diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Headphone Systems (http://www.diyaudio.com/forums/headphone-systems/)
-   -   DC Offset on Class A headphone amp. (http://www.diyaudio.com/forums/headphone-systems/104767-dc-offset-class-headphone-amp.html)

Dxvideo 6th July 2007 02:03 PM

DC Offset on Class A headphone amp.
 
Hi all,

I made a Class A headphone amp last week. Its based from originally headwize forum "A HeadWize Design Series Paper" article. "Class A MOSFET Follower" design.
http://img504.imageshack.us/img504/7674/headampt0.jpg
I use OPA134 op-amp for the buffer and it has 2 gain.
When first try, I faced there was -4 v offset on outputs (there was no output capacitor) And I couldnt fix that problem with adjusting the offset trimmer. Then I decided to put a 1500uF cap on output. However its a bit disturbing. Because the original design has no output capacitor. (There is no offset on op-amps output)
Anybody has any idea?
Thanks a lot for replies in advance.
Best regards..

Juergen Knoop 6th July 2007 02:13 PM

thats Vgs! :smash:

x-pro 6th July 2007 02:30 PM

2 Attachment(s)
Could you provide the link to the original schematics? The circuit you've posted here should produce about -4V on the output as the o/p voltage of the opamp is near 0V and a MOSFET needs 4V to open. I suspect that in the original circuit there was a capacitor between the opamp and the MOSFET. On top of this the polarity of the output electrolytics on your diagram is wrong.

Cheers

Alex

P.S. - I've attached a modded picture to show where the cap should go. I suggest about 1uF film would do.

Dxvideo 6th July 2007 02:42 PM

This is the original designs link, pls find it at the Class A MOSFET Follower title.
http://headwize.com/projects/showfil...=opamp_prj.htm
And the original circuit is here;
http://img516.imageshack.us/img516/7697/headflowug5.jpg
However I couldnt see any cap between op-amp and MOS!

x-pro 6th July 2007 02:49 PM

Quote:

Originally posted by Dxvideo
However I couldnt see any cap between op-amp and MOS!
It is not a complete circuit and so it is confusing and wrong on that particular picture. If there is no capacitor, the whole network of resistors to set the gate voltage of the MOSFET is completely useless.

Cheers

Alex

P.S. I have to add that that particular article is misleading in many places and some of the circuits shown are plain wrong and some are even dangerous to implement - i.e. "High Voltage Regulator"on fig. 20 .

Dxvideo 6th July 2007 02:56 PM

Ok then.
You say, I MUST put a series 1uF capacitor between op-amp output and MOSFET s gate. The output capacitor (1500uF) will be ommited in this case. And as I can understand it must be before 100ohm and trimpot node.
Am I right?
THX.

x-pro 6th July 2007 03:01 PM

Quote:

Originally posted by Dxvideo
Ok then.
You say, I MUST put a series 1uF capacitor between op-amp output and MOSFET s gate. The output capacitor (1500uF) will be ommited in this case. And as I can understand it must be before 100ohm and trimpot node.
Am I right?
THX.

Yes. You also may need to change the value of R3 - as the DC voltage on the gate should be about +4 +5V. Value would depend on the power supply voltage you are using. It would be essential to use a stabilized supply for this circuit - otherwise the output voltage would vary too much following the supply variations

Cheers

Alex

Dxvideo 6th July 2007 03:11 PM

I see.
I can adjust the gate voltage via R5 (theorically) after putting the DC blocking cap.
My power supply is a regulated one. I use 7812 and 7912 for regulating.
But thats interesting; in THIS situation, circuit works fine, except I feel some treble weakness.
And a final question;
May my op-amp have any defect from this happen?
Thz again.

x-pro 6th July 2007 03:18 PM

Quote:

Originally posted by Dxvideo
And a final question;
May my op-amp have any defect from this happen?


No, I don't think that opamp would suffer any damage.

Cheers

Alex

ilimzn 6th July 2007 03:25 PM

The schematic is wrong on many levels (some more serious than others). Perhaps the best solution would be to make a separate DC feedback from the output.


All times are GMT. The time now is 03:22 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2