Review and Measurements of New JDS Labs Atom Headphone Amp

[trl]

I've designed a headphone amplifier using OPA1612 as a gain stage and using OPA1622 as output buffers. Each channel has 4 outputs summoned, thus lowering the already really low noise by around -6 dB. I use TPS7A4700 and TPS7A33 low noise DC to DC converters with 4 µV/16 µV noise. Only thin film 0.1% 10 PPM resistors in the audio path. I'm currently working on my last revision with further improvements in the power supply.
[...]

Can you please find a scope to measure this beautiful headamp with sine & square waves at 20Hz, 1KHz, 10KHz, 15KHz and 20KHz? You could you the built-in signal generator from the scope, if available or you could use a very good DAC and audiocheck.net or similar website to generate the correct sines and squares.

It's very important to see when the output buffer starts to distort (to bend the sines or to flatten the output signal) and if they get overheated.

Also, Q401 is one of the best devices to test this headamp, right?

 

[amirm]

Thank you, Sir!
Would you mind, please, instead of testing 1 KHz sines to try 10 or 15 KHz sines phase tests, please (at least for future amps)? Usually at 1 KHz things are perfect on most amps. People complains about bass roll-off or trebles sounding too upfront or in-your-face etc., so wondering if phase-shift/delay might have something to do with that or not.

Here is the interchannel phase difference from 20 to 20 kHz:

It is well below 0.1 degree. So pretty benign.

 

[amirm]

As promised, here is the comparison of JDS Labs Atom against Massdrop THX AAA headphone amplifiers. Since the atom does not have balanced input, I tested the Massdrop THX using unbalanced which degraded its performance due to mains leakage.

The test is at 300 ohm. First, atom in low power mode versus gain 2 on THX:

We have identical power but the JDS labs has less noise/mains leakage.

Here is the performance in high-gain/gain 3:

Here, the Atom pulls ahead with more power with the trade off of slightly higher distortion (but still incredibly low).

Of course the THX has its balanced headphone jack so if you use that, it will produce more power.

Summary
The JDS Labs Atom is actually superior if your source is unbalanced as is your headphone connection at 300 ohm (e.g. Sennheiser HD-650 type headphone).

For best performance, you really need to use balanced inputs on Massdrop THX as I did in my review of that unit. Its switching supply likely has EMI filtering caps that bleed some amount of noise into their output, impacting unbalanced connection.

 

[wadec22]

As promised, here is the comparison of JDS Labs Atom against Massdrop THX AAA headphone amplifiers. Since the atom does not have balanced input, I tested the Massdrop THX using unbalanced which degraded its performance due to mains leakage.

The test is at 300 ohm. First, atom in low power mode versus gain 2 on THX:
View attachment 17646

We have identical power but the JDS labs has less noise/mains leakage.

Here is the performance in high-gain/gain 3:
View attachment 17647

Here, the Atom pulls ahead with more power with the trade off of slightly higher distortion (but still incredibly low).

Of course the THX has its balanced headphone jack so if you use that, it will produce more power.

Summary
The JDS Labs Atom is actually superior if your source is unbalanced as is your headphone connection at 300 ohm (e.g. Sennheiser HD-650 type headphone).

For best performance, you really need to use balanced inputs on Massdrop THX as I did in my review of that unit. Its switching supply likely has EMI filtering caps that bleed some amount of noise into their output, impacting unbalanced connection.

wow - thanks for doing that comparison. using the balanced inputs on the thx - does that only net you more power or do you get more power and less noise. how big does the gap between the two products become when using xlr on the thx?

 

[amirm]

wow - thanks for doing that comparison. using the balanced inputs on the thx - does that only net you more power or do you get more power and less noise.

Balanced *input* does not change the power. It just rids the unit of mains noise, and thus improve its performance.

To get more power you have to use its headphone/XLR balanced output.

 

[wadec22]

Balanced *input* does not change the power. It just rids the unit of mains noise, and thus improve its performance.

To get more power you have to use its headphone/XLR balanced output.

got it, yes that makes sense. how big does the gap become between the two when you go balanced input on the thx? significant?

 

[helloworld]

wow, I kind of regret buying 789......how better using the balanced in of 789 compare to unbalanced in of atom?

 

[maxxevv]

If you have a DAC like the SMSL SU-8 that has XLR outputs to Amp, then you'll maximise the abilities of the THX. Which from the measurements, at higher output, does measure better.

Its just like higher end options upstream will help with the downstream performance in most audio gear.

Buying the THX just gives you more options later should you upgrade / update your DAC. Its still an outstanding amp regardless of price. Headphone amps are one of those audio items that can remain unchanged for a full decade and it won't get out of date as long as its base performance is up to par. Its usually the DAC portion that gets superceded by newer stuff really fast.

So don't fret !!!

 

[trl]

Here is the interchannel phase difference from 20 to 20 kHz:

[...]

It is well below 0.1 degree. So pretty benign.

Looks perfect to me, many thanks!

 

[trl]

As promised, here is the comparison of JDS Labs Atom against Massdrop THX AAA headphone amplifiers. Since the atom does not have balanced input, I tested the Massdrop THX using unbalanced which degraded its performance due to mains leakage.
[...]

I assume @33Ohms THX will win, right? Or will it be tie?
I wonder what buffer THX has inside.

 

[gvl]

Turns out I got cold feet on the 789 for a good reason! Variable output and more compact form factor compatible with the mess on my desk at 1/3 the price. Do I hear jingle bells in November?

 

[mshenay]

Some more subjective stuff for you guys, with some points of reference


20181214015136_IMG_0130.jpg1500x1000



20181214015200_IMG_0131.jpg1500x1000

Running my modded Grado with a Soft inner word Hard outer word Sleved over top, like a PS Series but wood built around a Magnum V7 driver and dual sleeved to help make it… not so humpy in the lows. Most of the time I strongly prefer this headphone with either my modded SET or my Vali 1. I almost never listen to it on solid state with the exception of my HM601 which… sounds like syrup.

However line out from the 601 into the Atom is quite nice! It really doesn’t sound like it’s there, that is to say it’s quite transparent. So where as the Vali 1 and even the hm601s crapola opAmp sound hazy, with the Atom there’s clarity, so no haze but there’s still a very natural timbre but without any noise If that makes sense. To further elaborate, I would NEVER feed the HM601 into the Vali 1. That would just be too messy, too… soft and “romantic” no thanks. Rather I feed straight from either my EL Dac or most of the time my NFB10ES2 reference system.

How ever with the ATOM & HM601 I get a lot of similarities in tonality and detail even though the staging and imaging aren’t quite as impressive. My point being, the amp neither adds nor takes away. It reminds me a lot of my Head Amp Pico Power.

And to a lesser degree my iBasso PB2 which runs LME 49900s but in a fully balanced configuration. Now Switching to something nicer like a Soundaware P1, the P1 sounds… equally “not there” but there’s detail the P1 pulls out that I don’t get from the Atom, very fine transient information and staging cues that are just simply not audibly present with the Atom.

So I imagine, scaling up from the Atom MicroDetail/Transient response/imaging or staging cues and even “plakton” are what a nicer amp will bring assuming your DACs not the bottle neck.

An to further elaborate, while I feel the iBasso PB2 is tonally neutral, it’s not entirely transparent. The iBasso in particular is… very authoritative. It typically brings an aggressiveness into what ever system I have it in. An the Soundaware P1… does but doesn’t, I find the P1 again more clearly defines transient response, and even pulls out detail that the PB2 doesn’t. So in a sense it’s equally as impactful but without the perception of being “aggressive” that is to say it’s Transparent!

 

[jackenhack]

To tackle the distortion issue, use a passive attenuator to get input level down. The QA401 has -20 dB built right in, so trying that for starters seems like a no-brainer. Otherwise, you can always build one for yourself:

Yeah, I have a signal generator with a THD of somewhere around -140 dB THD and I've built a Twin-T notch filter that is very sharp at 1khz, so I can put that after the amp and get better resolution (although manual calculations) from the measurement ADC.

If you look at the PCB, it's from a year ago. Total overkill on the capacitors. I've redesigned the power supply and a lot of other stuff, so the current board has fewer components. Waiting to order the (hopefully) last revision of the PCB and if everything works out, I'll send one to @amirm if he wants one.

 

[jackenhack]

Incidentally, are there any series resistors in your input circuitry? If so, I would suggest shuffling things around so that the trace following the resistor becomes a whole lot shorter. This way the longest part of this rather long trace would see source output impedance only.

The last couple of revisions have the whole gain stage moved close to the input to increase the signal level to max as soon as possible to make it less sensitive to induced noise. So there's a lot of new design changes on it.

 

[pos]

As promised, here is the comparison of JDS Labs Atom against Massdrop THX AAA headphone amplifiers. Since the atom does not have balanced input, I tested the Massdrop THX using unbalanced which degraded its performance due to mains leakage.

The test is at 300 ohm. First, atom in low power mode versus gain 2 on THX:
View attachment 17646

We have identical power but the JDS labs has less noise/mains leakage.

Here is the performance in high-gain/gain 3:
View attachment 17647

Here, the Atom pulls ahead with more power with the trade off of slightly higher distortion (but still incredibly low).

Of course the THX has its balanced headphone jack so if you use that, it will produce more power.

Summary
The JDS Labs Atom is actually superior if your source is unbalanced as is your headphone connection at 300 ohm (e.g. Sennheiser HD-650 type headphone).

For best performance, you really need to use balanced inputs on Massdrop THX as I did in my review of that unit. Its switching supply likely has EMI filtering caps that bleed some amount of noise into their output, impacting unbalanced connection.

It would be fun to bridge this amp with a XLR to 2xRCA cable on input, and stereo jack to XLR-4 on output.
A pair of these amps is still "only" $200

 

[helloworld]

Hopefully they are going to make a pro version of this atom, with balanced xlr in and 4pin xlr out and 2x 3xlr out

 

[Sir Sanders Zingmore]

USD 37 to ship to Australia

 

[andreasmaaan]

USD 37 to ship to Australia

As of 1 July, I believe you may also now have to pay GST on the cost of the unit + shipping

 

[mshenay]

Hopefully they are going to make a pro version of this atom, with balanced xlr in and 4pin xlr out and 2x 3xlr out

I've been asking for a fully balanced design from JDS Labs for a while, so fingers crossed!!!

 

[helloworld]

I've been asking for a fully balanced design from JDS Labs for a while, so fingers crossed!!!

So they are going to release a balanced design?