This topic is of much interest to me as well. For the past fifty years I have lived with audible effects of ground loops in home stereo (and before that, mono), as well as in radio stations.
At the present I have a system using a Threshold FET ten/hl driving a Crown D-75A. Connected to the FET ten are a SACD player, phono preamp, and computer sound card. The FET ten is unbalanced of course, the D-75A is balanced, the phono preamp is unbalanced, the SACD player has both balanced and unbalanced outputs.
By far most of the noise in the system was coming from the computer. I installed a Jensen ISO-MAX between the sound card and the FET ten, and the improvement was unbelievable. The SACD player noise was not nearly as bad, but the ISO-MAX cleared that up too. Since I have only one ISO-MAX I am building a box for more transformers.
What I am working towards is isolation of all devices from the FET ten, as discussed in this article by Dave Davenport:
http://www.diyaudio.com/forums/diya...-grounding-interconnection.html#content_start
With the equipment chassis connected only by their safety grounds (or battery powered in the case of the phono preamp) no ground current should flow and there will be no ground loops. This may sound too good to be true but I have heard it, or rather didn't hear it.
Having read the articles linked above by marce, I wonder why the author does not consider transformers as galvanic isolation? Using a heavy shield to have a smaller voltage drop for ground currents seems like a Band Aid to me.
At the present I have a system using a Threshold FET ten/hl driving a Crown D-75A. Connected to the FET ten are a SACD player, phono preamp, and computer sound card. The FET ten is unbalanced of course, the D-75A is balanced, the phono preamp is unbalanced, the SACD player has both balanced and unbalanced outputs.
By far most of the noise in the system was coming from the computer. I installed a Jensen ISO-MAX between the sound card and the FET ten, and the improvement was unbelievable. The SACD player noise was not nearly as bad, but the ISO-MAX cleared that up too. Since I have only one ISO-MAX I am building a box for more transformers.
What I am working towards is isolation of all devices from the FET ten, as discussed in this article by Dave Davenport:
http://www.diyaudio.com/forums/diya...-grounding-interconnection.html#content_start
With the equipment chassis connected only by their safety grounds (or battery powered in the case of the phono preamp) no ground current should flow and there will be no ground loops. This may sound too good to be true but I have heard it, or rather didn't hear it.
Having read the articles linked above by marce, I wonder why the author does not consider transformers as galvanic isolation? Using a heavy shield to have a smaller voltage drop for ground currents seems like a Band Aid to me.
The author Tony Waldron is probably more of an expert on these sort of issues than most if not all of us on DIY audio, he contributes to the Compliance Club with Keith Armstrong, so I'd reccomend his comments even if they dont fit in with your beliefs.
This is not band aid this is how its done in the realy real world.
This is not band aid this is how its done in the realy real world.
I am no expert but the way it used to be done in the really real world of radio stations was to generate hum at power line frequency and appropriate harmonics and introduce the frequencies out of phase till the signal hum and buzz were canceled. Times change and I'm not saying I think this is a good solution for home audio.
Nonetheless I am reasonally certain that any time ground current other than DC flows in an unbalanced signal line there will be noise. It is just a question of how much. One could make the counter argument that any time a signal goes through a transformer there will be distortion. It is just a question of how much.
Sorry for being blunt with my reply, but I am trying to get accross a different view point and different techniques that with todays huge increase in electrical noise of all frequencies especialy RF, we need to explore all possible solutions. The main point with RF screening is screens connected at both ends!, this means we are going to have to tackle potential "ground loops" at some point in the system as we are always also going to have a protective earth as well as screens and signal return currents. As audio frequecies are in that awkward range where return paths will follow the path of least impedance, this path will vary between resistive and inductive with the frequency, so again quite often the return current may be going where we dont think it is going.
Waldron says:
"Galvanic isolation is the preferred alternative to the use of PECs and terminating cable shields at both ends. The method of isolation used: fiber optics, wireless, microwave, laser, infra-red, must have transmitter/receiver electronics rated at the full fault/surge voltage (rarely a problem)."
Can you tell me why transformers are not considered among the means of galvanic isolation?
Yes, twin core for signal and return, balanced or LVDS for digital with some form of isolation is the best solution, with a complete faraday cage shielding for RF/EMI immunity.
I dont know why transformers aren't listed, we use planar transformers quite often for critical digital and analogue signals with GND plane between windings to cater for capacitive coupling. Also used them to create a proper analogue island where the power from the SMPS was passed through the transformer, with the shield planes etc it greatly reduced the noise from both the SMPS's and the digital circuitry noise. One advantage of the above methods is a freedom from power surges, but they can be catered for.
I dont know why transformers aren't listed, we use planar transformers quite often for critical digital and analogue signals with GND plane between windings to cater for capacitive coupling. Also used them to create a proper analogue island where the power from the SMPS was passed through the transformer, with the shield planes etc it greatly reduced the noise from both the SMPS's and the digital circuitry noise. One advantage of the above methods is a freedom from power surges, but they can be catered for.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.