Transcendent Sound Forum
Latest Topics

  Author   Comment   Page 2 of 3      Prev   1   2   3   Next
Reply with quote  #16 


The posted ESP circuit can reduce audible AC hum but it doesn’t really "block" anything. It's definitely helpful and could be used together with the SOB and DAC in addition to the polarity change if that's not enough. It wouldn’t work for the original  MP as it uses the chassis as ground plane for the audio ground and the chassis needs to be insulated from the audio ground and only be connected to earth.

The "zero volt line" of the ESP circuit would be connected to the "audio ground" of the SOB (speaker ground, psu ground, RCA ground disconnected and insulated from chassis).The chassis of the SOB would be connected to safety ground/earth only.

It would be interesting to understand better what causes the hum. Because the ground loops which you get if you connect your SOB, DAC, MP to safety ground/earth have the reason in some potential difference of each unit  referenced to earth ground. That would also explain the bad SQ if you connect all three through the interconnects to only one safety ground.

The ESP circuit is a fix, changing polarity of the primaries of the power transformers could get rid of the cause for the potential difference to earth. But that needs to be tested practically. Sometimes there is a big difference when switching polarity, sometimes it's hardly noticeable.



Melvin Khoo
Reply with quote  #17 
Thanks very much Wolfgang, I am keen to try everything and will report back what happens. I would like to seek some clarification on what you said, please.
You indicated that both the MP and SOB chassis only should be connected to Earth ground, and must be isolated from / not connected to audio ground. This means, e.g. T13 in the MP must be insulated from the chassis, and so T13-B is not connected to the chassis. The question I have here is: since T13-B is the star ground for audio ground, and would now be disconnected from the chassis (and consequently from Earth ground), what "ground" is T13-B connected to then? In other words, how is the star ground referenced to "ground"? Wouldn't it be floating?
In the case of the SOB, the ground lug for audio ground is located between the speaker binding posts and the RCA output jacks. The ground lug is connected to the chassis, which is connected to Earth ground via terminal strip T1-C. If you note carefully, the star ground in the SOB is the ground buss located between the two speaker binding posts. This is in turn connected to the ground lug via a single black wire. I have taken a photo of my internal real panel to illustrate what I mean. So my question here is: do I disconnect the ground connection from the RCA jacks to the ground lug only, or should I also disconnect the black wire from the ground lug that comes from the ground buss between the binding posts?
My third question is about reversing the polarity of the transformers: are you referring to only the 250VA power transformer, or to the 120VA transformer as well? In addition, wouldn't reversing the polarity also reverse the phase between this device and the other devices? If so, would this have any SQ consequences?
Thank you for all the help so far!

Attached Images
Click image for larger version - Name: 20180131_023653-1.jpg, Views: 28, Size: 1.63 MB 

Reply with quote  #18 


Yes, if you fully disconnect the audio ground from earth ground it is floating referenced  to the earth ground and no typical ground loop can form. If you only connect the metal chassis to earth ground your chassis is safe in case of a fault but your audio ground could theoretically carry some voltage (if you touch RCA, speaker terminals) in case of a fault of some kind. That’s why both need to be connected to earth ground in order to be 100% safe and according to code.

If this connection results in a ground loop and you want to keep the safety rules you can only try to find the cause for the ground loop (like polarity of the power transformers for example). To keep the audio ground floating and only connect the chassis to earth ground is the easiest way with no neg. side effects to SQ but it’s not 100% safe in case of a fault. That has to be very clear if you consider doing this.

To use the ESP circuit is a compromise between floating audio ground and safety rules because it reduces the audible effect of a possible ground loop to some degree while keeping the earth connecting fully operational in case of a fault.  It’s 100% safe but not the best solution if you really want to get rid of the problem and not just shovel it below the audible level.It might help to reduce hum in most cases.

The MP belongs to different category regarding grounding as it uses the chassis as ground plane. You would have to insulate the soldering lugs with ground connection from the chassis and make connections with a cable in order to use the ESP circuit together with the MP. But that would be only my last choice. Let's first try to find out which gear is the main cause for the hum.

You only have to try different  polarity with the power transformers (250VA in case of the SOB). Changing polarity of the power transformer primaries has no effect on anything of the audio circuit. It will only change the potential difference of the audio ground to earth ground more or less.

If you don’t mind you could first test a few things before going any further.

First test:

If you connect only the SOB and the MP via interconnects  (MP out to SOB in)and  both to earth ground via power cable is there any hum? If there is hum simply by connecting the SOB to the MP we need to start here. No need for further testing.

If no hum, the problem must be connected  to  the DAC. In this case connect the DAC with interconnects to the MP but without any input connected to the DAC. Does this form the ground loop ?

If no hum, reconnect the input to the DAC (USB etc) . Any hum?

Second test:

If adding the DAC (with no input) has contributed to or created the ground loop together with the SOB/MP try to connect it directly to the SOB without the MP (just for checking, not listening to music). Is the ground loop still there?

If there is no hum, connect the DAC (with no input connected) to the MP (disconnected from the SOB) and listen with headsets if there is hum?

If there is hum in both cases is it the same amount?

It would help already a lot to know which gear is responsible for the ground loop or adds to it.The next thing would be to find out how a polarity change would affect the ground loop starting with the gear that is most likely responsible for the hum as a result of the connection test.

Reply with quote  #19 
Hi Wolfgang, I have done some tests and here are the results:

1.  SOB connected to MP only, both mains Earth grounded, single interconnect:

a.  Switched to empty RCA channel 1 - hums fairly loudly.

b.  Switched to shorted RCA channel 2 - no hum, very quiet.

2.  DAC connected to MP only, both mains Earth grounded, single interconnect, headphones connected, no streamer or other input connected to DAC:

a.  Switched to DAC on RCA channel 1 - hums fairly loudly.

b.  Switched to shorted RCA channel 2 - very, very faint hum, barely audible on headphones.

c.  Switched to empty RCA channel 3 - no hum, very quiet.

3.  DAC connected to SOB only, both mains Earth grounded, single interconnect, no streamer or other input connected to DAC:

a.  Fairly loud hum.

4.  SOB alone, no interconnect on RCA output jacks - very slight hum, only with ears less than 30cm from speakers.

So it is still a problem of ground loops in all three devices.

What should the next steps be?

Thank you for your continued help.
Reply with quote  #20 

it looks like the DAC could be the source for the ground loop in combination with SOB and MP.

What DAC do you use? Would be interesting to take a look inside. What power supply and grounding?

The next  logical step will be switching polarity of the DAC power transformer with the help of a modified power cable in order to find out if the potential that causes the loop can be reduced by switching the phase. 


Reply with quote  #21 
Yes Wolfgang, correct, this is why I said that I needed to lift the ground pins for 2 out of 3 of my devices in order for the ground loop humming to stop.

My DAC is the R2R Denafrips Terminator which has two power transformers, one 250VA and one 60VA, and both are shielded from the audio section, see the following Web page for photos of what it looks like:

I understand your suggestion to invert the primary wires on the power transformer(s), and instead of doing this internally in one or more of the devices, wouldn't it make more sense to do this on the power cords instead? I made my own power cords and connectors, so I could just simply flip the Live and Neutral connections in one or more of them to test if this solves the problem. How does this sound to you?

Thanks very much.

Reply with quote  #22 
I have gone ahead to flip the Live and Neutral wires inside one of my power cables, and did the same tests as set out in a previous post. Sadly, it did not resolve any of the humming issues in each case.

So I guess I will have to take the next steps, which I am assuming would be to keep all mains Earth connections to chassis intact, but decouple and isolate the star grounds in the MP, followed by the SOB, or the other way around, or both, and to see if any of these steps stops the ground loop humming.

Will update with results after I do these.

Reply with quote  #23 
Did you try the modified power cable also with the SOB and MP (one at a time)? No change at all with SOB/MP/DAC connected?


Reply with quote  #24 
Hi Wolfgang, I did the modified power cable test yesterday (where I switched the Live and Neutral wires inside the power plug) and this did not resolve the humming problems. The humming was just as loud. I will repeat this test again tonight.

Thank you.


Reply with quote  #25 
I just read this quickly to try and understand all the tests you have completed. I have a similar set-up; couple questions that may or may not rule out any one component.

Similar to test 4 above, have you fed the SOB with a battery operated (such as a phone) source? Should be strikingly silent.

Looking at your picture, is that covered cable in the back a twisted pair? From Tubes and circuits, that should be a single wire. Also notes that all circuit grounds should wire directly to bus, where your RCA’s touch chassis first. Did lifting that chassis connection fix the hum?

Same for the MP, listening to headphones. I didn’t read that you have confirmed each device is silent by itself as-built.
Reply with quote  #26 
Wolfgang, I confirm that using the modified power cable where Live and Neutral were flipped in the connector head did not make a difference to the ground loop hum. This was applied to all 3 devices one at a time.

Reply with quote  #27 
Hi Carson, Thanks for the help.

Both the MP and SOB alone are very quiet, no buzzing and only a very slight hiss if my ear is up next to the tweeter in the case of the SOB.

That white cable at the back is a twisted pair going from the binding post ground buss to the power filter cap ground buss. In the SOB build manual, it does state to use two wires twisted together to connect both ground busses. I used a twisted pair with silver plated copper strands since this was on hand.

The MP is also quiet with no humming if I only have a battery powered source connected.

Frankly, there's nothing wrong with any of the 3 devices. It's just a classic case of ground loop and I'm quite sure that the RCA interconnects are acting as the alternate ground paths between the 3 devices. This is why I said that the humming goes away if I lift the Earth pins in 2 out of 3 of the power cables.

Ideally, I'd like to keep safety Earth connected to chassis for all 3 devices, and at the same time keep a ground reference for both the SOB and MP. I'll leave the DAC unmodified as its still under the 3 year warranty. I believe the SOB should be easy enough to do that with an external ground box device, but the MP uses the chassis as ground plane in that tab C (the middle tab) in many of the terminal strips is used as ground connection which in turn is connected to the aluminum chassis. Not sure if it would be practical to undo this, and re-wire individual wires to the star ground point at Terminal strip T13-B.

Any suggestions would be greatly appreciated!  Thanks!
Kelvin Tyler
Reply with quote  #28 
Mel. See my post No3. Try just unscrewing input ground lug from chassis on SOB. Leave black wire in place. Make sure input RCA jacks are truly insulated from chassis.
Kelvin Tyler
Reply with quote  #29 
On occasion,I have 'solved' ground loop problems by the simple expedient of running the various connected devices off the same mains adaptor/multiblock. This gives all devices as common a ground source as possible.
Reply with quote  #30 
The link below is another helpful description of the problem which includes solutions like star grounding of the gear etc. Maybe it can also make it a little more clear why phase and polarity settings of the primaries can make a difference under certain conditions. Of course this article is about wiring studios but some  parallels can be found that can be also useful for home audio use.

I had a computer problem and lost some folders with circuit drawings  so I have to recreate the circuit for the polarity tester using my old tester. But I will post it soon because I think it can help to solve certain hum problems and it's easy to build and cheap.
Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.