using an impedance adapter to make an external DAC drive 32Ω headphones?

[leeperry]

hi there, I've got this DAC(using two LT1028ACN8 opamps as PCM1793 LPF): Spitfire 24-Bit DAC Amplifier

It's said to have a 100Ω output, and my Sony CD1000 headphones are 32Ω/104dB...could I somehow use this kind of adapter to make them both happy? RESISTOR ADAPTOR (3.5MM PLUG) - eBay

I've tried to plug the headphones directly to the RCA output, but it's a big failure...sound is very bright and harsh.

I also know someone who tried to plug 16Ω headphones to a headamp from Firestone Audio, and it instantly shut down(short?)...anyway, I know passive attenuation is baaad but I don't want to color the sound any further and my headphones are VERY sensitive(104dB) so they're very easy to please...they're said to need a gain of 2, which is very low.

I've tried using this adapter: Shure EA650 - In-Line Headphone Volume Control

when the volume is maxed out on the Shure toy(which is essentially a variable resistor at heart I think?), it reads 35Ω(my CD1000 is 32Ω) and at the min volume more like 600/800Ω....but the issue is that there's a major impedance mismatch between the 2 channels, so sometimes it's 500Ω/800Ω...and the stereo tracking is poor. OTOH, the sound is very balanced and very nice

I really want to listen to the discrete linear regulated PSU of my DAC and its LT1028AC opamps...nothing more, nothing less! I just need something to make the 100Ω output of the DAC match the 32Ω input of my CD1000, as I can manage the volume digitally in the computer anyway.

I guess I could use an impedance adapter and call it a day? how about 620Ω?

thanks for any insight,

 

[john@work]

leeperry,

i think your 32ohm headphones will need a source impedance of much less than 32ohms, less than 10ohms might be worth a try.
In addition they will (most likely) require less voltage than the specified 2 volts of your DAC.

The plugs you suggested all put an additional resistance between DAC and headphone. This will increase the source impedance, which is the opposite of what you need.

To operate these phones seriously you'll need a headphone amplifier (which has an output impedance suitable to drive 32ohms headphones).

But you may want to give it a try and install a fix voltage divider between DAC and headphones. This will lower the voltage dramatically (you will NOT be able to play loud!) but may work. You will have to try out some values. Starting point may be 100ohms (to DAC) and 22ohms (to ground).
Well, this will give about 1/20 the max output voltage of the DAC but will probably be enough to get a listening impression.

 

[leeperry]

ok, thanks for the tips!

when my cd1k is directly plugged to the DAC output, I cannot go over 10% volume on the computer...because my cd1k is so sensitive at 32Ω/104dB I think.

Using the Shure attenuator, the DAC sees 600Ω/800Ω and the headphones sound fine...the only issue is the poor stereo tracking due to the lousy pot it's using...I was kinda hoping to present the DAC w/ a high load like +600Ω w/o a pot, and I'd be good to go? SQ is great to my ears, as the LT1028AC can drive 32Ω AFAIK..

I understand it's not supposed to work, but it very much does...this Shure attenuator was meant to be used w/ headphones in the first place

BTW, DC offset is around 3mV...perfectly fine and harmless to the headphones, it's been working fine for weeks anyway.

 

[leadbelly]

1) IMHO, most of this thread is nonsense.

2) A simple opamp like the 2604 should be able to drive headphones adequately; since it is not in your case, my guess would be that there's a smallish cap inside that is rolling off the bass.

3) I would say if you're willing to modify it, it would be pretty simply to fix your DAC; if not, just build an external buffer with something like a BUF634.

 

[leeperry]

1) sure

2) it came w/ a 2604 stock...but the 1028 sounds FAR better

3) ok, great. thanks for passing by. I'm starting to suspect that there's more in that Shure attenuator than just a variable resistor

 

[john@work]

1) IMHO, most of this thread is nonsense.

2) A simple opamp like the 2604 should be able to drive headphones adequately; since it is not in your case, my guess would be that there's a smallish cap inside that is rolling off the bass.

3) I would say if you're willing to modify it, it would be pretty simply to fix your DAC; if not, just build an external buffer with something like a BUF634.

.. yes an opamp like '2604, '1028 or similar may drive headphones. But
a) output impedance is stated 100ohms, coupled 'without capacitors in signal path' .. 99% likely they got a 100ohms resistor in series to output
==> questionable how this works with a 32ohms headphone..
b) in this case the headphones will draw a considerable current, which has to be provided by power supply (of opamp/device). This may work.. or may not work.

so, leadbelly, nice 'first thinking' but please look a bit closer.
Of course, if problem b) is not valid then another solution would be to change the series resistor inside the DAC from 100ohms to something lowish (and reduce the amplification of the LPF/last stage, if possible.

But, leeperry, if the headphones sound good with a pot in front (not in theory but in practise i.e. to your ears) then why don't you try just building a passive adapter yourself, using two trim pots of 1k. You may adjust them separtely to exactly the same values and then control volume with the PC. Costs are pretty low. For maximum quality you may try different settig and replace pots with fixed resistors when you know the optimum values.

 

[leeperry]

well, it really sounds fantastic w/ the Shure attenuator tbh

that's what it looks like through RMAA(on the stock OPA2604): RightMark Audio Analyzer test: SPITFIRE

my only problem is that there's an impedance mismatch between the 2 channels and a poor stereo tracking due to the cheapo built-in Shure pot...and I can control the volume digitally, so I was kinda asking what kind of fixed resistor value I could use? at the min volume, the DAC sees 500/800Ω...so how about a 620Ω adapter like this? RESISTOR ADAPTOR (3.5MM PLUG) - eBay

is there a formula that could make a 100Ω DAC output match a 32Ω headphone? power supply is no problem, I get tons of bass, a clear soundstage and very nice trebles...I just would like to move to a fixed resistor to put an end to the poor stereo tracking/impedance mismatch

I was told that a 100Ω output could easily drive 600Ω headphones, so I'd need something like 15Ω output from the DAC?

anyway LT1028AC makes my day, I love this opamp and I don't want to destroy its SS w/ some nasty BUF634/NJM4580..

thanks!

 

[john@work]

I agree with the 150/600ohms statement.

Actually you have to try; best value will be a compromise and depend on headphone type.
But please remind that the pot actually is a voltage divider. So you will need two resistors.
My suggestion is to start with 100..470ohms from DAC output to headphone connection and 22..47 ohms from headphone connection to ground. Get some different values and try.

 

[leeperry]

a voltage divider? ok, so it kills two birds w/ one stone basically, together w/ increasing the impedance on the DAC side? oh well, I'll try to find a position that's pretty much stereo balanced and leave it as is then...the only thing worrying me is the impedance mismatch on the DAC side, what happens when one channel is 500Ω and the other 800Ω? FR mismatch?

 

[bocka]

Well, a nasty idea: As many people like passive i/v converters (simple resistor) wouldn't it be worth a try to direct connect the headphone to the dac? The tradeoff is the volume control has to be done by the dac and the output is current driven. But as many power amps only use a series resistor to make a headphone output this would the most simple solution for a headphone dac

 

[leeperry]

I actually tried it, but there's a huge FR mismatch...the sound is unbearably bright and harsh..

just like they show on that link if you click on "TechNotes": Auzentech

An externally hosted image should be here but it was not working when we last tested it.

 

[leeperry]

another solution could be one of those: SM PRO AUDIO NANO PATCH PLUS SILBER - U.K. International Cyberstore

 

[leeperry]

ok I'll send an email to that german manufacturer

apparently the Shure attenuator is called a "L-Pad": l-pad attenuator - Google Search

An L-Pad is a very simple, purely resistive, voltage divider attenuator. It consists of 2 resistors, one in series with the speaker, and one in parallel with it. The ratio between the 2 forms a voltage divider.

I'd like one w/ perfect stereo tracking and matched stereo output impedance

 

[leeperry]

ok, found some links!

http://www.diyaudio.com/forums/headphones/152993-headphone-attenuator.html

http://www.diyaudio.com/forums/pass...ce-relay-resistor-logarithmic-attenuator.html

http://www.diyaudio.com/forums/parts/13632-t-link-attenuator-constant-impedance.html

A PI and T attenuator

too bad they don't mean much to me

http://www.ehow.co.uk/about_6331454_l_pad-schematics.html

"L-Pad" is an audio electronics term that refers to a resistive attenuator network that reduces the signal to a speaker while matching the impedance to that of an amplifier. The network consists of two resistors, either fixed or variable, that are connected so that one is in series and the other is in parallel with the speaker load.

could I easily build one w/ a fixed value somehow? I've got a soldering iron and I'm not afraid to use it

 

[john@work]

impedance mismatch on DAC side will result in different voltage levels because of the 100ohms output impedance of the DAC. But I suspect this would be a real problem..

Of course it is a good idea to find a good position of the Shure attentuator and to keep this position. If you want to really fix it, measure the impedance and replace the pot by four fixed value resistors.

bocka,
this sounds interesting. I'm just not shure whether leeperry is willing to solder inside his DAC.. and this solution would give a much bigger 'impedance mismatch'.. probably a parallel resistor + capacitor would help and offer a great solution.

 

[leeperry]

yes, I'd really like to make a fixed L-Pad...if only I knew what resistors values to use?

 

[john@work]

ehm.. I did mention some values above (to start with).
It would be even better to find a good sounding position of the Shure attentuator, measure resistance between a) input and output (this will be R1) and b) between output and ground (this will be R2).

 

[leeperry]

well, you didn't mention "L-Pad" as this is what I want to do...and apart from cheapo attenuators, noone makes serious ones

but I'd rather get a passive attenuation than adding opamps/caps/psu in the audio path

ok I will measure everything you told me and will come back w/ figures..thanks for your help!

 

[leeperry]

ok, the manufacturer of this thing: SM PRO AUDIO NANO PATCH PLUS SILBER - U.K. International Cyberstore

tells me that it's not a "L-Pad" but that it will work fine w/ headphones...I've asked for more details to triple-check with you, but so far he told me:

The nanopatch+ has an inline-potentiometer so you can simply attenuate the volume of your DAC. You, of course, can use the nanopatch+ as an attenuator for your headphones..but not as a L-Pad. The resistors are in series with the speakers/Headphones, that's why it's not a L-Pad.

I guess the pot would work as a voltage divider? and not change the impedance on the DAC side? and it looks so neato in silver, tell me it can work...puh-lease

 

[leeperry]

hehe, I'm not that retarded after all...that's what an amp manufacturer just told me:

I've been thinking about your idea. It was/is a very good idea.

Essentially 100% passive headphone amp - for those people who have a low enough output impedance signal source (with a large enough output signal) to drive their phones.

I do not know if you know how good of an idea that is. I think it is really good.

So what I brought to the table was the info about the possible L-pad configuration.

I would like to try this out. I still want to open up one of the Shure headphone volume controls you showed me - to see exactly what they did inside, at least impedance wise. Then perhaps it could be emulated/replaced/improved with a stepped attenuator circuit of some kind.

The best thing that could happen would be the new product name: "leeperry 100% Passive Headphone Controller".

but he's also selling grossly overpriced passive stepped attenuators...when I can get a nicer looking/more convenient NanoPatch+ for 50 bucks