Capable Electronics Advice required please

[DF96]

The caps in parallel probably isn't that big of a deal with your application. I've only seen a real use for it for instances where it needs to maintain a phase coherency, if this was like a mic pre output, you would add or subtract caps in parallel to remove comb filtering effects when using it along with different mic preamps in the studio.

The only real use of parallel caps I have seen is the bias network for an RF PA transistor. Not necessary for audio.

 

[Ketje]

5uF at 125V stores 156mJ of energy. Looking at it another way, 5uF and 200R have a time constant of 1ms. I don't think that is going to harm a film resistor. Also, bear in mind that the short is being applied at a point which is already nominally at zero voltage so it is only as the voltages collapse that any current will flow; 125V is not dumped across the resistor. No need for CC.

Didn't you overlook the 20k cathode resistance in the discharge loop ?
Mona

 

[vintageaxeman]

well one of the major reasons why the 200 ohm is over rated is so that there is dissipation headroom. Because ideally, you want the resistance of the resistor pair in the channel to be the same. Otherwise, the phase angle of the output will change with temperature of the resistors. If you replace these resistors, they should be replaced with 2-5W 100ppm type. I use Mills/vishay MRA series or vishay z-foil types on these. The caps in parallel probably isn't that big of a deal with your application. I've only seen a real use for it for instances where it needs to maintain a phase coherency, if this was like a mic pre output, you would add or subtract caps in parallel to remove comb filtering effects when using it along with different mic preamps in the studio.

Hi Dave, you mention using Z-foils which IS what I was considering using but they are only 0.4watt. Are you saying they should be suitable?

 

[TheGimp]

It seems to me that the circuit should have a high value bleed resistor to ground.

The 20K is not in the discharge loop, the loop includes the power supply, Tube, capacitor, and resistor.

The 20K is in parallel with the output to ground. Technically one should do a thevenin's equivalent but I presume the impedance of the power supply and cathode follower are less than the 20K so it can be ignored. And it is in series with the current source.

 

[vintageaxeman]

Sometimes I think I should have paid more attention at school!

Thank you, Gimp, for responding to this thread, but my stage of learning in electronics doesn't seem to be helping me grasp the concepts you have laid out before me!.... Is there any way you can give me a layman's explanation?

 

[TheGimp]

Lets break it down into two paths, DC and AC.

The DC current path is from the power supply to the anode of the top output tube, through the tube, out the cathode, down through the 20K resistor, through the current source and to ground. The ground is contiguous with the power supply so it closes the loop.

The DC is blocked by the capacitors in the output C9,11,13 so so current flows.

Now the AC current path also stars at the power supply, down through the tube and out the cathode, but AC sees a very high impedance represented by the current source and 20K resistor.

The coupling capacitors C9,11,13 in parallel represent a low impedance to AC so the current flow is through them, through the 200 Ohm resistor, through the load attached to the output and back to ground.

Since the resistor side of the coupling capacitors are at ground either through the muting relay, or the load resistance, the capacitor will charge up positive on the tube side to the output voltage of the tube cathode. There will be approximately 125V across the capacitor.

If the feedback resistor (R19 at the left of the screen) has a DC path to ground then my comment about the high value bleed V can be ignored as R19 will perform this function.

 

[vintageaxeman]

That's very kind of you, Gimp. I think I understand that now. Many thanks for your help!

 

[DF96]

Didn't you overlook the 20k cathode resistance in the discharge loop ?
Mona

Yes, maybe I did. That makes it even clearer: the output resistors do not need to have high power handling for their own safety. Depending on the load and signal levels, it is likely that they don't need high power handling to avoid thermal distortion effects either.

It seems to me that the circuit should have a high value bleed resistor to ground.

There appears to be a DC path via R19 100k. Stop Press: just noticed that you have already spotted this.

The 20K is in parallel with the output to ground.

The 20k is in series with the cap discharge path. There are, in addition, other things which may appear in parallel with the 20k.

Anyway, it seems that nothing too strenuous is required of the components.

 

[vintageaxeman]

I am learning so much.
Thank you, guys. (or girls, if you are....)
Any more comments would be most welcome, or any general points on valve (tube) amp mods.
Cheers.

 

[TonyTecson]

those big output caps are not really mandatory, you can even get away with smaller value caps....say 1ufd, have done that and missed nothing, except the higher cost of bigger caps...

more so with that preamp driving a tube amp....

imho, topology and operating points will have the biggest impact than just mere one or two passive parts.....