying no sound because all components have an effect...It's a case of balancing the types so you get as close to neutral as possible. (Remember neural is a definite target, not a preference).

The electrical properties are more important..ie if the component burns out but sounded great....

If you can remove induction and magnetic materials, then again the circuit board creates a dielectric between tracks...

I guess you can hear a change in sound if you re-solder a connection and listen for about 5 mins…as it burns in.

There is no answer to the question because each person may prefer a different “sound”.

Quote “cheap carbon resistors (paralleled to increase power wattage)”

Here is a thought…does paralleling resistors increase wattage?<<<I know you are going to say WHAT?????

If a put two light bulbs in parallel across a supply does each lamp dissipate less wattage if the PSU remains up to voltage?

Will two half watt resistors across a supply dissipate less in each one if they are across the supply? Or do you need a one watt?

Regards
M. Gregg

[DF96]

I'm with SY on this: if you can hear the bleeder resistors then they must be very noisy and your circuit must have very poor PSRR. You should be able to reduce any perceived effect by putting the bleeder across the reservoir cap rather than the smoothing cap, as this moves it further away from the audio circuit. The downside is that you might need a slightly bigger resistor as the voltage wil be higher here.

[jrenkin]

Originally Posted by mahleriana

What i hear when changing bleeders is the same kind of sound difference if i had made a change somewhere else in the amp (i.e. cathode) with the same resistor type, with perhaps a slighter effect.
But the change only occurs now the amp was strongly upgraded towards transparency.

If changing the bleeder produces the same sonic change as any other resistor change anywhere in the amp, something is awry and I think it is called placebo.
We are quite prone to think something sounds different if we have changed something, even, perhaps even especially, if it doesn't.

[payloadde]

Originally Posted by M Gregg

Will two half watt resistors across a supply dissipate less in each one if they are across the supply?

Certainly, if they are both twice the number of ohms ...

[mahleriana]

As I tried a lot of things for my single ended amp, I can say that even bleeder resistors do influence the sound (but perhaps only in very straightforward low parts SE amps?)
I do use Mills there.
eric

[PreCD]

Originally Posted by DF96

I'm with SY on this: if you can hear the bleeder resistors then they must be very noisy and your circuit must have very poor PSRR. You should be able to reduce any perceived effect by putting the bleeder across the reservoir cap rather than the smoothing cap, as this moves it further away from the audio circuit. The downside is that you might need a slightly bigger resistor as the voltage wil be higher here.

This makes sense. At this time there is a 3.9 kOhm 2 Watt resistor on each cap. Tolerances are +-5%. I cannot see a reason to stray from the factory values. My 1st thought is to place a Mills non-inductive 5 Watt resistor in place of each them in the same value.