First kit build classic "47 Headphone Amplifier"

[paulca]

Well this went rather uneventfully.  I bought it because I have found soldering the pin headers onto maker modules very therapeutic, so wanted something with a bit more meat.



It took me about 45 minutes to solder it together and surprisingly to me I was soldering things faster than my 25W Weller could keep up with and I had to pause to let it heat back up.  This is a surprise because as a teenager I couldn't solder to save my life, but now soldering pin headers and ICs seems easy. 

On, "1 thousand", in with the solder, wait till it flows around and into the hole, off.  Next.

Also slightly disappointed that it worked first time.  I gave it 15V from the bench PSU, audio from my Burr Brown headphone DAC amp passthrough and connected my old  Sennheiser phones (I'm not stupid enough to connect my £300 Audio Techniq to something I built!).

It has some odd artefacts which add a kind of wooden, compared to the Burr Brown/Head Box amp, potentially clipping/square wave under tone, but I'm just being picky.  It has enough gain to "pop" the Sennheisers so that's a plus over the Head box amp which is nanny state limited.

[picandmix]

Nice job 

Why don't you upgrade your Sennheisers,  I got these just before Xmas, now even cheaper.

Showing as £169.00 on Sennheisrs own site, not a particularly old model either, got them on the strong recommendation of a friend who bought them when they first came out at nearly £300 !

Sound great to my ears, way better than my trusty old HD470s, plus the over ear design much better.

http://www.argos.co.uk/product/7906808

If you want something more ambitious you could consider this, just waiting for the parts to arrive from RS  myself.
http://www.diyaudio.com/forums/group-buys/302859-xrk971-pocket-class-headamp-gb.html

[paulca]

I currently have these:
https://www.amazon.co.uk/Audio-Technica-ATH-M50X-Monitor-Professional-Headphones/dp/B00HVLUR86/ref=sr_1_1?ie=UTF8&qid=1514812913&sr=8-1&keywords=audio+technica+ath+m50x

They were a bit more expensive when I bought them a few years back.  I put £300 in my original post, but thinking back I believe it was £120 for the Project HeadBox USB amp and £180 for the Audio-technica, so £300 ish all together.

[Zero999]

Well this went rather uneventfully.  I bought it because I have found soldering the pin headers onto maker modules very therapeutic, so wanted something with a bit more meat.

It took me about 45 minutes to solder it together and surprisingly to me I was soldering things faster than my 25W Weller could keep up with and I had to pause to let it heat back up.  This is a surprise because as a teenager I couldn't solder to save my life, but now soldering pin headers and ICs seems easy. 

On, "1 thousand", in with the solder, wait till it flows around and into the hole, off.  Next.

Also slightly disappointed that it worked first time.  I gave it 15V from the bench PSU, audio from my Burr Brown headphone DAC amp passthrough and connected my old  Sennheiser phones (I'm not stupid enough to connect my £300 Audio Techniq to something I built!).

Do you have a schematic for the 47 headphone amplifier? Searching for it using Google, gives a variety of different circuits, many of which require a dual power supply.

It has some odd artefacts which add a kind of wooden, compared to the Burr Brown/Head Box amp, potentially clipping/square wave under tone, but I'm just being picky.  It has enough gain to "pop" the Sennheisers so that's a plus over the Head box amp which is nanny state limited.

Do either of those amplifiers have tone control?

If not, then both amplifiers should sound the same. If either amplifier is affecting the sound, then it's either not designed/built properly or it's being overdriven. Another possibility is cognitive bias: you might believe there's a difference, because you built the one amplifier and not the other, they look different, the price or branding and therefore perceive a difference, when there isn't any.

Fortunately, it's very easy to determine this, with the help of a signal generator, an oscilloscope, some audio connectors and cables. Set both amplifiers to the same gain. Firstly connect one amplifier (doesn't matter which one) to your preferred music source and set the volume a comfortable listening level. Remove the music source and replace it with a 1kHz sine wave from the signal generator (don't put the headphones on when you do this). Look at the output using an oscilloscope and adjust the signal generator's level, until the 'scope shows a  500mV peak to peak sine wave. Swap the amplifier, keep the 'scope and signal generator settings the same and adjust the volume control on the other amplifier, until the 'scope shows a 500mV peak to peak sine wave. If the sine wave appears distorted, then the amplifier is bad.

Now you have both amplifiers set to exactly the same gain, you can see if one really sounds different from the other. Connect each amplifier to a set of input/output extension cables, of the same length, type and colour (they must look identical). Hide the amplifiers in a box. Make it so you don't know which input/output is connected to which amplifier. Now you can try swapping the input and your headphones between the two amplifier, without knowing, which is which. Can you hear any difference? Try tracing the cables back, to see which amplifier you're using. Repeat the test a few times. You might be surprised at how little difference there really is.

I've encountered my own cognitive bias, when testing audio amplifiers before, which is why I rely on my test equipment, more than my ears.

[paulca]

Remove the music source and replace it with a 1kHz sine wave from the signal generator (don't put the headphones on when you do this).
...snip...
I've encountered my own cognitive bias, when testing audio amplifiers before, which is why I rely on my test equipment, more than my ears.

I was testing on a different set of headphones entirely and it might have been overdriven.  You are probably right it's cognitive bias most likely.

On 1KHz test tones.  My brother does professional audio stuff, studio and live and his first mixing desk had it's test tone button directly above the master faders.  This is fine for studio work, but if you accidentally hit the test tone button when hooked up to a 2 kilowatt PA system in a club people start filing damage claims against you for hearing damage.  1 Khz pure sine wave at full volume through a gig rig would be incredibly painful.

He tried various ways of covering it up and eventually opened the mixer and removed the switch entirely.

Dumb place to put it.  It should be well out of the way where you are unlikely to push it accidentally.

[paulca]

Do you have a schematic for the 47 headphone amplifier? Searching for it using Google, gives a variety of different circuits, many of which require a dual power supply.

Best I can do, this came with it:


It seems to be a standard textbook stereo 2 stage amplifier with virtual ground and AC coupling.

It might be improved by buffering the virutal ground with a buffer amp, but meh!

[paulca]

So I wired it up to my proper headphones, after listening to the HeadBox all evening.  Straight amp swap and to be honest I can't tell the difference.  The 47 amp has a wedge more gain which makes it so much more appealing

Now.  A project springs to mind.  Adding a USB DAC to the 47.  It turns out I just gave my brother a PCM2702 sound card module.  I think I might order another of those. 

The only other thing I need is an impedence buffer to split the DAC line level output to an RCA Phono pair @ line level + the input to the 47 amp.

It has been bugging me for a while that I have to choose digital DAC and headphones only or Analogue with headphones and pass through for speakers.

Now how does one make an impedance balanced Y split for audio signals.  Google time.

[paulca]

Now how does one make an impedance balanced Y split for audio signals.  Google time.

So it seems as I expected I can accept the slight gain drop just by paralleling the 2 outputs or I can parallel and buffer with unity gain op amps with impedance resistors to get 600Ohm output impedance on both outputs with minimal gain drop.  I'd need another virtual ground supply to power the splitter though.  Might be worth it, I dunno.  Gain lost is gain you have to add back + noise.  Best to save it where you can.  On teh flip side each op amp and circuitry stage has the potential to add noise.

[David Hess]

It seems to be a standard textbook stereo 2 stage amplifier with virtual ground and AC coupling.

The second operational amplifier of each pair is being used to double the output current capability.  Not counting what is inside the operational amplifiers, it is really only one stage per side.

[paulca]

It seems to be a standard textbook stereo 2 stage amplifier with virtual ground and AC coupling.

The second operational amplifier of each pair is being used to double the output current capability.  Not counting what is inside the operational amplifiers, it is really only one stage per side.

Ah, yes, I see that now.  A unit gain op amp in parallel on the output, so it doesn't amplify just mirrors the actual amp but shares the current requirement on the output.  Nice idea.

[tooki]

I currently have these:
https://www.amazon.co.uk/Audio-Technica-ATH-M50X-Monitor-Professional-Headphones/dp/B00HVLUR86/ref=sr_1_1?ie=UTF8&qid=1514812913&sr=8-1&keywords=audio+technica+ath+m50x

They were a bit more expensive when I bought them a few years back.  I put £300 in my original post, but thinking back I believe it was £120 for the Project HeadBox USB amp and £180 for the Audio-technica, so £300 ish all together.

I have those cans, too. Great headphones! So far they’ve put up with all the homemade audio amps I’ve hooked them up to! :p

[Zero999]

It seems to be a standard textbook stereo 2 stage amplifier with virtual ground and AC coupling.

The second operational amplifier of each pair is being used to double the output current capability.  Not counting what is inside the operational amplifiers, it is really only one stage per side.

Ah, yes, I see that now.  A unit gain op amp in parallel on the output, so it doesn't amplify just mirrors the actual amp but shares the current requirement on the output.  Nice idea.

Yes, also note that R8 to R11 are there just to help the amplifiers current share and do not determine the output impedance of the amplifier, which will be much lower than two 47R resistors in parallel. Negative feedback will ensure V_OUT = (1+10/4.7)V_IN.

Speaking of output impedance, if you look at the schematic you'll see it has 15R resistors on the output, which act as current sensors for the output current limiting circuit. The op-amp will have an output impedance of 15R, when open loop, but if you look at the Electrical Characteristic table of page 5, you'll see the output impedance is only 0.3R, with a gain of 30dB.
http://www.ti.com/lit/ds/symlink/ne5532.pdf

[Lightages]

I currently have these:
https://www.amazon.co.uk/Audio-Technica-ATH-M50X-Monitor-Professional-Headphones/dp/B00HVLUR86/ref=sr_1_1?ie=UTF8&qid=1514812913&sr=8-1&keywords=audio+technica+ath+m50x

They were a bit more expensive when I bought them a few years back.  I put £300 in my original post, but thinking back I believe it was £120 for the Project HeadBox USB amp and £180 for the Audio-technica, so £300 ish all together.

I have those cans, too. Great headphones! So far they’ve put up with all the homemade audio amps I’ve hooked them up to! :p

IMHO, the ATH-M40X are better headphones and a lower price. Also IMHO, if you can hear a difference between amps and it is real, there is something seriously wrong with the design or implementation of the design of the amp. Any difference between amps should be buried in the mess of the transducer.

[paulca]

Yes, also note that R8 to R11 are there just to help the amplifiers current share and do not determine the output impedance of the amplifier, which will be much lower than two 47R resistors in parallel. Negative feedback will ensure V_OUT = (1+10/4.7)V_IN.

Speaking of output impedance, if you look at the schematic you'll see it has 15R resistors on the output, which act as current sensors for the output current limiting circuit. The op-amp will have an output impedance of 15R, when open loop, but if you look at the Electrical Characteristic table of page 5, you'll see the output impedance is only 0.3R, with a gain of 30dB.
http://www.ti.com/lit/ds/symlink/ne5532.pdf

V_OUT = (1+10/4.7)V_IN.

So correct me if I'm wrong here, but given a "normal" ball park of 500mV peak to peak on a line level input.  Taking the half wave for current (to avoid horrible trig) gives 250mV.  My headphones are 38 Ohm.

(.250 * (1+10/4.7)) / 38
.020

0.020A max current.  Which means even with the current sharing pair they are right on the current limit for the op amp which is 10mA.

I don't see the 15R resistors on the schematic.  Are you referring to a different version?

I found it will distort when driven hard, but this is "normal", though hard to tell if it's the amp or the phones, though I could probably test that with a scope and a 38Ohm resistor.  I put "normal" in quotes as I find most amps I have experienced in the last 10 years tend to limit long before you get to the point of clipping the power amp or headphones.  The gain on my soundcard differs between Windows and Linux.  I believe there are two interruptions of the gain structure between what Intel say it 'should' be run at and what they say it 'can' be run at.  Linux goes for the conservative and arguable correct gain where are windows driver nails it hard to MAX.  It's rather annoying balancing your output levels when you dual boot.  Tests so far where in Linux.  Presently windows volume never makes it above 50% and usually spends it's time around 30-40%.  So I don't get deafened or blow my speakers when I reboot.

[Zero999]

Yes, also note that R8 to R11 are there just to help the amplifiers current share and do not determine the output impedance of the amplifier, which will be much lower than two 47R resistors in parallel. Negative feedback will ensure V_OUT = (1+10/4.7)V_IN.

Speaking of output impedance, if you look at the schematic you'll see it has 15R resistors on the output, which act as current sensors for the output current limiting circuit. The op-amp will have an output impedance of 15R, when open loop, but if you look at the Electrical Characteristic table of page 5, you'll see the output impedance is only 0.3R, with a gain of 30dB.
http://www.ti.com/lit/ds/symlink/ne5532.pdf

V_OUT = (1+10/4.7)V_IN.

So correct me if I'm wrong here, but given a "normal" ball park of 500mV peak to peak on a line level input.  Taking the half wave for current (to avoid horrible trig) gives 250mV.  My headphones are 38 Ohm.

(.250 * (1+10/4.7)) / 38
.020

0.020A max current.  Which means even with the current sharing pair they are right on the current limit for the op amp which is 10mA.

Yes that's true. In your case, using two op-amps in parallel is not necessary, but there may be less sensitive headphones which require more drive.

I don't see the 15R resistors on the schematic.  Are you referring to a different version?

Sorry, I didn't make it clear that the 15R resistors I was referring to, are inside the NE5532 and are part of the op-amp's internal current limiting circuitry. The point I was making was that, the open loop output impedance is 15 Ohms, yet closed loop, with a gain of 30dB it's 0.3 Ohms. As long as the amplifier's closed loop gain is low compared to the open loop gain, the current isn't too high and the supply voltage is high enough, the output impedance will still be low, in spite of the added series resistors on output. Note that the feedback node must be after the resistor(s), not before, otherwise they will dominate the output impedance.

I found it will distort when driven hard, but this is "normal", though hard to tell if it's the amp or the phones, though I could probably test that with a scope and a 38Ohm resistor.  I put "normal" in quotes as I find most amps I have experienced in the last 10 years tend to limit long before you get to the point of clipping the power amp or headphones.  The gain on my soundcard differs between Windows and Linux.  I believe there are two interruptions of the gain structure between what Intel say it 'should' be run at and what they say it 'can' be run at.  Linux goes for the conservative and arguable correct gain where are windows driver nails it hard to MAX.  It's rather annoying balancing your output levels when you dual boot.  Tests so far where in Linux.  Presently windows volume never makes it above 50% and usually spends it's time around 30-40%.  So I don't get deafened or blow my speakers when I reboot.

Yes, that's true, it's difficult to tell whether it's the headphones or the amplifier which are responsible for the distortion. The impedance of the headphones might vary considerably, at different frequencies, so it would be more accurate to leave them plugged in and monitor the output with a 'scope.

[paulca]

Yes, that's true, it's difficult to tell whether it's the headphones or the amplifier which are responsible for the distortion. The impedance of the headphones might vary considerably, at different frequencies, so it would be more accurate to leave them plugged in and monitor the output with a 'scope.

Hmm.  If you clip the headphones by pushing the coil to it's limit or hitting the elastic limit of the diaphragm reflex would that then cause a reaction in the impedance and register on the scope anyway?  I was thinking of the coil bouncing back off the stops actually generating current.

[Zero999]

Yes, that's true, it's difficult to tell whether it's the headphones or the amplifier which are responsible for the distortion. The impedance of the headphones might vary considerably, at different frequencies, so it would be more accurate to leave them plugged in and monitor the output with a 'scope.

Hmm.  If you clip the headphones by pushing the coil to it's limit or hitting the elastic limit of the diaphragm reflex would that then cause a reaction in the impedance and register on the scope anyway?  I was thinking of the coil bouncing back off the stops actually generating current.

Yes, that's true. The impedance will become non-linear when the headphones are driven into clipping. Overheating and destruction of the headphones, is another problem with over-driving, but I've no idea whether it will occur before or after the clipping point. I suppose it depends on the construction of the headphones, the temperature, the extent of the overdriving and how long it occurs for.

[paulca]

Overheating and destruction of the headphones, is another problem with over-driving, but I've no idea whether it will occur before or after the clipping point. I suppose it depends on the construction of the headphones, the temperature, the extent of the overdriving and how long it occurs for.

It is a worry, especially with expensive headphones.  Different headphones behave differently.

* £20 Techniq headphones - Drove great into slightly distortion, "CLACKED" when they hit their physical limits and blew shortly after.  If you heard clacking you had seconds to find a volume control, but they were loud, cheap and didn't sound that bad.  I killed 3 pairs. 

* £10 Panasonic headphones - Distorted early, sounded horribly distorted at even higher levels, there was no point running them hard enough to kill them, but I did manage to blow a set.

* Ancient Tanoy headphones from the 70s I got with a TRIO amp.... my ears would bleed before I managed to pop them.  In the end they died because the TRIO amp died and melted the headphone coils.

* Sennheiser cheapo ~£100 phones, clacked hard and sounded hideous when driven too hard.  Kind of like a slap on the wrists that made you back the volume out cause the noise was unpleasant.  It's like a built in physiological limiter.  I never did blow them, the head band broke on one side after about 5 years.  They still work.

* Audio techniqa phones, this amp is the first one to drive them into distortion and I don't want to blow £120+ of headphones, no need they are loud enough, so I always back off when they start to sound dirty and clippy.  They don't clack or protest harshly though.  That worries me, they may struggle on trying to sound as good as possibly until they melt.

I did have a pair of DJ headphones I bought after researching "LOUD" headphones, but I was disappointed when they were incredibly high impedence and the loudest headphone amp I had didn't tax them too much.  I was told if I drove them with a proper professional headphone amp they would take huge volumes.  They always sounded underdriven and weak to me, so I bought the Sennheisers instead.

[Zero999]

Overheating and destruction of the headphones, is another problem with over-driving, but I've no idea whether it will occur before or after the clipping point. I suppose it depends on the construction of the headphones, the temperature, the extent of the overdriving and how long it occurs for.

It is a worry, especially with expensive headphones.  Different headphones behave differently.

* £20 Techniq headphones - Drove great into slightly distortion, "CLACKED" when they hit their physical limits and blew shortly after.  If you heard clacking you had seconds to find a volume control, but they were loud, cheap and didn't sound that bad.  I killed 3 pairs. 

* £10 Panasonic headphones - Distorted early, sounded horribly distorted at even higher levels, there was no point running them hard enough to kill them, but I did manage to blow a set.

* Ancient Tanoy headphones from the 70s I got with a TRIO amp.... my ears would bleed before I managed to pop them.  In the end they died because the TRIO amp died and melted the headphone coils.

* Sennheiser cheapo ~£100 phones, clacked hard and sounded hideous when driven too hard.  Kind of like a slap on the wrists that made you back the volume out cause the noise was unpleasant.  It's like a built in physiological limiter.  I never did blow them, the head band broke on one side after about 5 years.  They still work.

* Audio techniqa phones, this amp is the first one to drive them into distortion and I don't want to blow £120+ of headphones, no need they are loud enough, so I always back off when they start to sound dirty and clippy.  They don't clack or protest harshly though.  That worries me, they may struggle on trying to sound as good as possibly until they melt.

I did have a pair of DJ headphones I bought after researching "LOUD" headphones, but I was disappointed when they were incredibly high impedence and the loudest headphone amp I had didn't tax them too much.  I was told if I drove them with a proper professional headphone amp they would take huge volumes.  They always sounded underdriven and weak to me, so I bought the Sennheisers instead.

Are you hard of hearing? Perhaps listening to loud music through headphones has damaged your hearing?

I've never managed to make headphones distort, even a really cheap pair from Poundland, because my ears give up first and I have to turn the volume down. Decent headphone transducers are fairly cheap to produce. Headphones generally sound better than speakers because: the transducer is light so can respond quickly, the ear is in the near field region so there are no reflections and standing waves. Beyond a certain price you're paying for the comfort and robustness and beyond that you're just paying for the name. I don't think the Poundland headphones sounded that bad but they broke because they were cheap, weak crap. Personally I prefer speakers because I find sound directly going into my ears unnatural, so I don't use headphones much, even though I know they produce better sound.

[paulca]

I probably have damaged my hearing with headphones over the years, but I'm not hard of hearing... yet. "What?"

When I stopped being a teenager (mentally it's debatable that has actually happened) I moved from playing loud music on the HiFi, annoying the neighbourhood and playing it instead either in the car or on headphones.  So I can listen as loud as I like and not bother anyone... Except at traffic lights and meh, tough.

I like energetic music with attitude that needs power and heavy bass to be properly experienced, slight distortion can actually be an advantage as long as you don't clip the bass.  Rock, thrash metal and acid, trance, electro, house, dance stuff.  Stuff where it's natural habitat is a club or a gig where you expect to have whistling ears for at least 24 hours afterwards.

[paulca]

If you want to test your headphones for bass, try this:


At 2:15 to 2:28 is the real test.  There is a constant bass line playing but very, very few systems will present it.  I figure it drops to something like 5-10Hz.  Not much higher frequency than  a shiver.

[tooki]

IMHO, the ATH-M40X are better headphones and a lower price.

Really? I bought the ATH-M50x based on great reviews, but had never heard them before. And I do love them. How are the M40X better? (The $50 price difference was no big deal to me, and I got mine on sale anyway for a really, really good price.)

Also IMHO, if you can hear a difference between amps and it is real, there is something seriously wrong with the design or implementation of the design of the amp. Any difference between amps should be buried in the mess of the transducer.

I've been saying this for years, both with regards to headphones and speakers. It's a lot harder (relatively speaking) to make a good transducer than a good amp.

[tooki]

At 2:15 to 2:28 is the real test.  There is a constant bass line playing but very, very few systems will present it.  I figure it drops to something like 5-10Hz.  Not much higher frequency than  a shiver.

I hear the deep sound that I assume you're referring to. But using an online signal generator for comparison, I think it's only about 20Hz.

[paulca]

IMHO, the ATH-M40X are better headphones and a lower price.

Really? I bought the ATH-M50x based on great reviews, but had never heard them before. And I do love them. How are the M40X better? (The $50 price difference was no big deal to me, and I got mine on sale anyway for a really, really good price.)

When I was researching and buying mine I recall there being review articles that mentioned the M40X was as good and a bit cheaper.  However I couldn't find the M40X but found the M50 on sale.

[paulca]

At 2:15 to 2:28 is the real test.  There is a constant bass line playing but very, very few systems will present it.  I figure it drops to something like 5-10Hz.  Not much higher frequency than  a shiver.

I hear the deep sound that I assume you're referring to. But using an online signal generator for comparison, I think it's only about 20Hz.

The tune over that quiet portion goes "High...Low...........Very low and hold....... High...Low..... Very low..." and back to the beat.  The "Low" is probably about 20 Hz, but the very low, which is barely audible and is more just hearing the speaker moving I can tap my finger on the desk with the speaker deflections I'm hearing.  Unfortunately you can also here the sample aliasing and 'squeaking' of the mp3 encoding over it.

If I remember next time I have the scope out, I'll test it.

[tooki]

No, it’s the very low that’s around 20Hz, according to my comparisons with the signal gen.

[Lightages]

IMHO, the ATH-M40X are better headphones and a lower price.

Really? I bought the ATH-M50x based on great reviews, but had never heard them before. And I do love them. How are the M40X better? (The $50 price difference was no big deal to me, and I got mine on sale anyway for a really, really good price.)

The 40s have a flatter response. The 50s also have a strange phase shift in the lower treble. No headphone is perfect and each person will hear headphones a bit different because of the shape of the ear called the pina. Having said that, most people think the 40s sound better.

[tooki]

IMHO, the ATH-M40X are better headphones and a lower price.

Really? I bought the ATH-M50x based on great reviews, but had never heard them before. And I do love them. How are the M40X better? (The $50 price difference was no big deal to me, and I got mine on sale anyway for a really, really good price.)

The 40s have a flatter response. The 50s also have a strange phase shift in the lower treble. No headphone is perfect and each person will hear headphones a bit different because of the shape of the ear called the pina. Having said that, most people think the 40s sound better.

Not that I’m looking to replace my M50x, but I would be curious to compare them now! From what you’ve said (other than the phase shift, which I haven’t found other mention of) and other reviews and comparisons say, the M50x is probably perfect for me, in having ever so slightly boosted bass and prominent highs, in that I generally like slightly bright sound. (Though that said, the speakers and cans I like are all fairly flat, so it’s probably more accurate to say that I dislike the heavy bass boost of most mass-market transducers.)

My absolute favorite headphones so far are my Beyerdynamic MMX300, which are basically a DT770 Pro 32-ohm in a different housing with a boom mike. Unlike every other headphone or speaker I’ve ever owned, there is zero acclimation time when putting them on. (Even the M50x and my Ety ER4XR take a moment of listening to sound absolutely colorless.) The MMX300 is just instantly... perfect. (For my taste, of course.)

Heh, yes, I’m actually familiar with the word “pinna” (and used it in a FB discussion just the other day), on account of me having several large piercings in each of mine! :p

[paulca]

So I got the DAC soundcard module back off of my brother, so I can mod the 47 amp to be a USB digital amp.

I have a couple of 3.5mm breakout boards for easy soldering, so I might rig up a splitter so the DAC can provide input to both the headphone amp and the powered desktop speakers.

[picandmix]

Have tried those op amp headphone amps using 4556 chips, reportedly much better then the 5532, but still nowhere are good as the headphone output of my Arcam amp.

Just completed a discreet component build to that headphone amp I linked earlier to prove its worth and the difference is vast, its at least equal to the Arcam, possibly better.

The essence of that circuit is you can build as standard version as I did or,  build selecting  the parts to deliver very good channel balance and  select the components to give optimum performance with you particular cans impedance, high, low or medium, yet all done by a relatively simple low cost circuit.

Well worth a try.

While I did try that Star Wars theme,  and its a good warm up for the speakers, have you tried something a little more challenging over a wider spectrum  like this ?


Also you should a get much better sound if you download it and playback with something like FooBar2k