The DIY project headphone amp Objective2 ("O2") you mentioned is excellent and very well documented. Alas not as simple as I hoped because of its additional power-supply and protection circuits. But I'm intrigued to build the O2 once I can easily purchase a printed circuit board (PCB) around here.

cheers

The aforementioned FiiO E5 is a good example of a very cheap amp with a very low output impedance - about $20 with 0.7 ohms impedance.  The E6 is slightly improved. (note that it allows a higher input level before clipping, too).  The E9 has more power & voltage, but has a 10 ohm impedance, which is still low, but not as low. The E9 can be docked with the E7 or E17, which are both combo DAC/amps. The E7 has a lower output impedance, and I presume the E17 does too. (the E17 is very new, and has not been reviewed yet by NwAvGuy).  The E9 would be best suited to high impedance headphones, with the E7/E17 being more suited to low impedance headphones. If I didn't already have a headphone amp, I think I would have bought an E9 and possibly an E7. (but now I'd get the E17). The amp I have is the Musical Fidelity V-Can, which has an output impedance of 5 ohms. It's a bit pricey for what it does IMHO, considering the competition.  It's output drive (~3V RMS) is borderline for solo piano, with my AKG K601 headphones.  I think it's enough for typical music though.

That blogger designed his own amp - the O2. It appears to be a very simple op-amp design, but using good quality components. It has very good specs. (including a very low output impedance) You can build it yourself, or buy it pre-assembled  from JDSLabs.

Greg.

I did so now, and yes, it sounds differently with my AudioTechnica. With the Millenium HP-1 it sounds a little darker and slightly more bassier. But take this with a grain of salt, because didn't level it out very exactly, and I'm sure you know that would be very important for A/B-testing. Eventually I repeat it with a level exactly measured. But I'm quite sure about a sound-difference already.

...both full ack!

You made me curious, so I measured the voltage-drop of the HP-1 also when loaded with a 100 Ohm resistor. Measured open my testsignal was 1V at 400Hz. Loaded with 100 Ohm I found a drop to 0.667V. With the formula this means 50 Ohm output-impedance of the Millenium HP-1 headphone amp. So I can confirm the user-review in your link, thanks. When I bought it, I thought its output-impedance would be nearer at null, because it is advertised to drive low impedances down to 8 Ohm. It would be interesting to know, which headphone-amps have less than 50 Ohm and are not too expensive.

cheers

For the mixer, I would connect the headphone amp to a line out, for the very best fidelity. (I realise you were connecting to the headphone output for diagnostic purposes though)

Just btw, if this review is correct, that amp has an output impedance of 50 ohms, which is worse than your notebook: http://en.audiofanzine.com/headphone-am...15475.html  ;^)
If you haven't already, you might want to measure it's output impedance yourself.

Greg.

If it were me, I'd buy a headphone amp for  the mixer, if I had to use the mixer with headphones.  The output impedance is too high for low impedance headphones (including 40 ohm headphones), [...]

Just for interest I tried the headphone-amp Millenium HP1 between mixer-headphone-out (100 Ohms) and two different headphones (60 Ohms and 300 Ohms). Indeed it clears things up and the left hand has more definition and detail. The other side of the medal is, that two audio-device in row (mixing-console and headphone-amp) suck tone also,  more than using the headphone with the notebook-headphone-out (25 Ohm) directly. And it is one gadget and power-adaptor more.

In my opinion the best solution is to use the low impedance output of the notebook directly, when playing with headphones. If a mixing console is used and the center of your musical environment, a headphone-amp helps when having high-impedance outputs with 100 Ohms. Eventually it is worth not to use the cheapest headphone-amps on the market.

cheers

If it's true that the Beyerdynamic range is optimised for a 100/120 ohm impedance, then I suppose they would be a good match to this mixer.

Greg.

If it were me, I'd buy a headphone amp for  the mixer, if I had to use the mixer with headphones.  The output impedance is too high for low impedance headphones (including 40 ohm headphones), but the output level is also too low for high impedance headphones. (75mW into 40 ohms = 1.7V RMS loaded, or 2.4V RMS unloaded. I'd prefer to see an unloaded level of at least 3V RMS, preferably even higher)   If  you really don't want to get an amp, perhaps you could buy some headphones for which you know the impedance vs frequency curve, and choose headphones with a relatively flat curve. http://www.headphone.com have the impedance vs frequency graphs for many different models. 

Greg.

Or I leave everything as it was at the beginning, using the notebook-output directly with my 60 Ohm headphone and be happy!

- groovy

Here are some more links which I think are very informative:
NwAvGuy: Headphone & Amp Impedance
Sound & Vision:Output Impedance: A Show and Tell
Benchmark Media:The "0-Ohm" Headphone Amplifier: The Sonic Advantages of Low-Impedance Headphone Amplifiers

Greg.
EDITED 11-Jan-16 to update link for the Benchmark Media article.

Before buying a headphone amp, perhaps you could try adding 75 ohms of impedance in series with the laptop output, to bring the total output impedance up to 100 ohms, to match the output impedance of the mixer. See whether it makes the sound closer to the sound from the mixer, just as an experiment. (this will drop the maximum volume, it's important to make tonal comparisons at the same volume)

Crazy, that's it! The difference between 25 Ohm output-impedance of the notebook and the 100 Ohm of the Yamaha mixer determines the tonal characteristic with this headphone Audio-Technica ATH-M40fs! Never thought, headphones are so picky with the headphone-output.

Although I had not 75 Ohms resistors you suggested, I used my old adapater with 47 Ohm trimmers in row, that I soldered in the past for another experiment. When this resistance is between the notebook and the headphones, the sound degrades similar to what I hear when these headphones are attached to the Yamaha headphone-out. Exaggerated attributes could be then: more muddy, less life, less breath, less natural. (off course I compensated the volume loss by the added 47 Ohms). This effect can't be simulated with treble or bass knob of channel 7/8 the same.

- groovy

Or just use the tone controls and move on, as you said.

Greg.

FiiO seem to have really good value stuff.  Here's a technical review of the E5 that Michael mentioned: http://nwavguy.blogspot.com.au/2011/03/...e-amp.html
Points to note:
a) it seems to be designed for low output portable devices, because it has a maximum input level of 760mV RMS.
b) it is more suited to low impedance headphones, because it has a rather low maximum output level of 1.3V RMS.
c) It has a very good (low) output impedance of 0.7 ohms. 

For high impedance headphones, their E9 would be more appropriate, although unfortunately it's 10 ohm output impedance may be a problem with some low impedance headphones. (especially some in-ear monitors, I have read)

Note that it is not always correct to aim for the lowest possible output impedance. For one example, Beyerdynamic seem to optimise their headphones for use with a 120 ohm output impedance (120 ohms is a standard), and this is backed up by the fact that their own A1 headphone amp has an output impedance of 100 ohms. (close enough to 120 - not sure why they used 100 instead of 120 though) It's a bit of a murky area. When in doubt, contact the headphone manufacturer and ask them.

Greg.

- groovy

Anyway, assuming your meter is correctly reading at 400Hz, your internal resistance is actually 25 ohms.   Check: Vout = Rl/(Rl+Ri)*Vint = 100/(100+25)*1 = 0.8V RMS = correct. ;^)

I agree that even if your output impedance was 120 ohms, that would have been fine for the line inputs of the mixer. 

How long is the cable that you are using between the laptop and the mixer? Keep it as short as possible, anyhow.

Yes, 60 ohms is definitely "low" compared to 600 ohms!!!

1V RMS is not too bad. As I said, my USB interface has this level as well.  I really do not think this will be a problem, unless your mixer has a lousy signal-to-noise ratio.

Greg.

Michael