Simple Class A Headphone Amp Using THAT1646 - Page 2 - 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 3rd January 2012, 07:53 PM   #11
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
Thanks sgrossklass.

Yes, it can be used without the build-out which is there primarily for added short-circuit protection. I've listened to it both ways.

Without the build-out I've driven capacitors >>100 nF. Even with a ridiculous 1 uF I've not seen any sign of instability though as you might expect with 1 uF there is significant triangulation of squarewaves.

This circuit is also being used a cable driver and in that application it's driven over 1000 feet of STP.

WRT to build-outs I can't speak for what's best driving 'phones. I'm sure it depends a lot on the phones.

For balanced cable-driving I've found that build-outs should be optimized. That discussion begins here:

Pro Audio Design Forum • View topic - Balanced Outputs

And some comparative measurements of the SSM2142, DRV134 and THAT1646 both with and without added build-out:

Pro Audio Design Forum • View topic - Balanced Outputs
  Reply With Quote
Old 5th January 2012, 02:59 PM   #12
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Quote:
Originally Posted by mediatechnology View Post
WRT to build-outs I can't speak for what's best driving 'phones. I'm sure it depends a lot on the phones.
To give you an idea, here's an impedance plot for UE triple.fi 10s (if that isn't a rollercoaster ride I don't know what is - they are among the most critical ones around), and here's some of the variation you can find among "big" cans:
Click the image to open in full size.

Headphone cables typically run in the low single-digit nF range.
Quote:
Originally Posted by mediatechnology View Post
For balanced cable-driving I've found that build-outs should be optimized. That discussion begins here:

Pro Audio Design Forum • View topic - Balanced Outputs
How did you pick up the signal? If you're measuring single-ended, the full conductor-to-shield capacitance figures in, which would approximately cancel when balanced. And in STP that's pretty big when compared to conductor-conductor capacitance.

Additional output resistance essentially counteracts two factors that can cause overshoot and ringing:
1. Amplifier output impedance becoming inductive towards higher frequencies
2. Output stage being slowed down by capacitive loading, which adversely affects 1.

Obviously choosing too high an R eventually restricts bandwidth too much.

Thankfully all of these effects are not terribly relevant at audio frequencies, at least with the kind of cable lengths you'll find at home or in normal-sized studios.
  Reply With Quote
Old 5th January 2012, 03:18 PM   #13
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
Thanks for the headphone impedance plots. My tests using Sony MDR7506's sounded about the same both with and without build-out. Without seemed a tad tighter in the low end but not much. I chose the added build-out to minimize Isc. You can have it either way.

The cable driving tests were recovered differentially by a THAT1246 or a cross-coupled pair of THAT1286 in some tests. That info is buried in a really long thread.

Burdick (Benchmark) in his clean audio installation guide chose 30 Ohms/leg build-out (for cables) noting that it was not maximally-flat. He had the requirement to transmit HS timecode through his DAs over very long distances and for that reason tolerated the slight peaking. The 40-50 Ohm range (per leg) appears to be maximally-flat. What was interesting to me is that for a given recovered level the peak current requirements and IM of the line driver are significantly lower with a slightly higher build-out. Obviusly there can be too much of a good thing.

I do realize that most people aren't driving 1000' of STP.

The test with it's peak 6V us slew rate may not be real-world either.

Last edited by mediatechnology; 5th January 2012 at 03:33 PM.
  Reply With Quote
Old 9th January 2012, 11:08 PM   #14
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
I updated the schematic to show the build-out resistor as optional to avoid any confusion that might cause as well as add the mod to use the DRV134.

It's been awhile since I ran capacitive load tests and wanted to confirm that without any build-out it's stable at loads >>100 nF. It is.

A also ran some tests both with and without a 27R build-out driving driving MDR-7506's phones at 100 Hz and the THD is a good 20-30dB lower without the build-out.

For those that have strong feelings against using build-outs to drive phones there's no reason that you have to with this simple circuit other than the additional short-circuit protection it provides.
  Reply With Quote
Old 11th January 2012, 08:50 PM   #15
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Quote:
Originally Posted by mediatechnology View Post
A also ran some tests both with and without a 27R build-out driving driving MDR-7506's phones at 100 Hz and the THD is a good 20-30dB lower without the build-out.
BTW, I'm still not convinced that this measurement means a whole lot. These 20..30 dB are probably corresponding to the reduction in output impedance, so back-EMF will have proportionally less of an influence. One would have to measure actual acoustic distortion for a meaningful result.

When short-circuit protection is desired, I'd normally suggest some extra collector series resistance for the driver transistors... but I don't think an EF output running open loop has enough PSRR to avoid a negative effect on distortion, plus voltage swing is reduced somewhat.
  Reply With Quote
Old 11th January 2012, 09:02 PM   #16
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
Quote:
BTW, I'm still not convinced that this measurement means a whole lot. These 20..30 dB are probably corresponding to the reduction in output impedance, so back-EMF will have proportionally less of an influence. One would have to measure actual acoustic distortion for a meaningful result.
I agree. I ran tests similar to what John Siau at Benchmark has done recently.

See: Do Specifications Lie? Do Our Ears Lie? Where does the Truth Lie? An Examination of Headphone Amplifier Performance Specifications | Benchmark Interaction

And: The "0-Ohm" Headphone Amplifier: The Sonic Advantages of Low-Impedance Headphone Amplifiers | Benchmark Interaction

What I did find to be more significant was the reduction in cross-talk:

Click the image to open in full size.

There appears to be significant due to inter-conductor induction.

And an improvement in frequency response flatness:

Click the image to open in full size.

I really want to thank you for pointing out your findings WRT build-outs in your OP.
  Reply With Quote
Old 18th January 2012, 02:10 PM   #17
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
With regard to my earlier post on the subject of reduced crosstalk.

Lowering the build-out makes it look really good at the connector.
I didn't see that from the beginning.
At the transducer end there's no improvement in crosstalk.
Both ricardo from the Pro Audio Design Forum and John S. set me straight on that.

Regardless one can have it either way: Build-out or none.
  Reply With Quote
Old 5th February 2012, 04:34 PM   #18
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
I had some prototype boards made and thought I'd share the pic.

Click the image to open in full size.

This board is configured so that I can use it as a stereo headphone amp, dual single-ended line driver, balanced line driver with cross-coupled inputs, a studio "re-balancer," and dual output distribution amp.

It is also a very capable transformer driver.

I'm primarily going to use mine as an HPA and a second set of two to provide balanced sound card outputs for test instrumentation.

The outputs are internally ballasted and can also be paralleled to double drive current.

Those jumper links to cross-couple or parallel the inputs prove to be very handy.

I estimate the output impedance (it's open loop) to be <2 Ohms. It can be run with or without build-out resistors into large capacitive loads. 30 Ohm loads can be driven to the 100 mW range with decent performance.

The inputs are balanced using either a THAT1240, 1243 or 1246. There's a Phoenix block for level controls for HPA use. There is no voltage gain but I've found that with the low-impedance phones I'm using it isn't an issue. Having precise unity gain was more important for its' line driver role.

I've run it on supplies as high as +/-18V or as low as +/-9V. Things head south around +/- 6V. I also included a soft clipping circuit which tracks the rail voltage.

As expected for class-A, the current drain is huge. I use 85 mA Iq/channel typically so the totals run about 200-225 mA.

I'll post some RMAA but I'm not sure how accurate they'll be since the sound card masks most of what's going on. I need to setup a null test which is how I originally tested it.
  Reply With Quote
Old 12th February 2012, 02:51 PM   #19
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
I did some FFTs:

Iq was set to 85 mA per channel.

30 Ohm load at 100 mW. Rbuildout = 0R.
Green is output, Red generator monitor.
The driver adds ~ 0.0017% THD.

Click the image to open in full size.
Dual Class-A Headphone Amplifier with 30 Ohm load at 100 mW.

600 Ohm load, single-ended, +20 dBu. Rbuildout 0R.
Green is output, Red generator monitor.
Note that the output THD is essentially identical to the generator.
To fully see what the THD is a null test is required.
The driver subtracts ~ 0.0002% THD.

Click the image to open in full size.
Dual Class-A Line Driver with 600 Ohm load at +20 dBu.
  Reply With Quote
Old 6th June 2012, 08:42 PM   #20
diyAudio Member
 
Join Date: Apr 2006
Location: Dallas, Texas
Update showing the DRV134 build and the schematic:

Click the image to open in full size.

Click the image to open in full size.

These are a few applications for this high current output in addition to driving headphones:

1) A stereo headphone amp.
2) Dual single-ended class-A line output.
3) Fully-balanced line output with cross-coupled differential input.
4) A high-current line "re-balancer."
5) Dual single-ended or differential transformer driver.
6) Dual output distribution amp.
7) Parallel operation (without external ballast resistors) for double output current.
8) Passive EQ network driver.
9) Control Room Monitor driver.

Last edited by mediatechnology; 6th June 2012 at 08:47 PM.
  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
TSHA.. A very simple Headphone Amp idea.. Dxvideo Headphone Systems 15 1st February 2009 07:04 PM
Simple Headphone Amp Nico Ras Headphone Systems 69 8th August 2008 12:19 AM
SImple class D amp croccodillo Class D 0 29th May 2008 01:56 PM
Discrete simple headphone amp kubeek Headphone Systems 15 7th September 2007 04:13 PM
Capacitor type for simple headphone amp GianLorenzo Headphone Systems 3 20th December 2004 08:45 AM


New To Site? Need Help?

All times are GMT. The time now is 05:18 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Đ1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2