Thanks for your reply Salas. My DCB1 was originally connected that way but without the parallel diodes, I've ordered some to try, a straight wire link did not work. I was getting a small amount of hum through the system with my new amps (My-Ref FE's), not enough that I could hear it from my sitting position but it's there nonetheless. The article I linked to points out that equipment signal rca socket body's should be connected to star ground by a wire link, however that made no difference whatsoever so I'm back to square one.
It's transformer hum so I might try removing the transformer from the chassis using longer leads to see if that makes any difference.
It's transformer hum so I might try removing the transformer from the chassis using longer leads to see if that makes any difference.
Last edited:
It was not me who wrote the article.
For what it's worth I have now read several articles on single-ended system grounding to try to get a handle on this and virtually all of them show signal input rca body's connected to main star ground via a short length of wire.
I'm just trying to find out what is correct and the best solution for eliminating hum. Maybe I should just give up.
For what it's worth I have now read several articles on single-ended system grounding to try to get a handle on this and virtually all of them show signal input rca body's connected to main star ground via a short length of wire.
I'm just trying to find out what is correct and the best solution for eliminating hum. Maybe I should just give up.
There are many "correct" ways until we face a problem. Then we follow alternative "correct" ways. Since in your system the all returns to one point scheme is not eliminating some harmonic noise, its normal you try signal and PSU returns safely lifted/floated. If its due to your Tx field, a bit of rotation will be changing the buzz and you will know you got to distance it.
I should have the diodes in a few days, it will be interesting to see if they help with this issue. I have tried rotating the transformer but it made no difference. Maybe removing it from the case will but i'll try the diodes first.
Maybe using isolation transformers is best when a system has both class 1 and Class 2 components. Hopefully I wont need them though.
Maybe using isolation transformers is best when a system has both class 1 and Class 2 components. Hopefully I wont need them though.
Last edited:
I doubt you will not find an easy enough way. There are hundreds DCB1s of all versions out there, maybe thousands*, working silent as mouses. Yours is no different i.e. not inherently problematic for hum and buzz.
*(Never investigated the sum of boards given out since single layer black).
*(Never investigated the sum of boards given out since single layer black).
I do not wire the RCA shield to the system ground independently. I assume that gets connected via PCB ground.
This unit gets better every day as it breaks in.... I am going to measure the jfets. Only laziness stopped me. Now, I'm hooked. Thank you Salas for a great piece
And I have yet to tweak it. I can only imagine a balanced version would be even better!
Nah, I've seen inside too many so called professionally built products over the years. Some really shocking stuff around although things have improved a lot in recent years. The 5 LED group layout could just be put down to their leads being bent out of place. I stand by my original point 
Nice work!
Nice work!
davym,
If you take a look at post 4, you will see the board traces go from the gnd pnt at the molex connector to the power supply common -- there is no need to wire the RCA sheilds to anything else.. they are already going there via the molex connector
Mine is wired RCA -> Potentiometer -> Molex. Simple as that.
If you take a look at post 4, you will see the board traces go from the gnd pnt at the molex connector to the power supply common -- there is no need to wire the RCA sheilds to anything else.. they are already going there via the molex connector
Mine is wired RCA -> Potentiometer -> Molex. Simple as that.
Last edited: