For those interested, although these products and threads are old, I picked a RM6750 subwoofer from junk store - it looked like new and worked - so I was wondering why it was sent there... It had an annoying AC hum when powered - even when nothing was connected to it and volume knobs were at minimum volume. While it might not be a problem at high volumes, it was quite audible in silence and it is not acceptable in HiFi setups. I guess that its owner was annoyed and looked for other device...
It appeared that all Polk powered subwoofers share the same power amplifier board - shown below - and this board has a design defect which results in ~10mV 120Hz AC hum.
The cause of it is incorrect board layout in the power section - there is a ground loop between common (zeroV) point of power capacitors and the ground points in the circuit (black wire to speaker).
The ground is connected by a thin wire link (JP2?) shown in the picture, and a second link to input amplifier (TDA7294 or SK...) common net and the chassis ground. As the result, ripple voltage drop on capacitor connection is applied to the input of the power amplifier. This also will result in the additional distortion in the amplifier, but distortion in subwoofers is not clearly audible (driver speaker has higher distortion anyway). This is a common mistake of un-experienced electronics engineers.
There is an easy fix in two steps (someone may find that first step is sufficient).
= First, solder a thick (~16-14AWG) stranded link between the common connection between the large 4700uF capacitors and the point where black speaker wire is soldered. The wire is on the green side of PCB. Ideally, it can be from the point of a pin of the capacitor with "QC PASS" label, and the wire must be as short and as straight as possible. This wire is in parallel with JP2? and reduces the resistance of this link. This will bring the hum down by ~16dB.
= Second, cut the wide trace from the common connection of capacitors to ground bracket and to other components - this trace goes from lower capacitor to the lower edge of the board. Solder a new wire (use ~22-26 AWG wire) from this bracket to the point where black wire is soldered.
What we want to achieve is connecting the common if filter capacitors to common ground where negative feedback of TDA7294 (black wire) first, and then have this new common ground point be routed to other noise-sensitive nodes in a star-like topology.
Unfortunately, I closed the box and did not take a picture of rework. After the modifications, the hum is not heard anymore, and I can use this box in HiFi system. I am marking it on the picture I found on Internet.
On the side note: The subwoofer consumes ~7 watts when it is in standby mode (LED is red). It is not great for your budget. It is desirable to switch off the AC power with common switch or connect it to a switched power outlet on the receiver to save power.
It appeared that all Polk powered subwoofers share the same power amplifier board - shown below - and this board has a design defect which results in ~10mV 120Hz AC hum.
The cause of it is incorrect board layout in the power section - there is a ground loop between common (zeroV) point of power capacitors and the ground points in the circuit (black wire to speaker).
The ground is connected by a thin wire link (JP2?) shown in the picture, and a second link to input amplifier (TDA7294 or SK...) common net and the chassis ground. As the result, ripple voltage drop on capacitor connection is applied to the input of the power amplifier. This also will result in the additional distortion in the amplifier, but distortion in subwoofers is not clearly audible (driver speaker has higher distortion anyway). This is a common mistake of un-experienced electronics engineers.
There is an easy fix in two steps (someone may find that first step is sufficient).
= First, solder a thick (~16-14AWG) stranded link between the common connection between the large 4700uF capacitors and the point where black speaker wire is soldered. The wire is on the green side of PCB. Ideally, it can be from the point of a pin of the capacitor with "QC PASS" label, and the wire must be as short and as straight as possible. This wire is in parallel with JP2? and reduces the resistance of this link. This will bring the hum down by ~16dB.
= Second, cut the wide trace from the common connection of capacitors to ground bracket and to other components - this trace goes from lower capacitor to the lower edge of the board. Solder a new wire (use ~22-26 AWG wire) from this bracket to the point where black wire is soldered.
What we want to achieve is connecting the common if filter capacitors to common ground where negative feedback of TDA7294 (black wire) first, and then have this new common ground point be routed to other noise-sensitive nodes in a star-like topology.
Unfortunately, I closed the box and did not take a picture of rework. After the modifications, the hum is not heard anymore, and I can use this box in HiFi system. I am marking it on the picture I found on Internet.
On the side note: The subwoofer consumes ~7 watts when it is in standby mode (LED is red). It is not great for your budget. It is desirable to switch off the AC power with common switch or connect it to a switched power outlet on the receiver to save power.
