AU-717 chassis safety grounding?

[AudioModern]

I am wrapping up a rebuild of my AU-717 and was been thinking of installing a new mains cable with a ground wire, and properly grounding the chassis.

I know from testing continuity that the RCA sockets (shield) are connected to the chassis, so obviously this means the audio ground isn't 'floating'. If I ground the chassis via a new mains cable, I may introduce hum.

I built a Pass F5 amplifier recently that floated the audio ground from the chassis using a CL-60 thermistor and monolithic diode bridge in parallel. The audio ground is essentially isolated from the chassis unless there is a fault causing the audio ground to be energized.

So I was thinking... I know there is a spot between the two driver assemblies where two black wires connect to the chassis. One is from the equalizer board (can't remember where the other connects). What if I tied these two black wires together, then put a monolithic diode bridge / CL-60 between them and the chassis. Assuming (big assumption) this is the only place where audio ground connects to the chassis, then it would now be floating. And if I installed a new mains cable grounded to the chassis, then it should now be safely grounded without hum.

I've read another thread where someone suggested chassis grounding without floating audio ground should be ok, but I have a feeling there's a reason why manufacturers tend to isolate audio ground from the chassis...

Any thoughts? Of course I will experiment by grounding the chassis to a receptacle directly (easy with Schuko sockets) and see if there actually is any hum, before I do anything more complicated.

 

[Hyperion]

Of course I will experiment by grounding the chassis to a receptacle directly (easy with Schuko sockets) and see if there actually is any hum, before I do anything more complicated.

Yes, do this. I don't think you should go inside an AU-717 and start selectively lifting chassis grounds and adding components in there - you'll get yourself into a world of hurt.

If your hum tests are satisfactory, replace the mains cable and connect the ground lead to an existing 'star ground' chassis point, usually where many black wires meet at a chassis fixed solder post. Don't worry about ground loops until you get one, they most often happen when the different items that make up a system are connected to more than one mains power socket.

 

[DougBrewster]

If your hum tests are satisfactory, replace the mains cable and connect the ground lead to an existing 'star ground' chassis point, usually where many black wires meet at a chassis fixed solder post. Don't worry about ground loops until you get one, they most often happen when the different items that make up a system are connected to more than one mains power socket.

Would that advice be valid for the three items (receivers and amp) in my signature?

 

[Hyperion]

Would that advice be valid for the three items (receivers and amp) in my signature?

Yes it would. - so far I have done this to AU-D11 II, B-2102, C-2102, AU-X1, AU-919, AU-X701, AU-717, G-7000, AU-X711, AU-6600 (and a few others), - without any problems and NO problems with ground loops.

I think there has been discussion before on why grounded (3 core) power cords were not fitted as standard. I think it was because the electrical safety laws/conventions in the many countries where this gear was is/was used were so varied, overall it was safer not to have it - not sure on this

Now, things are a lot clearer and I personally think improving the electrical safety of such gear, by fitting a ground capable mains lead is a worthwhile exercise. Add to that, the 'figure of 8', 2 core (single insulated) 'shaver lead' style of mains cable does not inspire confidence safety-wise and can often show signs of damage after all these years of use. So even if you never connect the ground or earth core, fitting a new, double insulated, 3 core cable is a step up in electrical safety.

 

[thanhky1985]

Vâng, nó sẽ như vậy. - cho đến nay tôi đã thực hiện điều này với AU-D11 II, B-2102, C-2102, AU-X1, AU-919, AU-X701, AU-717, G-7000, AU-X711, AU-6600 (và một một vài người khác), - không có bất kỳ vấn đề nào và KHÔNG có vấn đề gì với vòng lặp trên mặt đất.

Tôi nghĩ trước đây đã có cuộc thảo luận về lý do tại sao dây nguồn nối đất (3 lõi) không được lắp theo tiêu chuẩn. Tôi nghĩ đó là do luật/quy ước về an toàn điện ở nhiều quốc gia nơi thiết bị này được sử dụng rất đa dạng, nhìn chung sẽ an toàn hơn nếu không có nó - tôi không chắc về điều này

Bây giờ, mọi thứ đã rõ ràng hơn rất nhiều và cá nhân tôi nghĩ rằng việc cải thiện độ an toàn điện của các thiết bị như vậy bằng cách lắp một dây dẫn điện có khả năng nối đất là một việc làm đáng giá. Thêm vào đó, kiểu cáp nguồn 'hình 8', 2 lõi (cách điện đơn) không mang lại sự tự tin về mặt an toàn và thường có thể có dấu hiệu hư hỏng sau ngần ấy năm sử dụng. Vì vậy, ngay cả khi bạn không bao giờ nối đất hoặc lõi đất, việc lắp cáp 3 lõi mới, cách điện kép là một bước tiến về an toàn điện.

I am in the middle of a project to replace the 3 prong power cord and ground it directly to the wall outlet for the sansui 2000A.

 

[AE Bob]

As long as you only have one device with a 3 wire cord (with earth ground pin), you won't have an issue. The problem occurs when you have more than one device connected to earth ground. The issue occurs because with multiple devices connected to ground, a loop has been created from ground through the chassis and patch cords to the other chassis and back to ground. This ground loop will have a current induced in it from the 60 (50) Hz we are bathed in due to the fields generated by our house wiring. The problem is that the shield of the patch cords which carries this "hum" current is also carrying the audio reference ground (0 signal). When they get mixed you get ground loop hum. As long as only one device goes to earth round, no loop is created, and there is no ground loop hum.

 

[lgdavis]

Recently I was listening to my FM tuner (Pioneer TX 9500), and decided to go over to it and change the station.
As soon as I touched it, felt a spark jump from my finger - and the tuner immediately went to noise.
Knew right away that I had zapped the input FET .
Of course it's unobtanium, but finally found one on the auction site - from China.
Replaced the FET, and the power cord - with a 3-wire grounded one.

So if you have FM - tuner or receiver, take heed .

 

[Hyperion]

In the UK we don't seem to have as many instances of hum loops as maybe the US or some other countries. I can only guess that our 'Ring Main' household wiring architecture means that more often than not, sound system devices are connected to effectively the same ground.

 

[AE Bob]

The issue is not that the devices are connected to different grounds. The problem occurs because the loop created by the multiple grounded devices (in the same grounded outlet) and the connection between the devices by the patch cords, creates a single turn loop (shorted single turn transformer), which picks up the 60 (50) Hz magnetic fields that surround us from house wiring and utility transmission lines etc. That single shorted turn has AC current in it, and part of the single turn loop is carrying the audio signal. This will not occur if only one device in the system is connected to ground through a three wire cord.

 

[Jefft91]

I first want to clarify that I'm in the U.S., and don't know much about how the rest of the world deals with the polarity in plugs, or if they do. In the U.S., all new electrical devices use a polarized plug where one prong is wider than the other, so the plug and the recepticle are keyed so that the neutral prong cannot plug into the hot side of the recepticle.

I've been following this thread, and wondering why one couldn't just use a polarized plug that has no ground. If we were to re-wire a device using a polarized plug, and assure that the common (return leg) was connected to the chassis, then there should never be voltage on the chassis (unless there is a short somewhere, which should trip the breaker). With this approach, we would eliminate ground loops and eliminate the potential to put line voltage on the chassis.

 

[thanhky1985]

I first want to clarify that I'm in the U.S., and don't know much about how the rest of the world deals with the polarity in plugs, or if they do. In the U.S., all new electrical devices use a polarized plug where one prong is wider than the other, so the plug and the recepticle are keyed so that the neutral prong cannot plug into the hot side of the recepticle.

I've been following this thread, and wondering why one couldn't just use a polarized plug that has no ground. If we were to re-wire a device using a polarized plug, and assure that the common (return leg) was connected to the chassis, then there should never be voltage on the chassis (unless there is a short somewhere, which should trip the breaker). With this approach, we would eliminate ground loops and eliminate the potential to put line voltage on the chassis.

In the house circuit breaker diagram. The neutral wire is colored white Connected to the ground wire at the circuit breaker panel. That happened when you connected the ground wire from the socket to the case. The negative current went into the ground wire and went back to Neutral wire. If your circuit breaker has the word test on a button next to the switch, it will turn off the breaker. This is also a concern for many people when their home has Get that kind of circuit breaker. It's called a Gfci.

 

[AE Bob]

Connecting either side of a two wire cord (even the large blade of a polarized plug) directly to the chassis WOULD BE ABSOLUTELY WRONG AND DANGEROUS. Most of the US devices have a connection from the AC cord to the chassis through an "isolation network". The network typically consists of a small value disk capacitor in parallel with a 2.2 or 3.3 meg ohm resistor. The impedance of the network assures there cannot be enough current in any loop to create hum, as well as assuring there is no where near enough current to be a shock hazard.

 

[thanhky1985]

I first want to clarify that I'm in the U.S., and don't know much about how the rest of the world deals with the polarity in plugs, or if they do. In the U.S., all new electrical devices use a polarized plug where one prong is wider than the other, so the plug and the recepticle are keyed so that the neutral prong cannot plug into the hot side of the recepticle.

I've been following this thread, and wondering why one couldn't just use a polarized plug that has no ground. If we were to re-wire a device using a polarized plug, and assure that the common (return leg) was connected to the chassis, then there should never be voltage on the chassis (unless there is a short somewhere, which should trip the breaker). With this approach, we would eliminate ground loops and eliminate the potential to put line voltage on the chassis.

Here is a video of what I am talking about. It shows that your thoughts are real.

 

[thanhky1985]

Has anyone in our group tried isolating the ground wire from the transformer. I mean in each transformer of 1 amp.There are many coils that produce different currents. And each coil has its own ground wire. But my sansui 2000a has 3 ground wires. All gathered at 1 point. I don't know about other amps. But as far as I know.When you encounter background hum. Just isolate the ground wire of the power supply circuit for the power board and the power supply circuit for the tone board. Then the noise will disappear. I will try that in a few days

 

[Hyperion]

The issue is not that the devices are connected to different grounds. The problem occurs because the loop created by the multiple grounded devices (in the same grounded outlet) and the connection between the devices by the patch cords, creates a single turn loop (shorted single turn transformer), which picks up the 60 (50) Hz magnetic fields that surround us from house wiring and utility transmission lines etc. That single shorted turn has a current in it, and part of the single turn loop is carrying the audio signal. This will not occur if only one device in the system is connected to ground through a three wire cord.

Thank you for this Bob, I hadn't fully appreciated that when the ground loop is formed it makes the various conductors into an aerial/antenna and this is what creates the problem. Previously I was under the impression that just the presence of more than one grounding point, maybe at slightly different potentials was responsible. Or simply by disconnecting the two (or more) ground paths introduced a discontinuity into the grounding scheme causing 50/60 Hz stray fields to interfere with the signal path screening.

That said, I still haven't had any problems connecting the earth core of the mains cable to the star ground point in various pieces of audio equipment.