Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Tricky Headphone Amp design for AK4396
Tricky Headphone Amp design for AK4396
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
Thread Tools Search this Thread
Old 12th May 2012, 01:47 AM   #1
SunRa is offline SunRa  Romania
diyAudio Member
Join Date: Aug 2004
Location: Romania
Default Tricky Headphone Amp design for AK4396

Hi there, long time since not posting in the tubes forum

Now, I have this AK4396 DAC I build some time ago and I want to build a headphone amp for it to drive my 50ohms cans.

These are the design goals:

- minimum number of stages
- no solid state voltage amplification stages (I would agree I guess with unity gain buffers and current buffers)
- smart techniques for mitigating noise (balanced topology, AIKIDO - stuff , CCS, etc.)
- I need analogue volume control

The Plan

I set my eyes on the AIKIDO headphone amplifier configured as a White Cathode Follower. I plan building this with 6n1p (input) and 6h30 (output) tubes. The first one is a pretty nice, relatively high gain, low noise part while 6h30 needs no introduction, its high current capabilities being perfect, I assume, for this role.

The Problem

Now the balanced AH4396 outputs +/-1Vpk and that's a nice thing, but the AIKIDO amplifier will have a gain of around 16 with B+300V and ~50ohm output impedance.

1. I don't have much experience with headphone amps but isn't that a lot? I mean, I am driving my Sennheisers directly from the output of the AK4396 (directly through the output caps, no other stages between) and I find them to be really loud (I have no real bass though and they lack some resolution, I clearly need more current).

2. If I go the Aikido route I still don't get volume control. I don't think putting a purely resistive load in front of the headphone amp is a good idea.

Alternatively I could use an Aikido Cathode Follower or a Broskie Cathode Follower, which are unity gain amplifiers (plenty of current), but are giving high-z outputs ~250-300 ohm. Great if I'd have some Sennheiser 605 , but I don't. An I still don't see a way of having volume control on it.

The obvious way would be to chain one of these unity gain amplifiers with an Aikido headphone amplifier and stick a pot between them. It should give nice noise rejection figures, these are elegant topologies, impedances would be matched at each point but 6 tubes and 3 stages seems to me a bit inelegant.

So, would you have any suggestions for me?

Your help is much appreciated!
  Reply With Quote
Old 12th May 2012, 05:17 AM   #2
tikiroo is offline tikiroo  New Zealand
diyAudio Member
Join Date: Nov 2007
Location: New Plymouth
I'm starting to get some parts together for the same type of thing (using 6gf7 dissimilar triodes). I've got 32 ohm Grado headphones and I was planning to apply a bit of feedback to get the gain and output impedance down. The Tubecad site has some examples of this, there is an example in one of the latest blog entries with 12at7's and 6AS7's:

Aikido Push-Pull
  Reply With Quote
Old 12th May 2012, 05:39 AM   #3
qusp is offline qusp  Australia
diyAudio Member
qusp's Avatar
Join Date: Oct 2009
Location: Brisbane, Australia
why does it have to have analogue volume? have you tried any of the inbuilt dac volume controls lately such as in the ESS and I thought the AKM? it has gotten to a point where the expense, size and complexity of analogue controls are nolonger IMO justified. in my experience the new breed of digital direct dac control leaves any analogue control, regardless of cost in the dust. channel matching is perfect, no impedance issues to worry about, smooth stepless control and no contacts to age.

yes 16x for grados is totally over the top, you dont need more than 2-3 or so. you need curent gain for grados, not voltage gain, but they do need more than +/-1v, but if you match the voltage gain well to the headphones you will hardly need the volume control.

also, its better to avoid calling anything tube based balanced; by definition it cannot be balanced. it can be balanced in effect, but till there is the possibility of a differential it remains something else. circlotrons make excellent headphone amps though, ive been looking at Johns circuits too lately for something like this.

which Sennheisers are you talking about?
  Reply With Quote
Old 12th May 2012, 06:25 AM   #4
SunRa is offline SunRa  Romania
diyAudio Member
Join Date: Aug 2004
Location: Romania
Hi, I just realized that using 6h30 in each stage, although a waste (3.2A only for heaters) will give me a smaller gain of only about ~6, which should be fine enough for my HD 558 (50 ohms).

Studying the Aikido amplifier I understand it is a plate follower providing gain in the first stage and a cathode follower in the second, I wonder if I could stick the pot between these two stages

@qusp. Thanks for the digital attenuation idea. AKM4396 apparently supports this, although I have no idea how to implement it. A micro-controller is needed and I need to learn tampering with the registries. However the real issue I am concerned of: I don't understand how attenuation is implemented in these chips. And is this bit perfect any-more once I go this route? Also not sure from datasheet if attenuation works both with 16 and 24 bits or if it performs any dithering before it starts cutting bits. I have a lot more trust in the ESS implementation (32bit wide path, etc.). However I can't touch a kit like that to soon because of the price. What I could do though, is to design a dac board around AK4397, which is a native 32bit DAC and digital volume control should be well implemented. Mhh.. you got me thinking

@tikiroo Thanks for the link, interesting! I'd want for this one to stay away from negative feedback though (although I admit that a small amount on such a simple circuit shouldn't be detrimental).
  Reply With Quote
Old 12th May 2012, 07:18 AM   #5
SunRa is offline SunRa  Romania
diyAudio Member
Join Date: Aug 2004
Location: Romania
As for controlling the gain of the first stage in an Aikido there's a starting point here: Aikido Tube Amplifiers and here: http://www.tubecad.com/october99/page9.html

The first link discusses input stage variations and the reason for dual triode in that position and the second shows the math for the Common Anode / Plate Follower configuration.

I can not but feel amazed about the great value of the TubeCad site. Great stuff in there.

Last edited by SunRa; 12th May 2012 at 07:23 AM.
  Reply With Quote


Tricky Headphone Amp design for AK4396Hide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
New AK4396 DAC board design support needed SunRa Digital Line Level 122 8th September 2011 01:41 PM
Headphone amp design andyoooooobun Solid State 0 28th January 2011 04:24 PM
Looking for a Simplistic Headphone Amp design merlin el mago Headphone Systems 24 7th June 2010 12:55 PM
Headphone amp design Omicron Headphone Systems 9 21st September 2007 04:24 AM
Tricky op-amp based VR problem matejS Power Supplies 22 29th August 2007 03:27 AM

New To Site? Need Help?

All times are GMT. The time now is 03:19 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.79%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio