Hey gang, seems like I only post when I realize I suck at actually designing/building, but I have no choice...
I built an OTL very similar to DIY Audio Projects Forum • Another Completed OTL Headphone Amp. It sounds fine, other than a few small issues
1. There is a slight hum even when the amplifier is off, when it is plugged it. If I unplug it completely the hum goes away, but if it's plugged in and off (via a power strip) then there is a slight hum/buzz. If I touch the ground pin on the plug (even tho my amp isn't connected to the ground pin) it stops.
2. Sort of building on that point, if the headphones are like half plugged in, then I get pronounced hum, even when the amp is off. If they are fully plugged it, the hum is faint but noticeable. If I touch the non-grounded terminals of the jack (tip or ring) I get a loud hum aswell.
3. When I turn the amp on, it hums very loudly for a few minutes, then that hum goes away (the hum from #1,2 remains)
I'm using an Antek Toroid. I currently have the tubes removed. Disconnecting the PSU ground solves the problem, but I'm measuring 0V AC on the transformer so I'm not sure how the PSU is doing anything. I also checked with a DMM, my Live-Gnd voltage is .8V more than my Live-Neutral, so I think my plug is wired properly.
Any help is greatly appreciated. Thanks so much.
I built an OTL very similar to DIY Audio Projects Forum • Another Completed OTL Headphone Amp. It sounds fine, other than a few small issues
1. There is a slight hum even when the amplifier is off, when it is plugged it. If I unplug it completely the hum goes away, but if it's plugged in and off (via a power strip) then there is a slight hum/buzz. If I touch the ground pin on the plug (even tho my amp isn't connected to the ground pin) it stops.
2. Sort of building on that point, if the headphones are like half plugged in, then I get pronounced hum, even when the amp is off. If they are fully plugged it, the hum is faint but noticeable. If I touch the non-grounded terminals of the jack (tip or ring) I get a loud hum aswell.
3. When I turn the amp on, it hums very loudly for a few minutes, then that hum goes away (the hum from #1,2 remains)
I'm using an Antek Toroid. I currently have the tubes removed. Disconnecting the PSU ground solves the problem, but I'm measuring 0V AC on the transformer so I'm not sure how the PSU is doing anything. I also checked with a DMM, my Live-Gnd voltage is .8V more than my Live-Neutral, so I think my plug is wired properly.
Any help is greatly appreciated. Thanks so much.
Last edited:
From your description it appears that the central ground of your power supply (like the xformer center tap or, generally, the power ground) are not connected to the mains safety ground. So some parts of the circuit can float and present a high impedance to ground which may pick up mains field.
You say it disappears when you touch the ground pin of the plug (not sure which plug you mean, I am assuming either mains or headphone out).
You could try to temporary make that connection to see if it goes away.
Jan
If it hums when turned off you mains switch is not switching both lines OR you have some kind of (broken) death-cap before the breaker. FIX !
Also your chassies seems not properly grounded via the ground ( third lead) in the wall outlet, this is a security issue and should be fixed asap.
If you are uncertain about any of the above, leave the amp to a qualified technician !
Thanks for all the input. I have since tried a different transformer, a different PSU (a 21st century Maida by Tom Christensen) and the issue persists. I have checked my outlets and they are wired properly. It is clearly a grounding issues but I'm not sure what the issue is.
Here is a wiring diagram. All of the black lines are grounds. It is a modified star ground.
Here is a wiring diagram. All of the black lines are grounds. It is a modified star ground.
An externally hosted image should be here but it was not working when we last tested it.
ok, I admire your diagram. It rivals my chicken-scratch on the back of used old envelope schematics drawn with my daughter's pencil crayons. 
Are those MOSFETS on the E88CC/6DJ8? DN2540? What's with using the resistor to reference the gate? To ground? but its not the real ground is it? aha..... Then your "modified" star ground goes from that point all the way over to another star ground?
I don't thing that will work so well. Your PSU noise is going straight to your "modified" star ground point, where the cathode of your input tube is. This could easily be the source of your noise.
1. A quick fix might be to self-reference the CCS so it doesn't bring PSU noise right to the cathode of your input valve/tube. Ok, yes, this will take some adjustment, a resistor or two and a nice multi-turn potentiometer for each channel. but it will take that PSU noise right away.
2. A more permanent fix would be to use a ground bus instead of your "modified" star arrangement. This assumes you don't have enough wire to do a proper star ground... Just use one nice thick piece of solid copper wire and wire all grounds to it.
3. Now, while you're at it.... DN2540 is not that expensive, so why not cascade a couple of them? If it is DN2540 then make sure the upper MOSFET has a nice sized heatsink and use 15V zener diodes to protect the gates.
Still... I can't explain why you are getting hum with the power off, unless it is actually not off. In this case there might be other problems. Best post a picture.
Are those MOSFETS on the E88CC/6DJ8? DN2540? What's with using the resistor to reference the gate? To ground? but its not the real ground is it? aha..... Then your "modified" star ground goes from that point all the way over to another star ground?
I don't thing that will work so well. Your PSU noise is going straight to your "modified" star ground point, where the cathode of your input tube is. This could easily be the source of your noise.
1. A quick fix might be to self-reference the CCS so it doesn't bring PSU noise right to the cathode of your input valve/tube. Ok, yes, this will take some adjustment, a resistor or two and a nice multi-turn potentiometer for each channel. but it will take that PSU noise right away.
2. A more permanent fix would be to use a ground bus instead of your "modified" star arrangement. This assumes you don't have enough wire to do a proper star ground... Just use one nice thick piece of solid copper wire and wire all grounds to it.
3. Now, while you're at it.... DN2540 is not that expensive, so why not cascade a couple of them? If it is DN2540 then make sure the upper MOSFET has a nice sized heatsink and use 15V zener diodes to protect the gates.
Still... I can't explain why you are getting hum with the power off, unless it is actually not off. In this case there might be other problems. Best post a picture.
Last edited:
So I fixed the issue for power off (I believe) by connecting to earth.
The transistor is actually an MJE350 BJT. My PSU connects to that terminal, and the star is formed by connected the input/output grounds, with the signal grounds (which form another star themselves, as shown). I am getting some hum still. I tried an artificial CT using 100R resistors, and it helped a bit, but I still have noise, so maybe I have to use DC filament?
The power supply, I believe, has very little noise. I have an authentic 21st Century Maida, but I also made my own PCB's, one of which I am using. I would be extremely surprised if this was the source of my hum.
Is it possible that the hum is from the filaments only?
Thanks for the help by the way, I really truly do appreciate it greatly.
I would still suggest posting a photograph.
Is everything that is AC twisted pair or coaxial? I mean.. the filaments are connected to twisted pair... reference to ground with 100 ohm resistors (not floating)? And all signal is either twisted pair or coaxial?
also.. what is your source? any chance of a ground loop or hum from there?
Is everything that is AC twisted pair or coaxial? I mean.. the filaments are connected to twisted pair... reference to ground with 100 ohm resistors (not floating)? And all signal is either twisted pair or coaxial?
also.. what is your source? any chance of a ground loop or hum from there?
No Hun from source , inputs are grounded.
I have it prototyped, so most wires aren't twisted pairs. Maybe I should build it on the actual chassis and then troubleshoot?
I'm not sure what you mean by self reference the CCS.
I'll post pictures once I'm home from work.
I'm pretty confident that my PSU has extremely low ripple, but with only a DMM idk how to verify that
I have it prototyped, so most wires aren't twisted pairs. Maybe I should build it on the actual chassis and then troubleshoot?
I'm not sure what you mean by self reference the CCS.
I'll post pictures once I'm home from work.
I'm pretty confident that my PSU has extremely low ripple, but with only a DMM idk how to verify that
So I resolved that hum issue, but another one arose randomly that wasnt present before. When I don't ground the input (like if remove my alligator clip from ground to hlthe grid, I get loud hum (even if it's only 1 side ungrounded, I get hum in both channels). I have 330k grid leak resistors so I'm not sure what's going on
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.