SMPS advice for headphone amp - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies
Home Forums Rules Articles diyAudio Store Gallery Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

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 7th July 2015, 02:08 PM   #1
Bappy is offline Bappy  United Kingdom
diyAudio Member
 
Join Date: Jul 2015
Default SMPS advice for headphone amp

Hello,

I'm toying with an idea... I'm going to build a custom PC chassis for my next machine, it's going to be an HTPC/Steambox running an AMD APU (a 7870k if all goes according to plan)

I thought to myself, what could be a better feature of an HTPC chassis than building in an integrated audiophile headphone amplifier?

So anyway long story short, I came across plans to something called 'Objective2'.... I suspect from all the forum chatter about this opensource circuit many people here will be familiar with it.

Now my issue is with powering it. The PSU in the PC will naturally have high current 12vdc available in a molex, so I should be able to power the amp from there. However, the O2 takes 20vac as a supply or -8.4v0v+8.4v when operating on batteries. So I might be able to get away with a voltage divider on the 12vdc supply to get 0v, 6v and 12v to simulate -6v0v+6v BUT that means I would have to capacitor-couple the signal input to avoid effectively short circuiting the PSU by connecting 6v to the DAC's ground output on the sound card.

I have had the thought that I could remove the half-wave rectifier and the battery section of the O2, and replace it with +12v and 0v from the PSU, and run an SMPS in inverting buck-boost topology to get -12v to go into the other side of the O2's power regulator circuits.

The problem is that 1) I have no idea if this is even technically feasible 2) I have no idea how to design and or build an inverting buck-boost SMPS.


Would someone with the relevant knowledge please be so kind as to advise if this is a goer or not? Or advise if there's a better option?

If this is a feasible solution, any guidance on where to start with designing an inverting buck-boost SMPS to convert +12vdc to -12vdc with sufficient current handling capacity would be very very well received

so far I've read a number of product datasheets for PWMs (this one looks promising http://www.st.com/web/en/resource/te...CD00001232.pdf) as it has a demo circuit described that produces -12v from a 4.5-6v input (could use the 5v line of the molex unmodified) but I don't know if 100mA is enough current for the O2 and I don't know what changing component values does to the input voltage tolerances or the output voltages as I don't understand the maths well enough.


Thanks for reading
Bappy
  Reply With Quote
Old 7th July 2015, 02:49 PM   #2
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
You don't need much current for an O2. Simplest solution is an off the shelf DC/DC convertor taking a nominal 12 volt input and producing either -/+12 or 15 volts depending how you want to use it.

These are available from many manufacturers in different power ratings. This are just a couple of examples.

IR1215S - XP POWER - DC/DC CONVERTER, SEMI REG, DUAL 15V | Farnell element14

JCA0212D02 - XP POWER - CONVERTER, DC/DC, 2O/P, 2W +/-12V | Farnell element14
__________________
-------------------------------------------------------
Installing and using LTspice. From beginner to advanced.
  Reply With Quote
Old 7th July 2015, 04:42 PM   #3
wwenze is offline wwenze  Singapore
diyAudio Member
 
Join Date: Mar 2008
IIRC I tried LT1054CP with the same intention as you. The result was noisy.

I recommend a linear regulator after the SMPS. On both rails. Even if at the cost of a few volts. Even then I would suspect if the resulting noise would be suitable for earphones because dirty ground is a problem.
  Reply With Quote
Old 7th July 2015, 07:15 PM   #4
Bappy is offline Bappy  United Kingdom
diyAudio Member
 
Join Date: Jul 2015
Thank you both kindly for your quick responses

So I'm thinking that if that unit that provides -/+15v has a max voltage of 12v, and the pc psu will produce anything from 11.8 up to 13.5 (maybe its better on more pricey PSUs) I'd do well to use a 12v 1A 3-pin regulator on the input and follow it with something like a 1A schottky (which tend to drop 0.6 to 0.7v) to get a steady 11.3 or 11.4 volts.

Re: noisy ground, the designer of the O2 makes a lot of claims regarding the psu section of the O2's circuit cleaning uo the supply. am I understanding it correctly that this regulation only affects the plusitive and minusitive (sorry that's my maths teacher talking) and doesnt do anything to clean up the ground?

Thanks
  Reply With Quote
Old 7th July 2015, 08:00 PM   #5
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
The '12 volts input' is a nominal figure. Now the full data sheet for the 3 watt IR1215S device quotes -/+10% and I've seen other devices that accept 8 to 15 volts or more. I think its probably a non problem in practice tbh. It will be fine.

There wouldn't be enough 'headroom' or overhead to use a 12 volt regulator on a 13.5 volts supply. The reg needs around 2 to 3 volts more applied to the input in relation to the output in order for it to function properly.

Regulation and grounds...

Everything is measured 'with respect to something else' and that something else is what we call 'ground'. The point we call 'ground' in a circuit like the O2 is arbitrary, because its not connected to earth or mains ground... as such it could just as easily be the positive or the negative rail and we would then measure everything else to that point. So ground is just a point we decide upon and then all things are referred and referenced to that location.

We try and keep grounds clean by ensuring that we don't contaminate them by feeding in 'dirty' currents. All this happens because wire and circuit board print has resistance. If the resistance were zero there would be no problem because any current flowing couldn't cause then cause an unwanted volt drop and by implication, modify the wanted signal in some way.

The O2 is all low power with little chance of any circulating ground currents... and so all you need do is ensure that the convertor ground connects to the main O2 ground which is the junction of the power supply caps at the O2's power input. Connect the -/+15 volt convertor output to the appropriate caps on the input of the O2's regulators.
__________________
-------------------------------------------------------
Installing and using LTspice. From beginner to advanced.
  Reply With Quote
Old 7th July 2015, 10:07 PM   #6
Bappy is offline Bappy  United Kingdom
diyAudio Member
 
Join Date: Jul 2015
Legend thanks. I had thought it woukd be ok, afterall if the pc psu ground was too noisy for audio, you'd never get acceptable audio quality out of a pc soundcard. And if there is ripple on the ground, the same ripple will be present on the ground of the input signal since all the dc voltages on a pc psu share a common 'ground' of arbitrary actual potential. If you've got the same noise on the input as you have the output, and the headphones are not grounded to anything else, it will in theory all be silent... I think

I'll have a play about and probably start ordering bits for prototyping a psu for the O2 soon. If you're happy that 100mA is enough for the O2 I'll trust your expert judgement and get some of those units to test with.

Thanks very much again

Will post my progress
  Reply With Quote
Old 8th July 2015, 12:29 AM   #7
wwenze is offline wwenze  Singapore
diyAudio Member
 
Join Date: Mar 2008
The difficulty gets high when trying to power both from computer PSU. Powering the amp with another power supply is ok.
  Reply With Quote
Old 8th July 2015, 01:18 PM   #8
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
Quote:
Originally Posted by Bappy View Post
Legend thanks.
Lol, hardly that but thanks

Yes, that's basically it with grounds.

The opamps in the 02 draw minimal current, just a few milliamps each and so a 100ma supply is fine for headphone use. If the amp is a permanent fixture running of the PC then there is a lot of circuitry on the 02 board that can be omitted.
__________________
-------------------------------------------------------
Installing and using LTspice. From beginner to advanced.
  Reply With Quote
Old 8th July 2015, 01:21 PM   #9
Bappy is offline Bappy  United Kingdom
diyAudio Member
 
Join Date: Jul 2015
Hi, now this is confusing me.

I have one forum member effectively telling me "use IR1215S - it will be fine" and another effectively saying "that wont work"....

I have some basic knowledge of electronics and what I know is this;

If the ground of the O2 is not at the same potential as the ground of the computer's motherboard, there will be problems. Bad and melty problems.

Unless of course the ground of the O2 has been decoupled from the PSU by means of a transformer, in which case, the decoupled ground of the O2 will happily come up to (or drop down to) the same potential as the motherboard/soundcard's ground and the -/+ rails will just be 15v relative to whatever that happens to be. <- please shout if I've made some retarded error here!


Now, looking at the little PSUs that Mooly has suggested, I can see that the -vin pins are NOT called common and there are no documents pertaining to the internal design of those units.

If I connect +12v to +vin and 0v to -vin; relative to -vin, what are the measured voltages on -vout, 'common' and +vout? do I get -15, 0, +15 relative to -vin? or 0, +15 and +30? do I get maybe -955, -940, -925 with respect to -vin or some other, equally obscure/random/unstable voltage?

effectively what I want to know is, on those little transformers, can I connect -vin straight to 'common' without melting the little unit? if that's both safe and the intended usage, I see no reason why it would be 'difficult' to power both the O2 and the Motherboard/soundcard from the same 12v souce on the computer's PSU?

ta
  Reply With Quote
Old 8th July 2015, 01:32 PM   #10
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
The 15-0-15 output on those little power supplies are totally isolated from the input.

So you connect the input across your PC's 12 volt line and the output is free to be used as you wish. So in practice you would couple the '0' of the 15-0-15 output to the '0' of the input, or as some call it the - or minus. Lol, that's confusing.

Think of a 1.5 v AA battery. It has a plus and minus marking. The minus is what we call 0. The plus end is 1.5 volts above that.
__________________
-------------------------------------------------------
Installing and using LTspice. From beginner to advanced.
  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
Looking for valve based headphone amp advice mcandmar The Lounge 19 19th October 2013 01:47 AM
Advice sought on thermionic headphone amp Simon B Tubes / Valves 46 10th May 2012 10:31 PM
My first Headphone amp need expert advice Wessel Lemmer Headphone Systems 10 10th March 2012 02:12 PM
need basic custom headphone mixer/amp advice oshunANDurth Headphone Systems 4 20th April 2011 07:05 AM
5v SMPS for headphone amp insta Headphone Systems 6 24th April 2009 04:30 PM


New To Site? Need Help?

All times are GMT. The time now is 10:40 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2