That's what I suspect also. There must be something I'm just not getting. I understand the circuit and have rebuilt it 3 times. Each time attempting to make it neater and cleaner. Same results every time
A couple things that seem weird. Maybe this will give a clue to the problem
1.The 25k volume pot on the opamp circuit has no effect for the first 75% of it's travel. You can not sense a volume change until the last 25%. I thought maybe the wiring was wrong so I tried it in reverse. The pot does have effect throughout but you have to turn it backwards.
2. I built the tone control circuit from the 49720 data sheet. It's a simple bass and treble control. The circuit seams to work but it causes the same problem at the amp.
The thing that the tone control circuit and the gain stage have in common is the power supply. The zener stabilized supply is connected to the GC supply.
First the opamp control. I think you will find that's because you are using a log pot here... and you would have to wire it "backward" for it to operate conventionally with the opamp.
Normally, as the wiper moves toward the ground end of the track the signal reduces on the wiper, in other words normal volume control action. The opamp works in reverse, the gain increasing with decreasing resistance (so more output) as the pot resistance reduces. Linear pot I think here
Sorry, never thought to mention that earlier... and linear ones are usually better matched unless you go for an ALPS etc.
I'll come back to the GC as it's still not easy to fully take in the wiring from a pic as I'm not familiar with the GC pcb's etc.
Normally, as the wiper moves toward the ground end of the track the signal reduces on the wiper, in other words normal volume control action. The opamp works in reverse, the gain increasing with decreasing resistance (so more output) as the pot resistance reduces. Linear pot I think here
Sorry, never thought to mention that earlier... and linear ones are usually better matched unless you go for an ALPS etc.
I'll come back to the GC as it's still not easy to fully take in the wiring from a pic as I'm not familiar with the GC pcb's etc.
I'll come to you in a little while...
In the meantime where have you connected the ground for the preamp... is it to that centre point with all the wires ?
The zener supply gets it's feed from the white and blue wires ?
What happens if no source is connected to the preamp... does it still do strange things then as the volume is turned up ?
If it's OK what happens if you short both preamp inputs to their grounds... as in shorting plugs in the preamp input... is it still OK as the volume turned up ?
Tea time
In the meantime where have you connected the ground for the preamp... is it to that centre point with all the wires ?
The zener supply gets it's feed from the white and blue wires ?
What happens if no source is connected to the preamp... does it still do strange things then as the volume is turned up ?
If it's OK what happens if you short both preamp inputs to their grounds... as in shorting plugs in the preamp input... is it still OK as the volume turned up ?
Tea time
Have to see what the above tests show... I'm struggling to suggest anything at the moment.
What I am trying to determine is if the problem is caused by the GC supplying current into the speakers, and whether it's that current that causes a contamination of some signal ground etc and hence instability.
I know it sounds odd but what happens if you cut the opamp "PSU ground" connection totally. The preamp is still grounded via the signal ground... and then taking it further what happens if you cut ONE of the signal grounds from pre amp to GC.
If you think about it, it should still all be grounded since the preamp uses a common ground for both channels (unlike the GC which is on two PCB's.
If you try the above just measure on ohms after each "cut" to confirm the preamp is still reading as connected to ground in the GC.
What I am trying to determine is if the problem is caused by the GC supplying current into the speakers, and whether it's that current that causes a contamination of some signal ground etc and hence instability.
I know it sounds odd but what happens if you cut the opamp "PSU ground" connection totally. The preamp is still grounded via the signal ground... and then taking it further what happens if you cut ONE of the signal grounds from pre amp to GC.
If you think about it, it should still all be grounded since the preamp uses a common ground for both channels (unlike the GC which is on two PCB's.
If you try the above just measure on ohms after each "cut" to confirm the preamp is still reading as connected to ground in the GC.
I built the zener supply on another board. It looks exactly like you drew it in post 37. 3 wires run from the board to the GC supply.
White to the + 34 v
Blue to the -34 v
black to the star ground in the GC
3 wires run from the outs of zener supply
white +15v
blue -15v
black ground
You can see where they enter the gain stage at the bottom of the photo.
The ground is all the way to the right. It goes through the D/R/D parallel and hooks to the end of the center traces.
White to the + 34 v
Blue to the -34 v
black to the star ground in the GC
3 wires run from the outs of zener supply
white +15v
blue -15v
black ground
You can see where they enter the gain stage at the bottom of the photo.
The ground is all the way to the right. It goes through the D/R/D parallel and hooks to the end of the center traces.
Thank you! I'll be at my workshop later today. I'll run the tests and report my findings.
OK... another thing you could try is temporarily connecting the speaker negatives to the "star" wiring and not the actual GC pcb's. That would eliminate one possible source of contamination.
It's a difficult topic... you have to think of each wire as a resistance... and I find it helps just to think of it as a DC problem for this. Imagine all the wires had say 10 ohms resistance.
Now imagine the speaker with say 1 amp flowing.
Can that one amp develop an unwanted voltage along any track or wire that alters the input or feedback voltages. If the answer is yes there is a problem.
It's a difficult topic... you have to think of each wire as a resistance... and I find it helps just to think of it as a DC problem for this. Imagine all the wires had say 10 ohms resistance.
Now imagine the speaker with say 1 amp flowing.
Can that one amp develop an unwanted voltage along any track or wire that alters the input or feedback voltages. If the answer is yes there is a problem.
Can't remember if I put a link to this before,
http://www.diyaudio.com/forums/solid-state/101321-3-stage-lin-topology-nfb-tappings.html
it illustrates the problems of incorrect grounding. Read all the posts but try and understand the principle of how the grounds can be contaminated and how that can cause disruption to the audio signal.
http://www.diyaudio.com/forums/solid-state/101321-3-stage-lin-topology-nfb-tappings.html
it illustrates the problems of incorrect grounding. Read all the posts but try and understand the principle of how the grounds can be contaminated and how that can cause disruption to the audio signal.
I'll come to you in a little while...
In the meantime where have you connected the ground for the preamp... is it to that centre point with all the wires ?
The zener supply gets it's feed from the white and blue wires ?
What happens if no source is connected to the preamp... does it still do strange things then as the volume is turned up ?
If it's OK what happens if you short both preamp inputs to their grounds... as in shorting plugs in the preamp input... is it still OK as the volume turned up ?
Tea time
With no source connected.
Very loud hum, same behavior from GC. At full volume there's 10 volts at the GC output.
With the inputs shorted.
No hum, same behavior from GC. At full volume 7.6 volts at GC output
That's good...
I'll post back (it may be tomorrow) what I would do. Basically it involves connecting the preamp up in stages to try and find where it's going wrong.
There's one more thing you can try... and it should be silent still.
With the GC inputs shorted as before, and still leaving them shorted connect the grounds of the inputs together with a bit of wire.
This happens anyway when you connect the preamp but lets do it one step at a time.
I'll post back (it may be tomorrow) what I would do. Basically it involves connecting the preamp up in stages to try and find where it's going wrong.
There's one more thing you can try... and it should be silent still.
With the GC inputs shorted as before, and still leaving them shorted connect the grounds of the inputs together with a bit of wire.
This happens anyway when you connect the preamp but lets do it one step at a time.
Something else to try on the preamp. In your picture is the pot fastened to a front panel ? metal ? grounded ? pots can pick up all sort of stray noise.
Might be worth disconnecting the wires on the PCB that go to the pot...that leaves the opamp at a gain of 1. See if it still hums and does odd things.
If I had it in front of me it would be a quick job to run through the checks... explaining them all takes a lot of time... and not easy to explain either exactly what to do.
I would keep the preamp inputs shorted during all this testing... obviously put the GC inputs back to normal.
Might be worth disconnecting the wires on the PCB that go to the pot...that leaves the opamp at a gain of 1. See if it still hums and does odd things.
If I had it in front of me it would be a quick job to run through the checks... explaining them all takes a lot of time... and not easy to explain either exactly what to do.
I would keep the preamp inputs shorted during all this testing... obviously put the GC inputs back to normal.
going back to post114.
That shows a non-inverting amplifier with the gain control in the Negative FeedBack loop.
If that is converted to an inverting amplifier then the gain can be varied to attenuate (-XdB) and amplify (+YdB) over a continuous range.
Disconnect the 100uF from Signal Ground. Input your Signal at the 100uF.
Connect the +IN to Signal Ground using the pads of the Rin resistor. Change this resistor from 470k to ~ 22k to match Rf.
Keeping the NFB resistor values as shown, the range of Gain is from -2dB to +18dB.
If the pot is moved to the 3way junction of Rf, -IN and the 2k7 then a much wider range of volume adjustment becomes available.
Retaining the post114 Non-inverting stage gives a Gain range of +5dB to +19dB.
That shows a non-inverting amplifier with the gain control in the Negative FeedBack loop.
If that is converted to an inverting amplifier then the gain can be varied to attenuate (-XdB) and amplify (+YdB) over a continuous range.
Disconnect the 100uF from Signal Ground. Input your Signal at the 100uF.
Connect the +IN to Signal Ground using the pads of the Rin resistor. Change this resistor from 470k to ~ 22k to match Rf.
Keeping the NFB resistor values as shown, the range of Gain is from -2dB to +18dB.
If the pot is moved to the 3way junction of Rf, -IN and the 2k7 then a much wider range of volume adjustment becomes available.
Retaining the post114 Non-inverting stage gives a Gain range of +5dB to +19dB.
Last edited:
"With the GC inputs shorted as before, and still leaving them shorted connect the grounds of the inputs together with a bit of wire."
Still silent!
"Might be worth disconnecting the wires on the PCB that go to the pot...that leaves the opamp at a gain of 1. See if it still hums and does odd things."
Same bad behavior with pot disconnected.
The pot is mounted to the aluminum chassis.
Andrew, I like the range of gain that your talking about. It would be nice to get some - gain if needed instead of an automatic boost.
Were you suggesting the inverted circuit as a solution to the noise problem?
Still silent!
"Might be worth disconnecting the wires on the PCB that go to the pot...that leaves the opamp at a gain of 1. See if it still hums and does odd things."
Same bad behavior with pot disconnected.
The pot is mounted to the aluminum chassis.
Andrew, I like the range of gain that your talking about. It would be nice to get some - gain if needed instead of an automatic boost.
Were you suggesting the inverted circuit as a solution to the noise problem?
Last edited:
It's just explaining what to do that's the hard bit... and some of the things I suggest might sound odd
Disconnect the power supply to the preamp... the + and - supplies to the zener regulators. So all you have left is the ground connections to the preamp. Still leave the output from the preamp connected to the GC.
Does it still hum and do weird things as the main volume is turned ?
Disconnect the power supply to the preamp... the + and - supplies to the zener regulators. So all you have left is the ground connections to the preamp. Still leave the output from the preamp connected to the GC.
Does it still hum and do weird things as the main volume is turned ?