If multiple inputs have a common grounding connection at the Source, then connecting the multiple input interconnects CREATES LOOPs
You have to reduce the interference current in these loops to reduce the interference voltage that is added to the input signal.
One normally ADDs resistance into the LOOP at a location that does not interfere with passing a Signal. The added resistance reduces the loop current and that reduces the interference voltage along the signal return wire.
You have to reduce the interference current in these loops to reduce the interference voltage that is added to the input signal.
One normally ADDs resistance into the LOOP at a location that does not interfere with passing a Signal. The added resistance reduces the loop current and that reduces the interference voltage along the signal return wire.
Hello Andrew
"You have to reduce the interference current in these loops to reduce the interference voltage that is added to the input signal."
How can interference current & voltage be created when you only connect two or more inputs together and nothing is connected at the end of them?
Differently stated, the end result, which is a ground loop, is created as soon as you connect inputs together... or... these inputs individually connected to individual outputs from a preamp. It's a freak-out situation!
For your last sugg I tried with 100 ohms resistors on shield to inputs in various ways...5 out of six with resistors and one directly connected...all six resistors..and some tests have been made with direct connections to star-point...Zilch!
Ooooooh! Thay wacky wabbit s'gonna make me cwazzy!
Luke
Hi Luke123,
Have you tried tying all of the input grounds together, and then connecting them to the star with a single wire? Like this:
Mike
Have you tried tying all of the input grounds together, and then connecting them to the star with a single wire? Like this:
Mike
Consider a single channel amplifier connected to a Source.
The input cable is a two wire connection.
That input connection is referenced to Power Ground.
Now double that up to a two channel amplifier with two interconnects from Source to amplifier.
The two Signal Grounds are referenced to the single Power Ground.
At the Source the two Signal Grounds are almost certainly connected internally.
Now you have a LOOP.
any interference impinging on that loop creates a current around the loop.
The part of the loop from source to amplifier that carries the signal will now have an interference current on it and with the inherent resistance of the bit of the connection will have an interference voltage that is added to the signal.
Multiply that by 10 inputs and you have dozens of loops all collecting interference. Equals big hum !
The input cable is a two wire connection.
That input connection is referenced to Power Ground.
Now double that up to a two channel amplifier with two interconnects from Source to amplifier.
The two Signal Grounds are referenced to the single Power Ground.
At the Source the two Signal Grounds are almost certainly connected internally.
Now you have a LOOP.
any interference impinging on that loop creates a current around the loop.
The part of the loop from source to amplifier that carries the signal will now have an interference current on it and with the inherent resistance of the bit of the connection will have an interference voltage that is added to the signal.
Multiply that by 10 inputs and you have dozens of loops all collecting interference. Equals big hum !
By the way...
This could be very useful for better understanding my actual situation.
So here is the Ground scheme of my amplifier.
Luke
I haven't perused the entire thread, so this may be redundant..
Your two ground connections at the main transformer complete a very large ground loop.
Any output current to a speaker will raise the input signal shield as a consequence of IR drop. This setup is pretty much a six channel unstable oscillator.
You need to rethink the ground such that the power currents through ground do not impact the input grounds.
jn
Oh yes Michael, believe me!
The very first thing I did after realizing buzzes were still there.
Add to this inclusion of 100 ohms resistor(s) with the tests.
Tested in a lot of different ways...
Oh by the way, you show a 12 channels amp while I only have 6 ch.
Luc
Last edited:
Andrew
Sure you're right andhis is what I concluded in my post #119.
You understand why I consider to try Line transformers and I need six of them.
So Gnd can be floating.
I will try these tomorrow with a pair of these I have to put my hands on.
Luke
Hi JN
Wow...this picture is almost a month old and it's old story. Gone!
The new Gnd config is here at post#117 and it is a star ground with a ½"x½"x5" copper bar.
Luke
Last edited:
You still have to control the currents.
Think of this as a mind experiment. With all the inputs connected to sources, think of yourself as a current.. Travel along one of the input shields...where do you end up? Can you get back to the source equipment via another shield?
For any conductive loops, can a time varying magnetic field go through it? Are any of them signal grounds?
You show a massive star, did you twist the opposing current wires together?
How do you deal with the shield connections at the input jacks? Pins 1, 3 and 5 will be connected to an external common ground, where they form a big loop. If the other side input connector also connects to the source unit ground, you again form a big ground loop using your input shields, but it now climbs entirely through your main amp chassis.
Given the piecemeal schematics of the system I/O, it is rather difficult to follow what your amp is doing.
jn
stage 1.) You can learn a lot about how to ground an amplifier properly by building an uncased single channel on your work-bench.
stage 2.) You learn more when you insert that single channel in a metal chassis.
stage 3.) There is even more learning by extending those hard learned lessons from the two stages above to a metal cased two channel amplifier.
If you miss out stages one, two and three, it is very difficult to jump straight to stage 9 and hope the learning will be obvious in the mess of cables going literally all over the place.
Go back to stage 1 and learn how to build an uncased mono block, properly !
stage 2.) You learn more when you insert that single channel in a metal chassis.
stage 3.) There is even more learning by extending those hard learned lessons from the two stages above to a metal cased two channel amplifier.
If you miss out stages one, two and three, it is very difficult to jump straight to stage 9 and hope the learning will be obvious in the mess of cables going literally all over the place.
Go back to stage 1 and learn how to build an uncased mono block, properly !
Hello
Bingo! The input transformers solved the problem!
So JNeutron was right to confirm my experience.
I just forgot I had a pair of Sowther 3197 600-600 ohms transformer so I tried them.
So I'll check for Jensen or Cinemag and I know I'll end out with much better noise specs as these have double-EShields and offer a better impedance match with my inputs.
I will add a follow-up to finalize this.
Thanks to all for suggestions and help!
Luke
Why are ground issues so f-ing confusing? If there are loops then shouldn't you disconnect an end of the loop to stop the offending current?
In principal, it seems simple. In practice, it can be very complicated. When dealing with multiple inputs as in this example, isolation may be the only practical solution.
jn
Hello
I knew it since the beginning that transformers was the solution.
It's only that in my mind at that time six of these beats would have cost me an arm and a leg.
Just for a matter of budget, as a reference, I first thought of the Neutrik NTL-1 which is a good one...not the best but just a good one and these would have cost me more than $550.
So just imagine the premiums!
After checking prices, finally these were not that expensive as I realize that Neutrik's prices just suffered from money exchanges.
And forget about Sother and Lundhal...expensive and much less bang for the buck.
At the end the American made premiums are far less expensive than I first thought and there are the best.
So I checked for Jensen JT-11P-1 and Cinemag CMLI-15-15B.
And my preference goes definitely for Cinemag's.
Really nice and collaborative people, there.
Luke
I knew it since the beginning that transformers was the solution.
It's only that in my mind at that time six of these beats would have cost me an arm and a leg.
Just for a matter of budget, as a reference, I first thought of the Neutrik NTL-1 which is a good one...not the best but just a good one and these would have cost me more than $550.
So just imagine the premiums!
After checking prices, finally these were not that expensive as I realize that Neutrik's prices just suffered from money exchanges.
And forget about Sother and Lundhal...expensive and much less bang for the buck.
At the end the American made premiums are far less expensive than I first thought and there are the best.
So I checked for Jensen JT-11P-1 and Cinemag CMLI-15-15B.
And my preference goes definitely for Cinemag's.
Really nice and collaborative people, there.
Luke
I do not understood why it is so hard to solve the ground problem.
You are not the first people and for sure not the last one that design and made an amplifier with more than 2 channels but seems that which for others works for you does not work.
All the time there are simple solutions and complex solution, cheap solutions and expensive solutions.
It's your choice!
You are not the first people and for sure not the last one that design and made an amplifier with more than 2 channels but seems that which for others works for you does not work.
All the time there are simple solutions and complex solution, cheap solutions and expensive solutions.
It's your choice!
- Status
- Not open for further replies.