Review and Measurements of New JDS Labs Atom Headphone Amp

[magicscreen]

Why are you using the Khadas Tone Board for the Atom (the transparent amplifier)?
Is the IMD hump not a big audible problem?

 

[Veri]

Is the IMD hump not a big audible problem?

Most likely it is inaudible. Hard to say. Suffice to say it is not a huge problem, more like a minor defect.

A lot of people here use KTB+Atom so no, it's actually a great combo

 

[hyperknot]

Why are you using the Khadas Tone Board for the Atom (the transparent amplifier)?
Is the IMD hump not a big audible problem?

Yes, it's a lovely combo! My post

 

[Irrenarzt]

Yes, it's a lovely combo! My post

Clean setup. I agree that the ESS hump is inaudible, or is at least to my ears. This setup is damn good for ~260 bucks and change (case and cables add an extra 60 bucks or so). What cables are those?

 

[hyperknot]

Clean setup. I agree that the ESS hump is inaudible, or is at least to my ears. This setup is damn good for ~260 bucks and change (case and cables add an extra 60 bucks or so). What cables are those?

Ghent Audio A01, asked to be in simple black. It's very cheap for a high quality 18 cm cable.

 

[DLS79]

The next step up the price ladder is probably what I have, Topping D50s and the Atom connected with JDS stack cables.

Edit: here's a quick shot for reference. Man I need to clean my desk, they where super dusty.

 

[Westsounds]

Tube buffers just add a little bit of distortion harmonics

I like that correction there

The thing is our ears/brains can like these things even if the meters / perfect measurement criteria don't

 

[solderdude]

Yes, people can have preferences for poorer sound reproduction fidelity while measurement equipment does not have a preference.

Funny part is that when there are no differences, say the unsuspecting listener is just told there are differences people can still here differences even when the measurements would show (because there were no real differences) the measurements did not change.

People are funny this way.

 

[Raikon_N]

Hi guys , just got my Atom today(Yes, finally!) and I am currently trying this setup:
Music from PC -> D30 -> MX3 -> Atom(from headphone out of MX3)

That way, headphone is fixed with the Atom. And I can switch to my passive speaker through MX3.
My question is, speaking with listening to my HE4XX headphone, am I loosing a good amount of sound quality compare with just PC -> D30 -> Atom.
My hearing is not that excellent to probably discern the difference but I'd just like to know.
TIA.

PS: I know, the stack is not good-looking.

 

[DLS79]

Funny part is that when there are no differences, say the unsuspecting listener is just told there are differences people can still here differences even when the measurements would show (because there were no real differences) the measurements did not change.

People are funny this way.

That's probably confirmation bias, i mean come on new/different must be better.

 

[L5730]

Just a couple of quick questions.
So assuming I use a passive unbalanced RCA switch to bypass the Atom to go to some powered speakers, would I need to switch the ground/shield connectors in a switcher too, or just switch the signal connections?
The Atom uses an AC/AC PSU, I see a barrel connector so I guess there is not a mains ground connection for a ground loop?

I ask this because a ground loop issues was solved with the connection to powered speakers by using their balanced input connections and modifying the audio cable from the unbalanced DAC.
RCA signal > XLR pin 2 (+ve) // RCA ground > XLR pin 3 (-ve) // RCA ground > cable shield // XLR pin 1 (ground) not connected.

EDIT: Corrected the above diagram and colourised the text.

Seeing how the Atom needs to be turned on in order for audio to go to it's RCA outputs, I'd rather simply bypass it altogether.

Ultimately I'd rather have a DAC/Amp combo device so that I can leave EQ APO setup with it for usual Windows OS based sound (Spotify, Youtube etc.) and a separate DAC connected to powered speakers. Two devices, two different chains. But a passive switcher and a couple of batch script shortcuts will do the job.

 

[trl]

I think that shorting pin1 of XLR to GND would be better, to short-circuit the input inverted amplifier from the speaker. At this moment pin1 is in free air and might get some interferences from the air (very slight chance anyway).

 

[Gradius]

You need to short pin 1 and 3 indeed. I have a Mogami, and those are in short circuit.

Posible configurations:

 

[L5730]

I do not wish to derail this thread, but I must respond.

@trl @Gradius
Thank you for your inputs, but I have a ground loop and therefor the ground of the XLR connection at the speaker (amplifier) input cannot be connected to the ground at the RCA output of the DAC.

I hear no hums or buzzes and have 4m of cable sitting in top of a PC case and a 240v AC power strip/extension. I've turned the active speakers up to max. volume and only hear a slight increase in hiss, which I would no matter how they are connected. In my situation this works just fine.

Read the thread for whole discussion if you are bored and have spare time.
https://www.audiosciencereview.com/...owered-audio-devices-discuss.5899/post-135551

At the very bottom:

If the sending end is the unbalanced two-conductor connector; both the shield and the “-” conductor of the pair are connected to the “shield” conductor of this connector. The “+” conductor of the pair is connected to the signal conductor of this connector. The shield is connected to nothing in the T.R.S. connector (receiving end) and the pair is connected to tip and ring similar to a balanced connection.

http://www.lavryengineering.com/wiki/index.php/Adapter_cable

Going back to my OP, with a setup like mine and using a passive switcher box to switch one unbalanced stereo input between two unbalanced stereo outputs, would I need to switch the ground and signal connections or just the signal connections (as is typical in off the shelf switchers)?
Does the Atom have a mains ground and thus the possibility of causing a ground lop with mains ground attached items? I don't think so, but I may be wrong.

 

[Budgeter]

The below is my calculation for power output of Atom amp, can anyone check it for me? It can output 250mW @300 ohms load (high gain), which means it will provide 300mW @250 ohms. How will these number change if I use low gain instead? I'm gonna get DT 880 Edition-250 ohms, which is rated @0.1 W, so I don't want to override them.


High gain is 4.5 x => With 2VRms input, we will have 9Vrms output => max 324mW @250 ohm (V^2/R). Assuming there is loss in reality, this theoretical result is around the number that Amirm measured so I guess it is correct. However, when switch this to 33 ohms load, it becomes 2.45W at max volume knob.

The latter result is doubled of what Amirm measured, does that means there is some kind of limitation in electronic part that only allow it produce 1.1W output at 33 ohms? Or should I say 1.1W is only the limit of free distortion range, not the power itself?

Now, Low gain is 1.0x => With 2VRms input, we will have 2VRms output => max 16mW @250 ohm. Is this number too small? However, I guess it is already enough to provide really loud volume, right?

 

[trl]

250mW @300 Ohms would be 8.66V RMS. given that Atom is unit-gain in low-gain, I would say that max. output level would be 2V RMS, so the desired 8.66V RMS to achieve 250mW @300Ohms would not be possible. But this is why Atom has the high-gain switch, right?

I'm using DT880 600Ohms with Objective2 on the 3.5X gain and it works fine with -1...0dB FS records, but with low-level recordings a powerful headamp might be needed. I would say that Atom will work perfect with your 250Ohms headphones.

 

[Budgeter]

250mW @300 Ohms would be 8.66V RMS. given that Atom is unit-gain in low-gain, I would say that max. output level would be 2V RMS, so the desired 8.66V RMS to achieve 250mW @300Ohms would not be possible. But this is why Atom has the high-gain switch, right?

I'm using DT880 600Ohms with Objective2 on the 3.5X gain and it works fine with -1...0dB FS records, but with low-level recordings a powerful headamp might be needed. I would say that Atom will work perfect with your 250Ohms headphones.

Well, it's complicated to say but I actually tried 880 600 or 250 ohms before. As in my experience, even at around 10'o clock, the volume is enough for me since I mostly listen on Youtube, which has pretty high average SPL. What concerns me is the possibility of overriding them, not really the volume itself. As my "educated-guess" above, probably even at 10 mW, the volume can reach really high level already, so there is no concern here (Based on my understanding from Solderdude site: https://diyaudioheaven.wordpress.com/tutorials/power-impedance-etc/ )

Besides, I wonder why amplifier output cannot correlate with position of volume knob? Maybe it is duo to the difference between potentiometer? Because of this, it's hard for me to applied these number in reality. (E.g: 100 mW-600 ohms @ 10'o clock knob, etc, things like that)

 

[solderdude]

I wonder why amplifier output cannot correlate with position of volume knob?

max. output power of an amp depends on max output voltage and current capabilities.
Max output voltage is determined by the internal voltage rails.
Max. output current could be limited by current limiting or other reasons.

The volpot position determines the output voltage.
The headphone impedance the output power at a certain output voltage.
At a certain volpot position a 600 Ohm headphone will have draw much much less power than when a 16 Ohm headphone is connected.

The volpot position to obtain a certain SPL thus depends on many factors:
1: gain of the amplifier
2: volpot taper (log, lin or other variant)
3: output voltage of the used source.
4: headphone efficiency (dB/V) not dB/mW and thus also headphone impedance.

 

[Budgeter]

max. output power of an amp depends on max output voltage and current capabilities.
Max output voltage is determined by the internal voltage rails.
Max. output current could be limited by current limiting or other reasons.

The volpot position determines the output voltage.
The headphone impedance the output power at a certain output voltage.
At a certain volpot position a 600 Ohm headphone will have draw much much less power than when a 16 Ohm headphone is connected.

The volpot position to obtain a certain SPL thus depends on many factors:
1: gain of the amplifier
2: volpot taper (log, lin or other variant)
3: output voltage of the used source.
4: headphone efficiency (dB/V) not dB/mW and thus also headphone impedance.

That's a detailed explanation, I get why we can't show power relative to knob position now. But can you tell me why can't we have a voltage output relative to volume knob position? Of course, it will be depended on input source/gain, but I think we should be able to give a spec in term of percentage instead (E.g: 70% of input source x gain @ 12 o'clock, etc).

 

[JohnYang1997]

That's a detailed explanation, I get why we can't show power relative to knob position now. But can you tell me why can't we have a voltage output relative to volume knob position? Of course, it will be depended on input source/gain, but I think we should be able to give a spec in term of percentage instead (E.g: 70% of input source x gain @ 12 o'clock, etc).

That's just voltage gain. Getting high gain is easy just change one resistor value. You do not get more power out of it. It will just clip earlier or plainly not work depending on whether the gain is before the volume pot or after. Power is limited by the power supply voltage for voltage output and output device output current for current output.