I have hum and DC-problems
I have an integrated amp with a J-fet input, DC-config, (no caps.). On that input I have a 47k Pot so I can regulate the volume.
All the signal earth is star config included the power supply earth. The chassi earth is not connected to the signal earth.
Everything sound good until the POT gets over ¾ of, the sound is higher. Then I have this hum and a DC-offset on the output.
If I don’t have any source cables connected to input everything works find even if I have the Pot at max volume.
When I have the hum and put my hand on the earth point and the chassie at the same time then the hum disappear. If I try some electrical connection in the same way, I don’t hold in the amp nothing helps.
Perhaps I need some filter components in the input config., help? Or some filter between the earth and chassi earth…?
Please can someone help me?
I don’t want to stay there holding my hand to make it work and listing at the same time.
Se picture:
I have an integrated amp with a J-fet input, DC-config, (no caps.). On that input I have a 47k Pot so I can regulate the volume.
All the signal earth is star config included the power supply earth. The chassi earth is not connected to the signal earth.
Everything sound good until the POT gets over ¾ of, the sound is higher. Then I have this hum and a DC-offset on the output.
If I don’t have any source cables connected to input everything works find even if I have the Pot at max volume.
When I have the hum and put my hand on the earth point and the chassie at the same time then the hum disappear. If I try some electrical connection in the same way, I don’t hold in the amp nothing helps.
Perhaps I need some filter components in the input config., help? Or some filter between the earth and chassi earth…?
Please can someone help me?
I don’t want to stay there holding my hand to make it work and listing at the same time.
Se picture:
Attachments
It seem like the input of your amp could use a resistor to Gnd.
Try connecting a 10K resistor between the lever pin of the pot and Gnd. If it works, you can try a higher value resistor.
Try connecting a 10K resistor between the lever pin of the pot and Gnd. If it works, you can try a higher value resistor.
Thanks I'm going to try it out.
I have tested a while now and I had a good effect when I connect the source gnd on the input side of the amp, before the Pot with a RC filter 0,1 uf parallel with 15ohm to chassie. However when this is a dual mono amp I don’t like to connect the gnd for both channels to a common spot… what do you say about it?
I have tested a while now and I had a good effect when I connect the source gnd on the input side of the amp, before the Pot with a RC filter 0,1 uf parallel with 15ohm to chassie. However when this is a dual mono amp I don’t like to connect the gnd for both channels to a common spot… what do you say about it?
Frankly I don't like it.....
Gnd is the reference point, and should be altered...
Differences in Gnd gives differences in signal..........
Gnd is the reference point, and should be altered...
Differences in Gnd gives differences in signal..........
OK then we have the same thoughts about it. I’m going to test your suggestion.
Any any other suggestions?
Any any other suggestions?
At this moment I have no other suggestions, but inserting a 10K resistor as told before, or alternative in series with pot (either on top or from the pot lever to the input of the amp.
It all depends of what causing the problem and how your source out impedance apperently mismatch the amp input impedance...
It all depends of what causing the problem and how your source out impedance apperently mismatch the amp input impedance...
Keep trying...
It seems like your j-fet input doesn't like low impedance input (when pot lever is 3/4 up and more)...
Do you have a schematic of the input of the amp?
It seems like your j-fet input doesn't like low impedance input (when pot lever is 3/4 up and more)...
Do you have a schematic of the input of the amp?
Progg70,
My suggestion might not be very practical, but I would suggest trying it, seeing that your setup does seem to behave somewhat strangely.
The symptoms you have are sometimes the result of feedback instability in the form of oscillation on the output. This can be (usually is) high frequency (above audio), but overloads something and can typically result in such symptoms. The solution is to monitor the output with an oscilloscope. This may be a problem for you, depending on where you are and what is available. A radio repair shop might help you even if you have to pay for the small task.
How to cure that is difficult to suggest without at least a circuit diagram. In your case it might help to insert a resistor from the volume control slider in series with the input, say about 10K. If you need much more to cure the problem, the solution could involve other things. I suggest this quite from the dark; if you could post a circuit diagram further ideas might occur, but very difficult as the problem might also simply lie in the physical construction.
Also, rereading your description, you say the chassis earth is not connected to the signal earth. You do not mean that the chassis is floating? ...or is it connected to mains earth only? I am wondering what exactly the chassis (which I presume is metal) "sees" regarding the circuit itself. It should have some reference to signal common (earth) in terms of the capacitances it offers to the circuit at various points. Perhaps some description of the construction can help us helping you.
Hope this helps.
My suggestion might not be very practical, but I would suggest trying it, seeing that your setup does seem to behave somewhat strangely.
The symptoms you have are sometimes the result of feedback instability in the form of oscillation on the output. This can be (usually is) high frequency (above audio), but overloads something and can typically result in such symptoms. The solution is to monitor the output with an oscilloscope. This may be a problem for you, depending on where you are and what is available. A radio repair shop might help you even if you have to pay for the small task.
How to cure that is difficult to suggest without at least a circuit diagram. In your case it might help to insert a resistor from the volume control slider in series with the input, say about 10K. If you need much more to cure the problem, the solution could involve other things. I suggest this quite from the dark; if you could post a circuit diagram further ideas might occur, but very difficult as the problem might also simply lie in the physical construction.
Also, rereading your description, you say the chassis earth is not connected to the signal earth. You do not mean that the chassis is floating? ...or is it connected to mains earth only? I am wondering what exactly the chassis (which I presume is metal) "sees" regarding the circuit itself. It should have some reference to signal common (earth) in terms of the capacitances it offers to the circuit at various points. Perhaps some description of the construction can help us helping you.
Hope this helps.
Hi
I agree with Johan, sounds like instability related to the input filter/impeadance/fb loop. The combination of the pot >3/4 with other circuit components, including the singnal source, is creating an occilation, probably asymetric, that is showing as DC at the speaker terminal. You should be able to see it with a scope. The small cap to signal GND from the wiper is a good idea to start. A smaller pot is another idea. Is the output impeadance of the CD player 47K? Sounds a little high, maybe 5K? Or could be ground loops.
I agree with Johan, sounds like instability related to the input filter/impeadance/fb loop. The combination of the pot >3/4 with other circuit components, including the singnal source, is creating an occilation, probably asymetric, that is showing as DC at the speaker terminal. You should be able to see it with a scope. The small cap to signal GND from the wiper is a good idea to start. A smaller pot is another idea. Is the output impeadance of the CD player 47K? Sounds a little high, maybe 5K? Or could be ground loops.
- Status
- Not open for further replies.