• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Cathode bias caps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Wow. There’s so many theories going on here that i might be soldering for a month of Sundays. I think 20 to 20 has the right idea, with a bit of luck the readings may be acceptable. The amp seems to be of a forgiving design. An example would be the primary filter cap read 70 instead of 50 uf, one cathode bias resistor read 270 instead of 180 ohms. And all the while it was performing well. Thanks for the replies.
 
I am not agree about this:
"""""1. The LF rolloff does not occur when the capacitor reactance is equal to the cathode resistor, but when the capacitor reactance is equal to the parallel combination of cathode resistor and cathode impedance (approximately 1/gm). Hence the rolloff will be higher than calculated above. You probably need twice the value calculated above."""


The formula 1/gm is relate only for CF.
In attach a diagram from Radiotron where are plotted the curves for same Rk and different C as bypass.
Lookin on that diagram is easy to find the relation between C and Fs.
With 1/10 of C the frequency is put at 10x time.

regarding the Ck on push pull it isn't necessary if there is Class A well tuned.
in Ab1 it is necessary; on Radiotron it is also specified.
In every case I suggest the use the separate Rk in the p-p Amp ( fixed bias is much better)

Walter
 

Attachments

  • DIY.pdf
    208.3 KB · Views: 98
Wow. There’s so many theories going on here that i might be soldering for a month of Sundays. I think 20 to 20 has the right idea, with a bit of luck the readings may be acceptable. The amp seems to be of a forgiving design. An example would be the primary filter cap read 70 instead of 50 uf, one cathode bias resistor read 270 instead of 180 ohms. And all the while it was performing well. Thanks for the replies.

So many options. Going back to the 1st post, what happened to the bad tube that brought it to your attention? Did it smoke that resistor?
 
joke aside and while we're talking about cathodic bias,
what is wrong with using a balancing potentiometer between two tubes with a low value resistance between each output of the potentiometer?
I do not see this technique used nowadays,so I ask myself the question...

It's hard to find a rheostat with the resistance and power rating nowadays... I still have 2 50R 12.5W rheostats from the 60's though.
EDIT: I stand corrected. They still make them but they're 50 bucks!: https://www.newark.com/ohmite/res50r/rheostat-wirewound-50-ohm-12-5w/dp/01F3491
 
The stereogram was losing volume on the right channel, so I replaced the out of spec resistors. After putting it back together again,I squeezed past the cabinet pushing it about a foot across the floor while it was playing music. It gave at a terrible loud noise like something had earth out. That was it. Can only blame myself. It was probably already weak but I helped it along.
 

PRR

Member
Joined 2003
Paid Member
I am not agree about this:
"""""1. The LF rolloff ... when the capacitor reactance is equal to the parallel combination of cathode resistor and cathode impedance (approximately 1/gm)..."""

The formula 1/gm is relate only for CF.
In attach a diagram from Radiotron ....

Your diagram shows that DF96 is correct.

Take the 1uFd curve (just because it is mid-graph and easy to see).

The test conditions specify that Rk is 2k.

2K with 1uFd is 80Hz.

But the *graph* shows -3dB at 170Hz.

If we believe DF96, then we are also looking at the effects of the tube internal cathode impedance. 1/Gm.

The pencil-calc approach is to look at many actual circuits and observe that the "shelf" from bypassed to unbypassed is often "about 6dB", which means 1/Gm is about equal to Rk. Sometimes 5dB, or in this case 8dB (probably because a pentode has added cathode current). For anywhere near this shelf-drop, we can assume that the "R" to be bypassed is "about half" of the Rk seen in the circuit. We do not need an exact guess because it will usually be an Electrolytic, historically huge tolerance, so we need to round-up generously.
 

Attachments

  • Waltube-1.gif
    Waltube-1.gif
    28.3 KB · Views: 189
It's hard to find a rheostat with the resistance and power rating nowadays... I still have 2 50R 12.5W rheostats from the 60's though.
EDIT: I stand corrected. They still make them but they're 50 bucks!: https://www.newark.com/ohmite/res50r/rheostat-wirewound-50-ohm-12-5w/dp/01F3491

I do not use rheostat as balancing potentiometer between two tubes, but simple potentiometer lineair of one or two watt with a resistance of 22ohm in case of problem on the potentiometer.
 
Alan4411 said:
Out of interest, in the real world situation, with say 25 or 50uF caps, what will the OP notice? After all there are quite a few amplifiers with those values fitted originally especially console types.
What he will notice depends on lots of things, including OPT quality and speaker quality and listener quality. However, it is generally best to design audio so that things which you might notice are reduced to a level where you probably won't notice them.

20to20 said:
The bypass caps are only necessary with shared a Rk.
You need bypasses for individual Rk (to prevent loss of gain and high output impedance) and for Rk shared between all four valves (to prevent interchannel crosstalk). You may get away with no bypass for an Rk shared between the two valves of a single PP output stage, but even then you might get better sound with a bypass - it all depends on details.
 
waltube said:
I am not agree about this:
"""""1. The LF rolloff does not occur when the capacitor reactance is equal to the cathode resistor, but when the capacitor reactance is equal to the parallel combination of cathode resistor and cathode impedance (approximately 1/gm). Hence the rolloff will be higher than calculated above. You probably need twice the value calculated above."""
You are free to disagree. You would be wrong to disagree. What I said is true. See any good electronics textbook.
 
While reading through the replies, I have to admit that some of the technical stuff has gone over my head. But playing the stereo today had me thinking just how well it sounds. I think they were made as the last hurrah to the valve era. Six speakers, two twelve inch cones, a pair of electrostatic tweeters. A quiet amp with the opt mounted about two feet away and shielding of earthed sheet metal screwed to the inside of the cabinet. At a glance none of the valves are brighter than any other, and all the resistors and caps are now in spec. As I said earlier that new valve was my only worry, and no wonder. One shop wanted to sell me one for $100. I declined the offer. At the moment i’m reading a PDF of Understanding Audio by W.G.Morley and a lot of what has been written here has been reinforced there. So once again thanks for the info.
 
Thanks for the offers, but I ended up asking a local ham radio club .And besides, unless I’m going to buy something from Zimbabwe, our dollar is a very poor relation. But I will keep it in mind. Now all I have to remedy is hearing my better half yelling out of the window to turn down the sound.That is a problem I do not know how to fix.
 
I use trimmers or 2W wirewound for bias voltage, but a rheostat can be used to trim cathode bias to match power tubes, too.
yes,i known but I have trouble making myself understood in English.
I wanted to talk about one or two dc balance potentiometers like the harman kardon A500 (100ohm potentiometer + 22ohm resistor)
attachment.php
 

Attachments

  • Schema A500 recadré.jpg
    Schema A500 recadré.jpg
    68.4 KB · Views: 222
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.