DC Offset on Class A headphone amp.

[Dxvideo]

Dear x-pro,
I've tried the series 2,2uF cap after the op-amp. But the -4v offset still exist on output and cannot adjustable with the trimpot! The trimpot cannot handle even the gate voltage. The midpoint (trimpots upper and lower pins) voltage is about 1v to GND and couldnt adjust even this voltage! And the more interesting thing is the circuit stil works fine!!!!
What now?

Dear ilimzn,
I made this circuit on a test board. And I afraid I have to remove all components before adding a DC feedback. So it will go to the trashbin if so.

Conclusion;
Another fiasco!

Thx all.

[x-pro]

Originally posted by Dxvideo
Dear x-pro,
I've tried the series 2,2uF cap after the op-amp. But the -4v offset still exist on output and cannot adjustable with the trimpot! The trimpot cannot handle even the gate voltage. The midpoint (trimpots upper and lower pins) voltage is about 1v to GND and couldnt adjust even this voltage! And the more interesting thing is the circuit stil works fine!!!!
What now?

As far as I understand you are using +/-12V supply. In this case you need to change R3 for 62K and R4 for 160K to be able to adjust the gate voltage between approximately 4 and 6 V so you will be able to set the output voltage to 0V.

Cheers

Alex

[Dxvideo]

But ilimzn mentioned; there are several mistakes in the circuit not only the coupling...

Anyway, if the circuit works with an output capacitor then will work without it.
So if I can balance the output DC offset with changing the resistor values then I will have a real Class A amplifier.
Are these values exact? Or may I accord them to E12 series? May be 56K and a 20K trimpot and a 150K?
Is it possible?
Thx..

[x-pro]

Originally posted by Dxvideo
But ilimzn mentioned; there are several mistakes in the circuit not only the coupling...

There is no such thing as an absolutely perfect circuit . I would add a 100K resistor from a non-inverting input of the opamp to the ground just in case the volume pot loses contact, but otherwise this circuit should work reasonably well providing you have enough heatsinking on the MOSFET and voltage regulators. It also would be sensible to add 22-47 Ohm resistor in series with the output to protect low impedance headphones in case of an overload 9thought it would impair the sound quality somewhat). The output offset would need to be accurately set after a warm-up as it would change with temperature.

Originally posted by Dxvideo

Are these values exact? Or may I accord them to E12 series? May be 56K and a 20K trimpot and a 150K?
Is it possible?

Yes, 150K and 56K should work just fine.

Cheers

Alex

[Dxvideo]

The output offset would need to be accurately set after a warm-up as it would change with temperature.

Yea I know that from my LM4702 amplifier with lateral MOSFETs. Until the MOSFETs warm up the 60-70° the bias current increase to the 110mA. And when they stop at the 70° then BIAS current stops at the 115mA. And when I try to adjust bias at lower temperatures and it goes upper and upper when it get warmer.

Anyway, I will try that resistor values and will report after the successfull try.

Best regards,

[Dxvideo]

Dear Alex,

I dont know the reason but I cannot adjust the bias voltage!
I've tried 150K+20K Trimpot+56K, 100K+20K Tr.+10K even 100K+20K+1K and without zener. However I couldnt get over 1,6v offset at gate.
At 100K-1K combination I can get only 1,6 v not more!
I am really confused. Whats the G-S resistance of that MOSFET? I suspicied that MOSFET sinks a huge current from gate. Then I cannot stabilise the mid point of voltage..
What do you say?
Thx again.

[Nordic]

Break the circuit up in two parts...

Lets look at the opamp part....
I would run this off a single rail of 24V or so...
with the output relative to 1/2VS i.e. about 12VDC, this is superimposed on top of the signal.... but you don't hear DC...

So you don't need a cap between the two stages... (but I don't like your circuit), You do need a sizeable cap (even 10000uf) on the output though or your bass will sufer drasticaly, the lower impendance your headset is.

Look at this valve hybrid circuit to get an idea... of how you could go about things...you know steal some ideas... the opamp on a single rail supply will funtion similarly to the valve there...

The Grid need to be a minumum number of volts above the Source, I think about 4V for that Fet, before it will conduct between the Drain and Source. The strength varies with the current presented to the grid... in the form of your signal.

So you just have to find the right biasing point .

[cliffforrest]

Originally posted by Dxvideo
Dear Alex,

I dont know the reason but I cannot adjust the bias voltage!
I've tried 150K+20K Trimpot+56K, 100K+20K Tr.+10K even 100K+20K+1K and without zener. However I couldnt get over 1,6v offset at gate.
At 100K-1K combination I can get only 1,6 v not more!
I am really confused. Whats the G-S resistance of that MOSFET? I suspicied that MOSFET sinks a huge current from gate. Then I cannot stabilise the mid point of voltage..
What do you say?
Thx again.

A mosfet's gate is isolated - it CANNOT conduct current in any circumstances! (unless it is punctured and destroyed.

[Dxvideo]

Dear Nordic,

You mean, should I keep my circuit at first situation?

[Christer]

What about just doing like this?