Noise with Computers, USB DACs, and Powerline Ethernet Kits

[Paul Birkeland]

The diodes replace the IEC earth to chassis lug connection, nothing else.

[mcandmar]

https://www.mouser.ie/datasheet/2/358/typ_DEH-1275811.pdf

Came across these on Mouser, they are targeted at the mhz range but i wonder if they would suffice to stop ground loops in audio equipment?

[DanceTomato]

I accidentally ordered a few STTH2R06A, rather than STTH2R06. I have attached here: https://imgur.com/a/ieMnrfB

Can I use these in place of the the non-A variant?  E.G. something like this https://imgur.com/mVpuFLf

[Paul Birkeland]

I would absolutely not use surface mount diodes in this application.

Order 1N4001, 1N4002, 1N4003, 1N4004, 1N4005, 1N4006, 1N4007, UF4001/2/3/4/5/6/7.

[Paul Birkeland]

https://www.mouser.ie/datasheet/2/358/typ_DEH-1275811.pdf

Came across these on Mouser, they are targeted at the mhz range but i wonder if they would suffice to stop ground loops in audio equipment?

They are 25 freaking dollars each if you buy 50 at a time!

[Deke609]

They are 25 freaking dollars each if you buy 50 at a time!

Neat idea though.  And if they actually work for breaking ground loops, they might be worth it for a lot of people.

One could also just install a tiny inductor between the earth ground and chassis ground, but I can't find a high current rated one on Mouser. These are all a lot cheaper than the Schurter thingy, but none are rated for much more than 5A, and most for a lot less: http://www.mouser.com/Passive-Components/Inductors-Chokes-Coils/Fixed-Inductors/_/N-wpczZ1yzvvqx?P=1z0z819Z1z0z7l5Z1z0wohhZ1yzjx1iZ1z0wqhyZ1z0jky5Z1z0wqh6Z1yzjx12Z1z0wognZ1z0wqobZ1z0wqgxZ1yyt8gaZ1z0wqhwZ1yztq3uZ1yyt8f1Z1z0woglZ1z0wqo9Z1yzmpecZ1z0wkc2Z1z0j7djZ1z0wqhsZ1yxadcvZ1z0j5rrZ1z0wogjZ1z0wqo1Z1yztpvpZ1yi47jgZ1z0wkc1Z1z0wqhqZ1z0jkw7Z1yu90k7Z1yzuwe3Z1z0woghZ1z0wqo0Z1z0woh9Z1yvgg2oZ1yzuvkkZ1z0wqhoZ1yzakszZ1yxh8woZ1yy85mjZ1y9gfh8Z1z0wogfZ1z0wqppZ1z0wpgmZ1yu90l4Z1z0wfhjZ1z0wqhnZ1yzuvk8Z1yy7e5iZ1z0wqn1Z1yzsc00&Keyword=inductor&FS=True

I'm not sure if that would pass electrical code requirements.  I'm guessing one probably needs a current rating on the ground connection that is at minimum the same current rating as the breaker at the panel (e.g., 15A or 20A).

cheers, Derek

[Paul Birkeland]

I would bet that those would pass 20A and pop a circuit breaker pretty easily. 

[Deke609]

Fair point - it only needs to handle the current for a split second.

[Paul Birkeland]

Still, I don't think these would get the desired result.  For audible noise, these would be a dead short (it's a piece of wire after all) and would only present significant impedance to high frequency noise that wouldn't otherwise be audible, and that we would otherwise want to shunt to earth as effectively as possible.

[Deke609]

Yeah, that makes sense. they'd only be appreciably reactive in the MHz range.  The application example diagram in the Schurter datasheet shows them used with a mainframe computer and peripherals.

[DanceTomato]

I would absolutely not use surface mount diodes in this application.

Order 1N4001, 1N4002, 1N4003, 1N4004, 1N4005, 1N4006, 1N4007, UF4001/2/3/4/5/6/7.

Thank you Paul B, I have ordered some 1N4007 (1A/1200V) following your advice.

[cddc]

The maximum peak signal voltage you could send into a Crack with the volume control all the way up is 1.5V.  If you have 500mV peak of noise current because of a polluted ground, that's not a small voltage at all!

If I understand the paragraph correctly, the noise current is coming from the ground pin in your power outlet / power strip, and the pair of twisted diodes is to block the noise current from entering the Crack, am I correct?

[mcandmar]

They are 25 freaking dollars each if you buy 50 at a time!

~$5 from Mouser.  Will add one to my next order and measure them against the 10r/.1uf trap i am using in my pre amp.

https://www.mouser.ie/ProductDetail/Schurter/DEHL-10-0001/?qs=%2Fha2pyFaduglz%2FafWzmtwPY7x75HshQWsFSrhm4Ey1uJ9e%252BfusRydw%3D%3D

[Paul Joppa]

If I understand the paragraph correctly, the noise current is coming from the ground pin in your power outlet / power strip, and the pair of twisted diodes is to block the noise current from entering the Crack, am I correct?

Usually, what we call "ground loop" hum/buzz comes from capacitive coupling from the power transformer primary winding to the signal amp power supply secondary (the high voltage in tube amps). The capacitance is small, on the order of 100pF,  so it presents a high impedance at power line frequencies and can be shorted to ground through a 10-ohm resistor.

The diodes are there in case there is a fault that presents a low-impedance source, such as a short from the power line to the signal ground. In normal use, they are not active and have no effect.

[cddc]

Usually, what we call "ground loop" hum/buzz comes from capacitive coupling from the power transformer primary winding to the signal amp power supply secondary (the high voltage in tube amps). The capacitance is small, on the order of 100pF,  so it presents a high impedance at power line frequencies and can be shorted to ground through a 10-ohm resistor.

The diodes are there in case there is a fault that presents a low-impedance source, such as a short from the power line to the signal ground. In normal use, they are not active and have no effect.

Thanks a lot for the explanation, PJ. I could hardly understand what you said, mainly due to my ignorance in electronics.

1.) so the capacitive coupling (first time to hear that coupling in the transformer  ) from the power transformer primary winding to the secondary windings will pass some small line current (110V 60Hz) to all secondary winding circuits (the 6.3 VAC heater circuit and the 2x0 VAC unrectified B+). But how the capacitive currents are grounded, I can't find any 10-ohm resistor in Crack?

2.) the twisted diodes are breaking the ground loop if I understood PB's 'ground breaker' correctly, so why they are not active and have no effect?