Hi,
I've built quiet a few 3875. I've found with them and now a TDA7297 that I get massive RF most of the time.
Original the 3875 came with a 220ohm input resistor for anyone experiencing RF.
I found that it only helped a tiny bit, so I omitted it for my history of using the 3875.
However playing with the TDA7297, it's been a huge issue. I have input capacitors for protection, but the RF has been an issue. At first it seemed as though grounding the heat sink did the trick, but somehow it came back.
I realized perhaps I could try the resistor on it.
The interesting thing is that a resistor before the input cap, on one channel, reduced the RF very significantly. This makes me wonder several things. First all these years I just needed to use 1kohm or larger. Right now I'm playing music with a 10kohm but it's eating more than I want. However it's very quiet. Yes I can hear something up close, but that level may never leave given my location. There may be enough parasitic capacitance in the tracings to create some where I live. I may end up using SMT resistors at the pins in the future, but I hate tearing up the signal.
The most curious thing is that both channels are silenced by one resistor, perhaps this means their shared signal ground is part of the parasitic capacitance loop, and therefor perhaps the resistor doesn't have to go in signal. I'm still playing around, but I thought this might be of fascination to those that have struggled with it as much as I have.
*Actually I may be entirely wrong, it's not Cp. It may be Lm. The only thing that works is creating resistance on the input line, which would force more RF to be grounded and not inducted. It's like trying to force a shield to work.
I've built quiet a few 3875. I've found with them and now a TDA7297 that I get massive RF most of the time.
Original the 3875 came with a 220ohm input resistor for anyone experiencing RF.
I found that it only helped a tiny bit, so I omitted it for my history of using the 3875.
However playing with the TDA7297, it's been a huge issue. I have input capacitors for protection, but the RF has been an issue. At first it seemed as though grounding the heat sink did the trick, but somehow it came back.
I realized perhaps I could try the resistor on it.
The interesting thing is that a resistor before the input cap, on one channel, reduced the RF very significantly. This makes me wonder several things. First all these years I just needed to use 1kohm or larger. Right now I'm playing music with a 10kohm but it's eating more than I want. However it's very quiet. Yes I can hear something up close, but that level may never leave given my location. There may be enough parasitic capacitance in the tracings to create some where I live. I may end up using SMT resistors at the pins in the future, but I hate tearing up the signal.
The most curious thing is that both channels are silenced by one resistor, perhaps this means their shared signal ground is part of the parasitic capacitance loop, and therefor perhaps the resistor doesn't have to go in signal. I'm still playing around, but I thought this might be of fascination to those that have struggled with it as much as I have.
*Actually I may be entirely wrong, it's not Cp. It may be Lm. The only thing that works is creating resistance on the input line, which would force more RF to be grounded and not inducted. It's like trying to force a shield to work.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.