Hi,
I have built a simple preamp, using a 6AU6 triode-connected with grid 2 as the anode and grounding the plate and grid 3. This gives a plate resistance of 18k and mu of 36. The cathode resistor is 1.2k and is unbypassed, which I think raises the effective internal plate resistance to almost 100k. The heater uses 6.3v AC, with a virtual (resistive) center tap set at +50v, with 1uF to ground.
I'm getting a bit of hum and I wonder if it's caused by not using a cathode bypass capacitor?
I have built a simple preamp, using a 6AU6 triode-connected with grid 2 as the anode and grounding the plate and grid 3. This gives a plate resistance of 18k and mu of 36. The cathode resistor is 1.2k and is unbypassed, which I think raises the effective internal plate resistance to almost 100k. The heater uses 6.3v AC, with a virtual (resistive) center tap set at +50v, with 1uF to ground.
I'm getting a bit of hum and I wonder if it's caused by not using a cathode bypass capacitor?
To answer your questions, I can't post a schematic because I don't know how. However, the preamp stage is integrated with the power amp (stereo) and the power supply consists of:
340v 700mA transformer secondary
MUR1400E rectifier diode bridge
100uF smoothing capacitor
2x 6D22S damper diodes in parallel (for soft startup)
5H 40ohm choke
470uF reservoir capacitor
(take-off for EL34 triode PP stages with 6SN7 drivers = 410v)
680 ohm resistor (to drop a bit of voltage)
10H 260 ohm choke
235uF capacitor
(take-off for 6SL7 splitter = 397v)
27k resistor
100uF capacitor
(take-off for 6AU6 preamps = 220v)
The heater circuit uses two 330 ohm resistors in series across the 6.3v winding. The artificial centre tap given by these resistors is connected to a voltage divider across the 235 capacitor after the 1H choke, consisting of 1M above and 200k below. There is a 1uF capacitor connected between the heater centre tap and ground (i.e. across the 200k resistor).
This positions the heaters at about +65v. This is to protect the heater-cathode insulation of the 6SL7 LTP splitter, which has a cathode potential of about 120v because of the 6AU6 pentode acting as a CCS in its tail. It's also supposed to help reduce hum.
There are actually two heater windings but they're both connected up in as described above. I use them to feed the heaters of the two channels; however, the right channel heater winding feeds both of the input 6AU6 triode stages because they are positioned very close together on the right-hand side of the chassis.
The preamp is stereo and I find that if I remove the 6AU6s I get no hum at all. The 6SL7-6SN7-EL34 pp power amps are well balanced and very quiet. There's no doubt the hum is coming from the 6AU6 stages, which are preceded by passive Baxandall tone stack plus volume control. One 6AU6 is somewhat noisier than the other but they both generate hum.
I hope that helps to clarify things. Sorry it had to be so long-winded!
340v 700mA transformer secondary
MUR1400E rectifier diode bridge
100uF smoothing capacitor
2x 6D22S damper diodes in parallel (for soft startup)
5H 40ohm choke
470uF reservoir capacitor
(take-off for EL34 triode PP stages with 6SN7 drivers = 410v)
680 ohm resistor (to drop a bit of voltage)
10H 260 ohm choke
235uF capacitor
(take-off for 6SL7 splitter = 397v)
27k resistor
100uF capacitor
(take-off for 6AU6 preamps = 220v)
The heater circuit uses two 330 ohm resistors in series across the 6.3v winding. The artificial centre tap given by these resistors is connected to a voltage divider across the 235 capacitor after the 1H choke, consisting of 1M above and 200k below. There is a 1uF capacitor connected between the heater centre tap and ground (i.e. across the 200k resistor).
This positions the heaters at about +65v. This is to protect the heater-cathode insulation of the 6SL7 LTP splitter, which has a cathode potential of about 120v because of the 6AU6 pentode acting as a CCS in its tail. It's also supposed to help reduce hum.
There are actually two heater windings but they're both connected up in as described above. I use them to feed the heaters of the two channels; however, the right channel heater winding feeds both of the input 6AU6 triode stages because they are positioned very close together on the right-hand side of the chassis.
The preamp is stereo and I find that if I remove the 6AU6s I get no hum at all. The 6SL7-6SN7-EL34 pp power amps are well balanced and very quiet. There's no doubt the hum is coming from the 6AU6 stages, which are preceded by passive Baxandall tone stack plus volume control. One 6AU6 is somewhat noisier than the other but they both generate hum.
I hope that helps to clarify things. Sorry it had to be so long-winded!
A few things to check
Take a multimeter and set it to AC volts, place it accross the B+ supply to the plate of the 6AU6 and measure the ripple voltage, if it is greater than 3V, try adding a resistor between the plate resistor of the 6AU6 and add a capacitor on the B+ side of the 6AU6. I would start with 100ohm 5 watt and 100uF cap, adjust accordingly. You should check to assure all grounds go to a single point.
Take a multimeter and set it to AC volts, place it accross the B+ supply to the plate of the 6AU6 and measure the ripple voltage, if it is greater than 3V, try adding a resistor between the plate resistor of the 6AU6 and add a capacitor on the B+ side of the 6AU6. I would start with 100ohm 5 watt and 100uF cap, adjust accordingly. You should check to assure all grounds go to a single point.
A few things to check
Take a multimeter and set it to AC volts, place it accross the B+ supply to the plate of the 6AU6 and measure the ripple voltage, if it is greater than 3V, try adding a resistor between the plate resistor of the 6AU6 and add a capacitor on the B+ side of the 6AU6. I would start with 100ohm 5 watt and 100uF cap, adjust accordingly. You should check to assure all grounds go to a single point.
Take a multimeter and set it to AC volts, place it accross the B+ supply to the plate of the 6AU6 and measure the ripple voltage, if it is greater than 3V, try adding a resistor between the plate resistor of the 6AU6 and add a capacitor on the B+ side of the 6AU6. I would start with 100ohm 5 watt and 100uF cap, adjust accordingly. You should check to assure all grounds go to a single point.
Konnichiwa,
This can be the case, especially if the valve is heated with AC. The cathode is a input node, with a capaictance to the heater. If the cathode is not AC grounded than a highpass is formed between cathode and heater trough which some heater noise can leak into the signal circuit. Also, the valve bceomes more subject to electrostatic pickup with the cathode at high(ish) impedance.
Normall good layout and implementation practice should avoid having hum even in such situations. Are you aware of of what is "best practice"?
Sayonara
ray_moth said:
This can be the case, especially if the valve is heated with AC. The cathode is a input node, with a capaictance to the heater. If the cathode is not AC grounded than a highpass is formed between cathode and heater trough which some heater noise can leak into the signal circuit. Also, the valve bceomes more subject to electrostatic pickup with the cathode at high(ish) impedance.
Normall good layout and implementation practice should avoid having hum even in such situations. Are you aware of of what is "best practice"?
Sayonara
Re: Re: Does a cathode bypass capacitor reduce hum?
That too.
Saying that pulling the 6AU6 stops hum because it's the source is like saying your car's engine stopped because you pulled the computer (fuel injection and all is still there, but it ain't gonna be workin). Or something. You've just stopped the LTP from operating, and that's all.
Tim
Guido Tent said:
That too.
Saying that pulling the 6AU6 stops hum because it's the source is like saying your car's engine stopped because you pulled the computer (fuel injection and all is still there, but it ain't gonna be workin). Or something. You've just stopped the LTP from operating, and that's all.
Tim
Hi,
The only correct answer is NO as Franz G stated correctly already.
It is a myth that bypassing a cathode REDUCES HUM for the simple reason that you're comparing apples to oranges:
What the unbypassed cathode resistor represents is a different circuit compared to the one with the bypassed cathode resistor.
As you don't bypass the cathode resistor, the cathode, as an input node, becomes more sensitive to picking up AC signal; logical as it's not being shunted to ground as it would be with a bypass across the cathode.
Since you already have the AC heater shunted at the virtual CT, the heater induced hum itself should already be greatly reduced.
However, circuit layout, twisting of the heater wires, ripple riding on the B+ etc., can still inject hum into the circuit.
A simple test would be to temporarily run the heater from a well filtered DC source or a battery, for instance.
In some cases such as phonopreamps or high-ish gain linestages the easiest way out is to resort to DC heating.
Some tubes are also more sensitive than others, so hum may vary from one brand to another depending on construction.
Last but not least, heater to cathode insulation should be kept well within the manufacturer's specs as disrespecting it will effectively couple both electrodes with the possible risk of a heater getting stuck to a cathode. (BANG)
Coupling (other than just capacitive) between the both of them can possibly start already with a potential difference of a mere 30V.
Which is why Philips (among others) recommends having heater and cathode at the same potential where no secondary emission can possibly occur.
This can simply be achieved by biasing the heater up to cathode potential. Or at least high enough to be within stated maxima.
Sorry if I repeated some of what other fellow members had stated already...Just thought it would be useful to condense it in a single post.
Hope this helps,
The only correct answer is NO as Franz G stated correctly already.
It is a myth that bypassing a cathode REDUCES HUM for the simple reason that you're comparing apples to oranges:
What the unbypassed cathode resistor represents is a different circuit compared to the one with the bypassed cathode resistor.
As you don't bypass the cathode resistor, the cathode, as an input node, becomes more sensitive to picking up AC signal; logical as it's not being shunted to ground as it would be with a bypass across the cathode.
Since you already have the AC heater shunted at the virtual CT, the heater induced hum itself should already be greatly reduced.
However, circuit layout, twisting of the heater wires, ripple riding on the B+ etc., can still inject hum into the circuit.
A simple test would be to temporarily run the heater from a well filtered DC source or a battery, for instance.
In some cases such as phonopreamps or high-ish gain linestages the easiest way out is to resort to DC heating.
Some tubes are also more sensitive than others, so hum may vary from one brand to another depending on construction.
Last but not least, heater to cathode insulation should be kept well within the manufacturer's specs as disrespecting it will effectively couple both electrodes with the possible risk of a heater getting stuck to a cathode. (BANG)
Coupling (other than just capacitive) between the both of them can possibly start already with a potential difference of a mere 30V.
Which is why Philips (among others) recommends having heater and cathode at the same potential where no secondary emission can possibly occur.
This can simply be achieved by biasing the heater up to cathode potential. Or at least high enough to be within stated maxima.
Sorry if I repeated some of what other fellow members had stated already...Just thought it would be useful to condense it in a single post.
Hope this helps,
Thanks very much for your replies. Firstly, I agree with those who say that several sources of hum are possible. As to understanding good practice, I think I do understand but that doesn't mean I that did it properly. I will have to track down the hum as best I can.
Secondly, I suddenly realize that one aspect of my circuit that may be unwise is that I have placed the passive Baxandall tone controls, followed by the volume control, before the first (6AU6 triode) stage. This means that the signal is firstly attenuated heavily (about 12dB, I think) before it is amplified. So the noise due to the 6AU6 triode is amplified together with the weakened signal. I did it this way because I didn't want to risk overloading the 6AU6 connected using the screen grid as the anode - apparently, this mode of connection does not work well with large signals. However, the risk of a poor signal-to-noise ratio
Maybe it would be better to place the volume control at the beginning, followed by the 6AU6 in convdentional triode mode (plate, grid 3 and grid 2 strapped together as the anode) to give a lower Rp, followed by the tone controls. This should attenuate the both the noise from the 6AU6 and the signal by an equal amount, giving a better signal-to-noise ratio. In that case, I guess I should bypass the cathode resistor, to achieve low Rp for feeding the tone stack. What do you think?
Secondly, I suddenly realize that one aspect of my circuit that may be unwise is that I have placed the passive Baxandall tone controls, followed by the volume control, before the first (6AU6 triode) stage. This means that the signal is firstly attenuated heavily (about 12dB, I think) before it is amplified. So the noise due to the 6AU6 triode is amplified together with the weakened signal. I did it this way because I didn't want to risk overloading the 6AU6 connected using the screen grid as the anode - apparently, this mode of connection does not work well with large signals. However, the risk of a poor signal-to-noise ratio
Maybe it would be better to place the volume control at the beginning, followed by the 6AU6 in convdentional triode mode (plate, grid 3 and grid 2 strapped together as the anode) to give a lower Rp, followed by the tone controls. This should attenuate the both the noise from the 6AU6 and the signal by an equal amount, giving a better signal-to-noise ratio. In that case, I guess I should bypass the cathode resistor, to achieve low Rp for feeding the tone stack. What do you think?
Hi,
Guido,
I'm well aware of what you're saying but I'm stuck with the fact that bypassing Rk will still give a circuit different than the one that was designed from the outset.
That's what stated in post #1:
To which I think the answer is no, it not CAUSED by the fact that Rk is not bypass.
Since the stage is part from an integrated amp, bypassing Rk will change the circuit parameters enough to be a potential source of other problems. The cathode could be a node for NFB input for instance, etc.
Other than that I feel hum is best tackled differently than by just bypassing Rk which is not always a practical solution as outlined above and which to my mind is a bandaid, not a solution as it intrinsically modifies the circuit parameters enough to turn it into a new one.
That, however is just my opinion.
Cheers,
Guido,
I'm well aware of what you're saying but I'm stuck with the fact that bypassing Rk will still give a circuit different than the one that was designed from the outset.
That's what stated in post #1:
To which I think the answer is no, it not CAUSED by the fact that Rk is not bypass.
Since the stage is part from an integrated amp, bypassing Rk will change the circuit parameters enough to be a potential source of other problems. The cathode could be a node for NFB input for instance, etc.
Other than that I feel hum is best tackled differently than by just bypassing Rk which is not always a practical solution as outlined above and which to my mind is a bandaid, not a solution as it intrinsically modifies the circuit parameters enough to turn it into a new one.
That, however is just my opinion.
Cheers,
fdegrove said:Hi,
Guido,
I'm well aware of what you're saying but I'm stuck with the fact that bypassing Rk will still give a circuit different than the one that was designed from the outset.
That's what stated in post #1:
To which I think the answer is no, it not CAUSED by the fact that Rk is not bypass.
Since the stage is part from an integrated amp, bypassing Rk will change the circuit parameters enough to be a potential source of other problems. The cathode could be a node for NFB input for instance, etc.
Other than that I feel hum is best tackled differently than by just bypassing Rk which is not always a practical solution as outlined above and which to my mind is a bandaid, not a solution as it intrinsically modifies the circuit parameters enough to turn it into a new one.
That, however is just my opinion.
Cheers,
Frank,
Keep it simple, and forget about the rest of the circuit.
After all, one should look at where the hum stems from. If that is the common cathode stage, it will reduce once the cathode is bypassed.
If not, it won't affect the circuit and the hum will stay
I am trying by the way to finish an article on how to tackle hum, and how to decently layout vtube circuits.
best regards
Just discovered a reason for a lot of the hum
I feel a bit stupid having to admit this! I have just discovered that the multi-output extension lead I've been using has only a 2-core cable. The ground pin is not connected to anything. However, my signal source is a video-CD player connected to a TV, which has a line output that is connected to the amplifier input. The VCD player and the TV are both run from the same extension lead as the amp, so they also are not grounded.
The only grounding comes from a Cable TV connection to the TV! This crazy arrangement works OK, with very little hum. However, it's a different story if I disconnect the TV-to-amp lead, which I often do for testing. Then, because there is no grounding at all, the amp picks up considerable hum very easily.
2-core extension leads are very common in Asia (where life is cheap and profit is everything). The only way I can be sure of a 3-core extension is to make one myself, which I will have to do.
I still intend to reorganize the preamp so that the volume control precedes it and the tone stack follows it, leading directly into the differential phase splitter of the main amp. I think this would be a better topology: the volume control will be at the input, where it belongs, the preamp stage will be a conventionally-wired 6AU6 in triode mode (with a low output impedance), and it will be easy to arrange a tone-defeat by simply switching the main amp input directly to the volume control instead of the tone stack output.
Anyway, thanks very nmuch for all your responses.
I feel a bit stupid having to admit this! I have just discovered that the multi-output extension lead I've been using has only a 2-core cable. The ground pin is not connected to anything. However, my signal source is a video-CD player connected to a TV, which has a line output that is connected to the amplifier input. The VCD player and the TV are both run from the same extension lead as the amp, so they also are not grounded.
The only grounding comes from a Cable TV connection to the TV! This crazy arrangement works OK, with very little hum. However, it's a different story if I disconnect the TV-to-amp lead, which I often do for testing. Then, because there is no grounding at all, the amp picks up considerable hum very easily.
2-core extension leads are very common in Asia (where life is cheap and profit is everything). The only way I can be sure of a 3-core extension is to make one myself, which I will have to do.
I still intend to reorganize the preamp so that the volume control precedes it and the tone stack follows it, leading directly into the differential phase splitter of the main amp. I think this would be a better topology: the volume control will be at the input, where it belongs, the preamp stage will be a conventionally-wired 6AU6 in triode mode (with a low output impedance), and it will be easy to arrange a tone-defeat by simply switching the main amp input directly to the volume control instead of the tone stack output.
Anyway, thanks very nmuch for all your responses.
Guido
Imagine a cathode follower with an cathode bypass! Does not work.
Do cathode followers produce hum, because there is no bypass? Absolutely not!
To use a cathode bypass cap is a question of local current feedback, influences the gain, impedance and distortion.
It is definitely not a question of hum or not.
Franz
Imagine a cathode follower with an cathode bypass! Does not work.
Do cathode followers produce hum, because there is no bypass? Absolutely not!
To use a cathode bypass cap is a question of local current feedback, influences the gain, impedance and distortion.
It is definitely not a question of hum or not.
Franz
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.