Thanks. I have tried grounding the circuit to chassis at the RCA socket with 10mm distance only and it works no problem. The sound seems to be also improved but I need more time to confirm this. I found 8v ptp resonance at 20kHz on one channel after yesterday's experiment so I need to debug it first.
Where things are grounded is extremely important.
I've learned through experience when dealing with pre-built stuff that I'm restoring not to mess with the grounds if there are no ground loops regardless of how wrong the grounding scheme is.
I've also learned that to solve some ground loops one sometimes has to attach a ground from one point to an unused terminal on a socket or terminal strip in the amp. Don't know why that works, but it sometimes does.
I've learned through experience when dealing with pre-built stuff that I'm restoring not to mess with the grounds if there are no ground loops regardless of how wrong the grounding scheme is.
I've also learned that to solve some ground loops one sometimes has to attach a ground from one point to an unused terminal on a socket or terminal strip in the amp. Don't know why that works, but it sometimes does.
Jan and Abraxalito are right. All were said in Jan's first reply. #1 wiring grounding at chassis at input is the best option. I think it sounds the best. At least I tried that on my power amp. Although I liked wiring 3 on my preamp, it happens that my preamp chassis connection point is after the volume pot , which really belongs to the input area, so the effect should be similar to wiring 1 grouding at the input.
Your next step may well be to explore transformer isolation as this tends to reduce the common-mode currents in the interconnecting cables.
Abraxalito, I am interested. But last time when I tried to use a common mode ferrite at the input I had mixed result. Where can I find a transformer / inductor / ferrite that doesn't distort the signal?
By the way, I took my family to Huangzhou less than 2 months ago. OMG West Lake was so incredibly crowded!
By the way, I took my family to Huangzhou less than 2 months ago. OMG West Lake was so incredibly crowded!
A common-mode ferrite is rather different from a transformer - I tend to think of a common-mode choke as a transformer turned through 90 degrees. All transformers are going to contribute some distortion, particularly at LF. By this I mean measurable distortion on a sinewave signal. However unless overloaded, I'm not aware of transformer distortion when listening to music.
I made my own isolation transformer using a very cheap ferrite core set (EE26 with 10K core material). Next time you're in Hangzhou message me and I'll show you where to buy them 😛 I haven't found a local source for the very thin wire needed, that I get from Taobao.
I made my own isolation transformer using a very cheap ferrite core set (EE26 with 10K core material). Next time you're in Hangzhou message me and I'll show you where to buy them 😛 I haven't found a local source for the very thin wire needed, that I get from Taobao.
Grounding is a small part of a system design. <full stop> messing with various interconnects to your input to your pre amp which may have flaws to begin with isn't a universal truth. Usually any unbalanced currents coming from input cables is a flawed system. This a symptom of a problem between boxes. Fully trouble shooting is best done 1st to identify what the problem is! Have you done this?
The easiest band-aid for single ended interfaces is increasing signal gnd resistance to force current to take other low resistance paths. ( your approach is reducing gnd resistance to lessen the I^2*Rg, this is less effective than reducing the bad current ) Differential inputs with screen shields is the best solution, this what pro sound does. Transformers if done right, besides being expensive often have other draw backs. Spec a transformer correctly then go shopping. Peak input levels and LF response determines the core size / cost. PS you can buy cheap transformers for autosound for less than 20, helpful to keep around for trouble shooting. Battery powered sources / amps are good tool too.
The easiest band-aid for single ended interfaces is increasing signal gnd resistance to force current to take other low resistance paths. ( your approach is reducing gnd resistance to lessen the I^2*Rg, this is less effective than reducing the bad current ) Differential inputs with screen shields is the best solution, this what pro sound does. Transformers if done right, besides being expensive often have other draw backs. Spec a transformer correctly then go shopping. Peak input levels and LF response determines the core size / cost. PS you can buy cheap transformers for autosound for less than 20, helpful to keep around for trouble shooting. Battery powered sources / amps are good tool too.
Last edited:
Wow so many solutions to a problem that hasn't even been identified.
This was the old way is improper because OEM are doing it another way is plainly sayin one way is my way. Have any of you realized all major OEM audio is Class II and we are probably* talking about Class I gear combined with II? the blind leading the blind. I think HiFiNut knows more that this group combined he realizes the XFMR to chassis is the heart of his input noise injection.
*the OP needs to show ground details what pieces of gear he is connecting up.
This was the old way is improper because OEM are doing it another way is plainly sayin one way is my way. Have any of you realized all major OEM audio is Class II and we are probably* talking about Class I gear combined with II? the blind leading the blind. I think HiFiNut knows more that this group combined he realizes the XFMR to chassis is the heart of his input noise injection.
*the OP needs to show ground details what pieces of gear he is connecting up.
- Status
- Not open for further replies.