Help me with headphone "le Classe A" - 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 29th December 2007, 04:24 PM   #1
diyAudio Member
 
nelsonvandal's Avatar
 
Join Date: Nov 2006
Default Help me with headphone "le Classe A"

I don't know any electronics or amp engineering, but I've been after an all discrete portable headphone amp, having tried a number of different opamps but never been totally satisfied with any of them. There's very little interest in this kind of amp in the DIY-world, so I've done some tweaking. I think I like bipolars better than JFETs, so I merged the front end of Hiragas le Classe A with the output stage of JISBOS by steinchen/amb. I've only built one channel. It's actually playing music, seems fairly stable, not much offset drift, very low noise. I'd prefer a low supply so I can use AAA's (9-12V) and have the amp class A biased. An alternative is class AB biasing and use 2x9V batteries. The different transistors is because I had matched pairs of BC327/337/549/559. I guess it would be nicer to use only BC550/560 and BD139/140. Some of the resistor values are are taken from the original circuits, some found by trial and error and some just randomly choosen.

I'm not clever enough to simulate circuits, so I don't know if there are major errors.

Can this be made into a good sounding amp? Can you help me with it? I would like to make it with an active ground channel, so this channel should be unity gain. Could it be made unity gain stable?

Please help me make this become a functional and hopefully good sounding amp!

Click the image to open in full size.
  Reply With Quote
Old 29th December 2007, 08:17 PM   #2
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
Default Re: Help me with headphone "le Classe A"

Quote:
Originally posted by nelsonvandal
I don't know any electronics or amp engineering,...
You certainly know the jargon You managed to match transistors, measure offset, bias it...

So what exactly do you want help with, its working, it is stable. there is little noise, what does it sound like?
  Reply With Quote
Old 29th December 2007, 11:35 PM   #3
diyAudio Member
 
nelsonvandal's Avatar
 
Join Date: Nov 2006
Default Re: Re: Help me with headphone "le Classe A"

Quote:
Originally posted by Nico Ras
So what exactly do you want help with, its working, it is stable. there is little noise, what does it sound like?
It's always hard to tell how it sounds when it's mono. There are no obvious flaws. Even with 9 V supply, it plays very loud without clipping. I wanted to know if there are any major errors, and was hoping someone could lead me in the right direction before I make a real amp out of it. Is the amp sane or not?

E g the feedback resistors is just more or less randomly choosen. Would higher or lower values make any difference if the proportions are the same?

R19 and R20 wasn't used in the amp I "copied", but they made the amp more stable at lower gain. It oscillated like hell at unity gain. The values are just randomly choosen, should I try other values? Should I even use these resistors?

Are biases to Q1-Q6 enough for class A operation? The current draw isn't more than about 5 mA for all of them.

Will BW-limiting caps make the amp more stable, and where should I put them if needed - parallell with R19/R20 or parallell with R11 or between base and collector of Q3/Q4 or Q5/Q6?

In le Classe A R9/R10 aren't used. Will Q5/Q6 add gain now? I added those resistors because of massive current draw otherwise.

Is it likely this could be unity gain stable?

What do you think the output swing is? Is it complicated to calculate? What could the open loop gain be?
  Reply With Quote
Old 30th December 2007, 06:10 AM   #4
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
What is the purpose of Q9 an Q10, it looks like all the wires are shorted?
  Reply With Quote
Old 30th December 2007, 07:05 AM   #5
diyAudio Member
 
nelsonvandal's Avatar
 
Join Date: Nov 2006
Quote:
Originally posted by Nico Ras
What is the purpose of Q9 an Q10, it looks like all the wires are shorted?
The wires shouldn't be shorted of course. I thought the JFETs could do some good isolating the rails of the small transistors from the output transistors to increase the rejection ratio. It seems like they actually do. Since I don't have an oscilloscope, I have to listen to the rails, and there's lower "noise" on those rails now. Maybe they shouldn't be there at all, maybe I should use some other kind of CCS. What do you think?
  Reply With Quote
Old 30th December 2007, 08:38 AM   #6
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
Quote:
Originally posted by nelsonvandal

The wires shouldn't be shorted of course. I thought the JFETs could do some good isolating the rails of the small transistors from the output transistors to increase the rejection ratio. It seems like they actually do. Since I don't have an oscilloscope, I have to listen to the rails, and there's lower "noise" on those rails now. Maybe they shouldn't be there at all, maybe I should use some other kind of CCS. What do you think?
There, fixed. A zero volt generator to the right. Works off 9V. Biased in class A bias current approximately 85 mA. should be loud enough with 3V p-p swing.
Attached Images
File Type: jpg hp amp.jpg (75.1 KB, 1047 views)
  Reply With Quote
Old 30th December 2007, 09:04 AM   #7
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
Quote:
Originally posted by Nico Ras


There, fixed. A zero volt generator to the right. Works off 9V. Biased in class A bias current approximately 85 mA. should be loud enough with 3V p-p swing.
Knocked one channel up, and it actually sound great! Not a bad design you started out with.
  Reply With Quote
Old 30th December 2007, 10:10 AM   #8
diyAudio Member
 
nelsonvandal's Avatar
 
Join Date: Nov 2006
Wow, that was quick! I'm really glad someone like you are helping me with this. Have you done any simulations or measurements?

Just some questions before I go ahead with it and buy the stuff that's needed. Since I want it portable, I'd really want a bit lower current draw to something like 20-30 mA/channel, both because of run time and not having to use large heat sinks. I use 300 Ohm sennheisers and almost never listen louder than 2 V, so I don't think I'll need all that current.

Could the bias be tuned down by changing R13/R14 and R16/R17, and to what values? I think using resistors like in your schematic is more attractive since it's cheaper and takes up less space, but is it still possible to use a VBE multiplier for bias adjustment, and are there other drawbacks of using one? Maybe I could have a switch for high or low current. Are those resistors forming some kind of local feedback loops? You see, I don't know much about this stuff. I just want a pure uncolored nonfatiguing sound.

Will I get a worse result using BC549/559, BC550/560, BC327/337, BD137/138. I have a lot of them lying around.

The circuit to the right - is it a virtual ground circuit? Should it be one of those/channel or one for the whole amplifier? What about output impedance? I'm a fan of active ground channel with very low output impedance for less crosstalk. I think the ground channel is equally important to left/right. Could I still use this amp as an active ground channel after a railsplitter/virtual ground?

PSRR, is it something to worry about in a class A amp with active ground? Should the rails to the amplifying stage be isolated?
  Reply With Quote
Old 30th December 2007, 11:05 AM   #9
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
Quote:
Originally posted by nelsonvandal
Wow, that was quick! I'm really glad someone like you are helping me with this. Have you done any simulations or measurements?.....
Sorry mate, I read somewhere 32 Ohms. I was wrong. I will reduce bias.

Yes I have simulated it, it looks pretty good. Transistor types are not that important now, knock it up with what you got just to satisfy yourself first. I use MPSA types because I have hundreds. You can use your BC types throughout, even for the output they will run a little warm at about 135 mW, but well within general purpose low power types.

To lower bias to about 25 mA, change R12 & R15 to 120 ohm. If the gain is too low or too high change R10, I did not know what sensitivity you need.

Of course class A draws a constant current from the supply so NiMH rechargeable batteries would work nice as they are designed for deep cycling. You want to know more about charging, e-mail me I don't think this forum is for battery chargers.

Furthermore power supply rejection is not really an issue for batteries and class A. If you want to use a power supply, just use a 9V regulator.

Almost any 9V transformer will work, you need less than 1VA. You get little encapsulated PCB types that is great for the job.

e-mail me I will send the simulations, it is too much trouble to post here.
  Reply With Quote
Old 30th December 2007, 11:13 AM   #10
diyAudio Member
 
Nico Ras's Avatar
 
Join Date: Nov 2005
Location: East Coast of South Africa
BTW I am impressed enough with the simulation I am starting to lay a PCB to build it. I am using a single ended class A at the moment (earlier posts) but I like what I see. I also use HD650. Would you be interested in a PCB layout. If yes what format you want it in?

I am thinking of using surface mount throughout it will cost almost nothing and it will fit into a little pocket box available from OKW in Germany, I have a few in my drawer.
  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
PCB Group Buy for the "DAC + headphone amp" Project Geek Group Buys 345 20th April 2010 12:23 AM
Beyerdynamic's Headzone headphone promise "perfect spatial impression" Nordic Headphone Systems 1 28th August 2007 09:10 PM
"Digital" headphone amp from Crystal/Cirrus peranders Headphone Systems 0 19th April 2002 07:36 AM


New To Site? Need Help?

All times are GMT. The time now is 09:24 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