Automatic biasing of output stage - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th June 2002, 04:43 AM   #1
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
Default Automatic biasing of output stage

I was going to add this to temperature compensating of o/p stage thread, but there temp is the means of altering the bias. This has nothing directly to do with temp except that hopefully it negates it's effects.

The circuit here has not been built, and it is only a concept diagram really. When the input voltage swings upward the rhs opamp senses the difference between input and output voltage because the upper fet needs several volts on it's gate before it starts conducting. The rhs opamp via the upper diode applies an offset voltage to the upper cap sufficient to make up for the dc offset between signal voltage and output voltage. Same for the lower half cycle.

Probably the caps should have high value bleed resistors so they will discharge over 10-20 seconds. Zero standby dissipation! One problem I can forsee is that if the output stage clips then the fets will both be driven fully on together

It's just an idea that might benefit from other people's input....
Attached Images
File Type: gif auto bias comp symm.gif (5.1 KB, 557 views)
  Reply With Quote
Old 13th June 2002, 08:22 AM   #2
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
I think part of my brain wasn't switched on before. Here is a much more workable setup. Now if only I liked P-channel fets... yuk

GP.
Attached Images
File Type: gif auto bias comp symm 2 opamp.gif (5.9 KB, 508 views)
  Reply With Quote
Old 13th June 2002, 08:55 AM   #3
Banned
 
Join Date: Dec 2001
Location: Zamboanga, City of Flowers, Mindanao
Send a message via Yahoo to Elso Kwak
Lightbulb Biasing MOSFET Outputstage

Hi Circlotron,
I believe Linear Technology has a dedicated IC for this application:
LT1166
http://www.linear.com/prod/datasheet.html?datasheet=113
  Reply With Quote
Old 13th June 2002, 09:29 AM   #4
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
Circlotron, I think you are in the wrong track here. The basic idea is the create some sort of control loop for the minimum current (gate voltage). The gate voltage must increase with higher output current. Your idea works only for a class A stage. If you check the idea behind LT1166, you see the control loop around current shunts.

The voltage at gate-source is the audio signal plus a DC-level added (if you want the transistor to conduct all the time).

Much can be done here, but you must have crazy ideas in order to filter out the good ones. You're not alone here, I have had the same ideas myself.
__________________
/Per-Anders (my first name) or P-A as my friends call me
Tube Buffered Gainclone in work |Thread || Diamond buffer |Thread for the group buy | Wiki
  Reply With Quote
Old 13th June 2002, 09:34 AM   #5
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
When I think about it, I have papers somewhere about auto-biasing amps. Check International Rectifier (Harris?) for application notes in the subject. I remember though the distortion figures wasn't very extreme, rather low actually.
__________________
/Per-Anders (my first name) or P-A as my friends call me
Tube Buffered Gainclone in work |Thread || Diamond buffer |Thread for the group buy | Wiki
  Reply With Quote
Old 13th June 2002, 01:07 PM   #6
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
Elso, I discovered LT1166 just a few days ago. Thanks, though.

Peranders, the way the cct is supposed to work is the input goes for example a little positive but the output does not yet follow because of the gate threshold voltage, so the opamp sees the difference between input voltage and output (source) voltage and starts putting a charge on the cap more and more and because the cap is in series with the gate signal it gets a dc offset added until the gate voltage is high enough that the source voltage finally begins to follow the gate. Almost, because of the resistive voltage divider on the opamp. When the source voltage catches up to the gate (almost) the opamp output drops away until the cap may need an additional charge. I think too that the caps should have a bleeder resistor of say 1M and the caps about ~47uF.
  Reply With Quote
Old 14th June 2002, 04:55 AM   #7
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
Righteo then. This is the latest incarnation of me thinking out aloud. Here is the same principle applied to a circlotron topology, with a few of the time constants stated. The operating principle, putting it a different way, is a peak detector for how much the source voltage is below the signal voltage, and it adds this difference voltage to the 10uF cap, so now the gate voltage is higher than the signal voltage (dc offset) by the amount that the source voltage lags behind the signal voltage.

e.g Vsig=10v, Vgate=10v, Vsource=6v (4v gate threshold)
Opamp says "whoa! Vsig > Vsource". It starts pulling up gate via 100k and diode and charging 10uF cap until Vsource=Vsig So now Vsig=10v, Vgate=14v, Vsource=10v (still 4v gate threshold but it appears across 10uF cap and so is effectively gone).

10M bleed resistors across 10uF caps allow caps to slowly discharge unless bumped up from time to time by opamp. When the music stops the caps slowly discharge down to zero and the mosfets gradually turn off completely, making the power drain *very* low. This part was unintended so it is pure serendipity.

GP.
Attached Images
File Type: gif auto bias circlotron.gif (31.0 KB, 401 views)
  Reply With Quote
Old 14th June 2002, 06:31 AM   #8
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default auto biasing

Circlotron,

The last part of your previous post raised my suspicion! I'm afraid you have introduced level dependent bias. As you say, without (music) signal there is no bias. But the bias should be there with or without music. You don't want to have strong xover distortion at low level, right.

So, the idea is OK, but you should look at the DC current in the o/p stage for the information to drive your opamp loops.

Cheers, Jan Didden
  Reply With Quote
Old 14th June 2002, 01:13 PM   #9
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
The opamp only works as a comparator. It charges the cap until the gate threshold voltage is reached and then stops. It is not level sensitive except for when the drive signal is below several millivolts or whatever the offset voltage of the opamp is. Seeing that would be the same level fed to the loudspeaker, that's not very loud!

Regardless of the signal level, the circuit always adds the same amount of dc offset to make the fet source voltage equal the instantaneous signal voltage. The dc offset will reduce as the fet heats up of course because the threshold voltage reduces. Actually it also means you could use unmatched fets because it would bias each one to whatever it needed.

Anyway, I am building it up tonight (Friday) out in the shed, so provided I don't set myself on fire or something, I will let you know how it goes. Good or bad, it doesn't matter. If it doesn't work then I'll learn something.

Actually I'm quite willing to concede that this setup, and hexfets too for that matter, may not be the ultimate solution for hifi. I think though that it / they could be a good thing for high power public address systems, able to drive a 70v or even 100v line directly. Given the size of mosfet modules you can buy nowadays, it may not make the best racecar or limousine, but a mighty fine truck. Anyway, enough yackety-yak. Out to the shed and see if my circuit is good or nonsense.

GP.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Input stage biasing - BJT steve_mak Solid State 24 11th July 2007 03:38 AM
Output stage biasing, with the driver stage andrew_whitham Tubes / Valves 0 19th May 2007 11:14 AM
Output Tube biasing for Toroidal Output Transformer gingertube Tubes / Valves 7 17th April 2005 04:40 PM
Biasing output stage FETs PeteMcK Solid State 7 7th February 2002 09:37 PM


New To Site? Need Help?

All times are GMT. The time now is 04:17 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