Classic-50mW - diyAudio
Go Back   Home > Forums > Amplifiers > Headphone Systems

Headphone Systems Everything to do with Headphones

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 24th October 2014, 01:36 AM   #1
diyAudio Member
 
jerluwoo's Avatar
 
Join Date: Feb 2006
Location: Ashland,Ky U.S
Default Classic-50mW

I would like to share a headphone amp design that I have been working on for the past few weeks. Since I only recently had a need to use hp's this is my first go at a design for them. Since you can basically throw a rock anywhere on the net and find opamp designs, I wanted something discrete and a topology rarely seen in today's designs. After digging through tons of old schematics, I kept seeing this topology used in low level gain stages starting all the way back to mid 50's when transistors first became available and then basically disappearing after the early 70's. Douglas Self called it a 2-transistor shunt-feedback amplifier Discrete design: More on 2-transistor shunt-feedback amplifiers So after some simulating, prototyping, tweeking, and some probably not so accurate measuring, I found it to have very good performance. This design of course won't be for everyone (eek look at all those capacitors) but I find it enjoyable to listen to for hours. It's simplicity to build also made it fun to tweek and experiment. So anyway I give you what I am calling the Classic-50mW, the classic part being obvious and 50mW since this is the power it can give into 32-600 ohm hp's.
Attached is schematic, frequency sims (open and closed loop), and in the zip is a .htm file generated by RMAA6 containing measurements of soundcard loopback alone compared with the amplifier loaded with 32 ohm inserted into the loopback. It's rather obvious the soundcard is the real limiting factor here.
Attached Images
File Type: jpg Classic-50mW.jpg (59.7 KB, 477 views)
File Type: jpg Classic-50mW openloop gain.jpg (74.9 KB, 458 views)
File Type: jpg Classic-50mW closedloop gain.jpg (67.0 KB, 403 views)
Attached Files
File Type: zip Comparison.zip (917 Bytes, 18 views)

Last edited by jerluwoo; 24th October 2014 at 01:38 AM.
  Reply With Quote
Old 24th October 2014, 08:04 AM   #2
diyAudio Member
 
Join Date: Jan 2012
With respect the schematic is a series feedback design , the feedback going to the emitter of Q1 and so raising the input impedance . R7 just supplies the bias current for Q1 as it is decoupled of AC by C1
  Reply With Quote
Old 28th October 2014, 04:28 AM   #3
diyAudio Member
 
jerluwoo's Avatar
 
Join Date: Feb 2006
Location: Ashland,Ky U.S
Oops looks like I messed up on the Comparison archive and left out the graphs.
Attached Files
File Type: zip Comparison.zip (31.8 KB, 16 views)
  Reply With Quote
Old 28th October 2014, 06:35 AM   #4
narmex is offline narmex  United States
diyAudio Member
 
Join Date: Oct 2011
Quote:
Originally Posted by jerluwoo View Post
Oops looks like I messed up on the Comparison archive and left out the graphs.
thank you jerluwoo for the additional information on this design!
how would you describe it's sound?

i'm trying to adapt it to 12v supply, that's the only PS i have right now, and also ideally have it driving a small speaker for desktop use, hope you wouldn't mind.

the various feedback paths get everything interconnected, and i'm have a hard time getting it all working together, the THD i get now is way worse than yours, around 0.74% as it is now.

any general advices on how to adapt it to 12v supply?
what kind of collector current should i shoot for for each transistor?

also i'm curious about how one design and simulate speed, could you give a brief lecture on that? thank you!

Last edited by narmex; 28th October 2014 at 06:37 AM.
  Reply With Quote
Old 28th October 2014, 07:53 AM   #5
diyAudio Member
 
Join Date: May 2013
Hi. Is the input a 10k pot??
  Reply With Quote
Old 28th October 2014, 09:38 AM   #6
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Quote:
Originally Posted by narmex View Post
the various feedback paths get everything interconnected, and i'm have a hard time getting it all working together, the THD i get now is way worse than yours, around 0.74% as it is now.
Are you using the above circuit or the one from here?

TBH that's the kind of result I'd expect for a 4-transistor AB amp driving a speaker at any kind of power. If that's an unloaded result, however, chances are your operating point is off. The output probably isn't anywhere close to half supply.

Even if you do get the operating point right, you can still expect to see appreciably worse distortion at a 12 V supply vs. 32 V. Either you leave the resistors as-is, which reduces standing currents significantly, or you decrease them as needed, in which case they make a worse current source approximation, and you may need to employ "real" current sources instead. This circuit also has a reasonably high output impedance pre-feedback due to the bootstrap, so the loop gain gained by that is drained immediately if you connect a low-impedance load (hence also why the speaker version is not bootstrapped, I presume - it would require one more current gain stage).

But yes, these circuits with multiple DC feedback paths tend to be real head-scratchers. Part values are typically optimized iteratively, going from desired voltages and device currents.
  Reply With Quote
Old 28th October 2014, 11:54 AM   #7
diyAudio Member
 
jerluwoo's Avatar
 
Join Date: Feb 2006
Location: Ashland,Ky U.S
Quote:
Originally Posted by narmex View Post
the various feedback paths get everything interconnected, and i'm have a hard time getting it all working together, the THD i get now is way worse than yours, around 0.74% as it is now.
The voltage at Q3 emitter is undoubtedly lower than what it should be, reducing available current output.




Quote:
Originally Posted by narmex View Post
i'm trying to adapt it to 12v supply, that's the only PS i have right now, and also ideally have it driving a small speaker for desktop use, hope you wouldn't mind.
Change value of R11 to 100 ohms. Change the value of R3 by increasing in value (1k 1.2k 1.3k etc. or replace with a multi-turn pot) until you get 6 volts at the emitter of Q3. Recalculate the value of C1= 1/(2*pi*20*R3) . Distortion will still be higher than with the higher B+.
Attached is test I did with 8ohm speakers at 10mW. Loud enough for close field listening with sensitive speakers. I built a second version also for speakers by changing only 2 components, but, this is the headphone section and not the proper place.

Quote:
Originally Posted by Sounduser View Post
Hi. Is the input a 10k pot??
Yes a dual audio taper.
Attached Files
File Type: zip Classic-50mW 8ohm.zip (34.9 KB, 33 views)

Last edited by jerluwoo; 28th October 2014 at 12:15 PM.
  Reply With Quote
Old 28th October 2014, 12:11 PM   #8
diyAudio Member
 
jerluwoo's Avatar
 
Join Date: Feb 2006
Location: Ashland,Ky U.S
Quote:
Originally Posted by sgrossklass View Post
Are you using the above circuit or the one from here?
I have not built or tested the a/b circuit, it was just an example based on this circuit. I am confident it will work well enough for a learning tool. Certainly not hifi in any respect.



Quote:
Originally Posted by sgrossklass View Post
Even if you do get the operating point right, you can still expect to see appreciably worse distortion at a 12 V supply vs. 32 V. Either you leave the resistors as-is, which reduces standing currents significantly, or you decrease them as needed, in which case they make a worse current source approximation, and you may need to employ "real" current sources instead. This circuit also has a reasonably high output impedance pre-feedback due to the bootstrap, so the loop gain gained by that is drained immediately if you connect a low-impedance load (hence also why the speaker version is not bootstrapped, I presume - it would require one more current gain stage).
Correct on all except that the circuit will not function with a current source load on Q2. Output impedance is ~6 ohms. If you look at the test on 8ohm load from previous post you will see a rise in output in the bass on the frequency plot. This is where the output capacitor impedance began to rise creating a higher load value.

Quote:
Originally Posted by sgrossklass View Post
But yes, these circuits with multiple DC feedback paths tend to be real head-scratchers. Part values are typically optimized iteratively, going from desired voltages and device currents.
I had to make a few compromises on device operating points so it would provide the stated power on a wide range of headphone impedances.

Last edited by jerluwoo; 28th October 2014 at 12:25 PM.
  Reply With Quote
Old 28th October 2014, 01:28 PM   #9
diyAudio Member
 
Join Date: Sep 2007
Send a message via Yahoo to linuxguru
Quote:
Originally Posted by jerluwoo View Post
... After digging through tons of old schematics, I kept seeing this topology used in low level gain stages starting all the way back to mid 50's when transistors first became available and then basically disappearing after the early 70's. ...
Interesting, not least because it also uses my favourite small-signal transistor - the 2sc1845. However, you can probably get better sonics and performance with a modified c.1969 JLH 4-BJT Class-A with a PNP input stage - it has about the same parts count as the 3-BJT schematic you've posted. The JLH can also be made to run comfortably on lower rails like 9 to 12V, single-supply.
  Reply With Quote
Old 28th October 2014, 06:08 PM   #10
diyAudio Member
 
Join Date: May 2013
What do cap 5 and 6 do??
  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
Pathos Classic one priax Tubes / Valves 5 12th July 2013 02:32 PM
Are 50mW pots enough for most audio applications? FenderBender11 Parts 3 8th June 2013 10:39 AM
classic rock rock12 Full Range 16 4th February 2013 07:52 PM
'Classic Enclosure'? sled108 Siliconray Online Electronics Store 1 14th November 2012 11:48 PM
Classic 16.0 Lowrider Tubes / Valves 38 6th August 2008 08:15 AM


New To Site? Need Help?

All times are GMT. The time now is 11:06 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2