Your diagram looks a bit odd but it looks like it already has a elevated DC for heater supply look were the black wire is, that certainly have a 51k resistor not a 5.1k you can use a multimeter to check between any heater (assuming they are wired to the same transformer winding) pin to gnd and you should see a DC signal (~1/2 of the supply voltage)
Thank you for the information. I am not 100% sure about the diagram because I found it online while searching for way to stop this noise. I think I should test it out with multimeter to make sure. Thanks for the idea...
I guess all filament wiring will be in parallel, but without a wiring schematic this is not 100 % sure. So if you add the 2 resistors, and ground in the center, this works for all tubes. So just check if all filaments are wired in parallel.
Regards, Gerrit
Regards, Gerrit
If it doesn't I am recommending TRYING this , its controversial in some peoples eyes , caused conflicting arguments that got a bit hot but in YOUR particular circumstance its worth a try.
What am I talking about ?--------- a screening can .
Yes I know all the supposed negatives of-
it shortens valve life etc but this is an EF95 B7A based valve in my experience this valve does not get very hot and every one I came across in government/Army/Navy/Air Force had screening precisely because of its sensitivity to RF pick up.
Make sure there is an air slot at the bottom to funnel the heat away .
This very forum has DIY instructions on making one (or did ) and if it doesn't work you haven't lost much but I have tried running my valve communications receivers without them and---- yes - RF harmonics occurred (pickup of stray signals ) .
Worth a try.
What am I talking about ?--------- a screening can .
Yes I know all the supposed negatives of-
it shortens valve life etc but this is an EF95 B7A based valve in my experience this valve does not get very hot and every one I came across in government/Army/Navy/Air Force had screening precisely because of its sensitivity to RF pick up.
Make sure there is an air slot at the bottom to funnel the heat away .
This very forum has DIY instructions on making one (or did ) and if it doesn't work you haven't lost much but I have tried running my valve communications receivers without them and---- yes - RF harmonics occurred (pickup of stray signals ) .
Worth a try.
Sorry capacitive coupling - coupling of the noise via stay capacitance. For example one path is from the heater wires into the grid of the first valve.
Interestingly your little spikes are 50Hz apart rather than 100Hz but they are very HF. So one possible method is they are being generated by the HT rectifier and getting back through the mains transformer and onto the heater supplies. Interesting everybody suggests using two resistor trick. I don't know why capacitors to ground (say 100nF) on each heater lead would not work. Adding caps around the bridge rectifier may also help.
Interestingly your little spikes are 50Hz apart rather than 100Hz but they are very HF. So one possible method is they are being generated by the HT rectifier and getting back through the mains transformer and onto the heater supplies. Interesting everybody suggests using two resistor trick. I don't know why capacitors to ground (say 100nF) on each heater lead would not work. Adding caps around the bridge rectifier may also help.
Last edited:
Thank you for the information. I was also thinking of doing this but finding these little cans will be a challenge. What if I cover the whole amplifier with a metal case? like a big huge bucket? just to see if it get rid of the noise? I tried covering all the tubes with a large white steel can but that didn't help.
@gerrittube
I went through the wires when I Removed the amplifier. I Will do so again and post a photo as well. It has blue wires and red wires connecting to the circuit board. None of them are braided. Blue wires come towards the 6J1 Gain tubes. yes sir! I must check to see if all the tubes are parallel connected.
It is said that this is dual mono circuits...! something called SEPP design. I have a feeling that there are two filament wirings going for each channel. So I guess I gotta add one for each.
I went through the wires when I Removed the amplifier. I Will do so again and post a photo as well. It has blue wires and red wires connecting to the circuit board. None of them are braided. Blue wires come towards the 6J1 Gain tubes. yes sir! I must check to see if all the tubes are parallel connected.
It is said that this is dual mono circuits...! something called SEPP design. I have a feeling that there are two filament wirings going for each channel. So I guess I gotta add one for each.
Sounds like a very interesting idea but honestly I don't understand how i should do it...! (me = not electronically inclined although I'm alright with soldering and understanding basic stuff)
How do i ground the capacitors? solder a wire to + or - side of the caps and connect it to ground...? ( this is just a thought, i have no idea)
Covering the whole unit with steel has been proved not to work Nofacemonster, as it does not stop internally generated RF hash from occurring or transformer/power supply pickup.
All my old equipment made in steel proves that as does an old high quality signal generator where the whole oscillator tube had to be individually screened to stop pickup.
Its a scientific fact , just try it you haven't much to lose just a bit of time and effort .
Langdrex (for one ) sell them -
B7G VINTAGE 2INCH SCREENING CAN AND SKIRTED BASE NOS VALVE/TUBE - Langrex
Remember to make an air slot at the bottom to allow air intake but it only takes minutes to see if it works or not .
All my old equipment made in steel proves that as does an old high quality signal generator where the whole oscillator tube had to be individually screened to stop pickup.
Its a scientific fact , just try it you haven't much to lose just a bit of time and effort .
Langdrex (for one ) sell them -
B7G VINTAGE 2INCH SCREENING CAN AND SKIRTED BASE NOS VALVE/TUBE - Langrex
Remember to make an air slot at the bottom to allow air intake but it only takes minutes to see if it works or not .
OK easy to try. Connect a 100nF cap between pin 3 (h1) of EF95 to ground on your PCB say grounded end of R16. Then connect a second 100nF cap between pin 4 (h1) of EF95 to ground on your PCB. Thats it. The cap can be 220nF does not matter. The idea is that any high frequency rectifier voltage will get shunted to ground instead of your grid. Ceramic capacitors are best for this.
Thank you for the alternative technique. I already ordered few resistors of 100ohms 1W and I might get them tomorrow to try out. If that doesn't work, I am going to try out your method although I have to order 100nf capacitors. By the way, when I search for 100nf capacitors, all they have is 100pf capacitors that looks like little flat buttons. Are they correct or there are 100nf caps as well?
The modern ones look like beads - try 100nF multi-layer leaded ceramic. Don't get 100pF they are 1000x smaller in value.
Here take a look to Merlin Blencowe he run a website and sells books about designing of vaccum tubes audio amplifiers and preamplifiers, tips and info that can be applied to all audio circuits, the link is a chapter of his book that can be seen in his site regarding heater supplies, please if this is dificult to understand or execute, please take your time to learn the basics before doing something, this is the easiest digerible info I could find about, stay safe!
The Valve Wizard
(Problem Solved)
First of all, I want to thank all of you who helped me to solve this problem with my little dot mkII, without your suggestions and advice I could have never done this.
I was right about it. Today I go the 100ohm resistors and what I did was solder a resistor to each of the filament pins of gain tubes (in my case Mullard C8100 and grounded them). It works 98% successful and that's all I need.
Now I can run any headphone on Gain setting 4 (highest) without any buzzing or hum. Here are the test results.
HD6XX there is no noise at all (100% Succesful in all gain settings)
V.E Monk Plus (98% Successful. Absolutely no noise unless I put gain setting to 4 and crank the volume to the highest end) This is unnecessary since it is unbearably loud at 5% volume in high gain settings.
V.E Bonus IEM (Same story as Monk Plus)
KZ ZS5 - This is also 100% successful, zero noise, zero buzzes, and hiss. Completely black background in all volume levels. (Unbelievable since it is a hybrid of Dynamic and balance amateurs )
The above stuff is based on my hearing. Not from graphs or any other things because I don't need that accuracy plus if it doesn't bother me, that's all that matters.
Conclusion
"Artificial Center Tap" with 100ohm resistors work flawlessly.
First of all, I want to thank all of you who helped me to solve this problem with my little dot mkII, without your suggestions and advice I could have never done this.
I was right about it. Today I go the 100ohm resistors and what I did was solder a resistor to each of the filament pins of gain tubes (in my case Mullard C8100 and grounded them). It works 98% successful and that's all I need.
Now I can run any headphone on Gain setting 4 (highest) without any buzzing or hum. Here are the test results.
HD6XX there is no noise at all (100% Succesful in all gain settings)
V.E Monk Plus (98% Successful. Absolutely no noise unless I put gain setting to 4 and crank the volume to the highest end) This is unnecessary since it is unbearably loud at 5% volume in high gain settings.
V.E Bonus IEM (Same story as Monk Plus)
KZ ZS5 - This is also 100% successful, zero noise, zero buzzes, and hiss. Completely black background in all volume levels. (Unbelievable since it is a hybrid of Dynamic and balance amateurs )
The above stuff is based on my hearing. Not from graphs or any other things because I don't need that accuracy plus if it doesn't bother me, that's all that matters.
Conclusion
"Artificial Center Tap" with 100ohm resistors work flawlessly.