I only did for one channel and after a while the other channel without choke gave out. It was fine after removing the choke. Will wait until I get the second choke to give my final verdict.
I guess I'm not one to speak, I only make single prototypes of my amps, so I haven't listened to good stereo in literally years! 😱
It had to be related to having a choke on only one of the channels but not the other. I am not sure it was the amp that gave out. Could have been the minidsp 4x10HD. I will wait to plug it in again until I have a stereo setup.
Here is a pic if you want to check my connections on the choke and the 3K9 resistor. The choke is called EPCOS - B82724J2142N1 - CHOKE, COMMON MODE, 2X27MH, 1.4A.
Datasheet: http://www.mouser.com/ds/2/400/b82724a_j-525424.pdf
The top par of pins on the choke goes to RCA ground on either connectors. The 3K9 resistor is connected to ground directly without a choke. The lower pairs of pins on the choke that is closest to the bottom of the box is for signal and source. Correct?
Datasheet: http://www.mouser.com/ds/2/400/b82724a_j-525424.pdf
The top par of pins on the choke goes to RCA ground on either connectors. The 3K9 resistor is connected to ground directly without a choke. The lower pairs of pins on the choke that is closest to the bottom of the box is for signal and source. Correct?
Attachments
Last edited:
I heard no difference in quality, but it was more quite without the computer RF noise. Just a little hiss, like the constant static from the power grid. Norway is known for noisy power but that is not so much of a problem. I must test it with both channels first, the other choke is ordered and will be here in a week.
To avoid adding a new thread I will tack my question on here:
I am building a stereo IRS2092 based amplifier and am trying to figure out the grounding. Here is my summary and let me know if it seems reasonable:
AC input ground - Connected to chasis of enclosure, this is the origin of my grounding point.
Meanwell 2 x 48VDC power supplies, I am wiring these in series for +/-48VDC (approved by manufacturer) so I connect V- of one supply and V+ of the other to ground along with the enclosure ground for each supply.
Speaker outputs I am going to ground the negative output of one amplifier and the positive of the other (with the input wired in reverse for this one). Doing this to allow bridging.
RCA inputs - I am planning to float these.
Microcontroller (Ardunio Uno) Including measurement circuits - Ground.
Does this seem reasonable?
I am building a stereo IRS2092 based amplifier and am trying to figure out the grounding. Here is my summary and let me know if it seems reasonable:
AC input ground - Connected to chasis of enclosure, this is the origin of my grounding point.
Meanwell 2 x 48VDC power supplies, I am wiring these in series for +/-48VDC (approved by manufacturer) so I connect V- of one supply and V+ of the other to ground along with the enclosure ground for each supply.
Speaker outputs I am going to ground the negative output of one amplifier and the positive of the other (with the input wired in reverse for this one). Doing this to allow bridging.
RCA inputs - I am planning to float these.
Microcontroller (Ardunio Uno) Including measurement circuits - Ground.
Does this seem reasonable?
Last edited:
The PE to chassis is correct. Beyond that, your Amp and PSU should have only one connection to chassis. Connect the 0V (at the amp or PSU) to the chassis and isolate inputs and outputs from the chassis.
Last edited:
Fairly sure that this is going to be a problem. Most amp designs have the input and output grounds connected to the ground/0V. If you connect the live output of one channel to ground/0V you will create a short.
Specifically, you need the signal input to go to one amp's input, and also to a gain-of-1 inverting opamp whose output goes to the other amp. Then you will have a bridged output between the two amps' "hot" outputs.
For bridged mode you do the phase inversion at the inputs to the amps, not at the outputs.
Ground loops can be formed from more than the earth ground loop...
If you have two separate (left and right) line level interconnects that are not very, very close to each other all the way from one piece of equipment to the other, their ground shields can and will form a loop. Magnetic fields induce currents into this loop. If this is the case, I would try putting some small resistance (e.g. 4.7R) into one or both ground paths and moving the two cables as close together as possible.
Invert one channel and you can use the amp as mono bridged or two channel with the speaker terminals of the inverted channel reversed.
This is HBRL & HBRR in the D.Joffe paper.
But you must put the added resistance in the correct line.
The added resistance REDUCES the interference CURRENT in the loop.
For the critical Signal Return lead (coax screen) the interference current times the screen+connector resistance/impedance gives the interference voltage.
If one adds the resistance to that "screen+connector resistance" then one makes the interference voltage worse, instead of less.
- Status
- Not open for further replies.