• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Topping NX1s Portable Headphone Amp Teardown

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,550
Location
Seattle Area
In the recent review of Topping NX1s Headphone Amp, question was raised regarding the reason there is a roll off in low frequencies. I speculated that it is likely caused by capacitor coupling. Capacitors are used to block DC output as they have infinite resistance to steady voltages (and become a short as you go to infinite frequency). A teardown was requested and the owner kindly agreed.

Opening the unit was very easy courtesy of smart assembly with just two very long screws that held the front and back together. Here is the front part of the PC Board with most of the components:

Topping NX1s Portable Headphone Amplifier Phone Teardown.jpg


Starting bottom right, there is a switching power supply to create stable voltages are needed for the operation of the unit (likely positive and negative voltages). It also handles power management for charging of the Lithium cell on the other side (edit: there is an IC on the other side which may be performing this function). Nice to see an isolation ring all around that segment to keep its noise away from the amplifier section to the left and top.

Amplification and gain are provided through a pair of OPA1652 OpAmps (operational amplifiers). The OPA1652 is a common audio opamp with very low distortion ratings:

1577472744744.png


0.00005% results in distortion of -126 dB which is more or less what we saw in the measurements.

There are a pair of opamps in each package and each provides the gain/buffering for each stereo channel. I could not read the markings on the second (top) Opamps but I am assuming it is a higher current, BUF634 OpAmp as seen in the application note for OPA1652:
1577472930503.png


Edit: turns out the buffer is LMH66, not BUF634. See: https://www.audiosciencereview.com/...able-headphone-amp-teardown.10577/post-293270


Anyway, the interest was in the cause of bass roll off. There is none in the above schematic but let's look at the other side where the battery mostly lives:
Topping NX1s Portable Headphone Amplifier Phone Top PCB Teardown.jpg


We see two 1 microfarad capacitors directly after the volume control. They are film capacitors which are quite large for their value. They are rated at 63 volts and likely there to clock DC from input travelling all the way through the amplifier to the headphones. A larger value capacitor of a different type would not have caused the roll off that we see.

Overall, the value remains incredible for all the components and circuits in NX1s.

P.S. Yes, the wire came out of its socket when I tried to disconnect the battery. :( It uses a very flimsy connector that was a pain to disconnect.

 
Last edited:

Cahudson42

Major Contributor
Joined
Sep 21, 2019
Messages
1,083
Likes
1,556
Question is: Does the NX3s have similar input blocking caps?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,550
Location
Seattle Area
Question is: Does the NX3s have similar input blocking caps?
Looks like it does:
1577475678586.png


But they are more than double the value (2.2 microfarad versus 1) so the roll off would be much less.
 

Cahudson42

Major Contributor
Joined
Sep 21, 2019
Messages
1,083
Likes
1,556
Hmmm. A 'bonus teardown' of the NX3s. :) Could we see the rest? Is circuitry similar? i.e. same opa1652?
 

NTomokawa

Addicted to Fun and Learning
Forum Donor
Joined
Jan 14, 2019
Messages
779
Likes
1,334
Location
Canada
two 1 microfarad capacitors
Ah! Ha! Caps on the input side! Haven't thought of that one.

Also why film caps? I am not an EE so I genuinely don't understand. Could one replace then with, say, 330uF electrolytics? Those are simple through-hole solderings that could be easily modified/replaced by the end user.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,588
Likes
38,288
Location
Gold Coast, Queensland, Australia
We see two 1 microfarad capacitors directly after the volume control.

The capacitors appear to be prior to the volume control, as is standard practice. If the capacitor was after the volume pot (wiper), DC would not be blocked on the pot (and they get noisy, fast), and response would vary even more with pot position.

The pot is marked A103. That means it is an audio taper (log) and 10K ohms (too low IMO- 50K would be better with 1uF). A 1uF capacitor along with your 600 ohm AP source resistance and the 10K pot has a -3dB point around 15Hz (just like your FR plot), (not including the parallel input impedance of the actual active stage following).

Raising the capacitor to 4.7uF would take the F3 down to around 3Hz. A small electrolytic could easily be substituted and the bottom end FR would be much better.
 
Last edited:

Keicar

New Member
Joined
Jan 24, 2019
Messages
2
Likes
3
Location
Adelaide, South Australia
Well, I don't think the output stage is a BUF634, as they only have one amplifier per package, so there'd need to be two of them for a stereo application. (Strictly speaking, I wouldn't call the BUF634 an 'op-amp', as the gain is fixed at unity - you'll note it's typically used inside the NFB loop of an op-amp, though.)
 

Keicar

New Member
Joined
Jan 24, 2019
Messages
2
Likes
3
Location
Adelaide, South Australia
Ah! Ha! Caps on the input side! Haven't thought of that one.

Also why film caps? I am not an EE so I genuinely don't understand. Could one replace then with, say, 330uF electrolytics? Those are simple through-hole solderings that could be easily modified/replaced by the end user.

Film caps can have measurably lower distortion than electrolytics in some applications, there's a series of articles by Cyril Bateman that provide interesting reading. Generally I'd err on the side of oversizing input coupling caps, as the source impedance is something that's beyond my control - but perhaps don't go overboard. You don't want the response to a DC offset at the input to be too slow, otherwise your coupling caps might as well not be there.
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
Buf634 is a buffer not an opamp. It doesn't have gain. And this circuit doesn't use buf634. Let alone the fact buf634 only has one channel, you need two buf63s. I suggest not to make assumptions on these teardowns.
 
Last edited:

pavuol

Major Contributor
Forum Donor
Joined
Oct 2, 2019
Messages
1,562
Likes
3,951
Location
EU next to warzone :.(
I wish that all electronics had "user replaceable battery". Using screws/connectors is still acceptable, just no glue/soldering/etc... (of course the connector in this case could be of better quality)
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,550
Location
Seattle Area
Buf634 is a buffer not an opamp. It doesn't have gain. And this circuit doesn't use buf634. Let alone the fact buf634 only has one channel, you need two buf63s. I suggest not to make assumptions on these teardowns.
I put in a question mark on the picture for a reason. I am not reverse engineering these devices. Only giving some high level ideas. As I said, I can't read the markings on the second IC. The interest here was whether there were capacitor couplings or not. You want to do better, step up and do your own teardowns. I only have so much time to spend on these things. :(
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,550
Location
Seattle Area
Well, I don't think the output stage is a BUF634, as they only have one amplifier per package, so there'd need to be two of them for a stereo application. (Strictly speaking, I wouldn't call the BUF634 an 'op-amp', as the gain is fixed at unity - you'll note it's typically used inside the NFB loop of an op-amp, though.)
It is an opamp wired internally for unity gain.

Anyway, I searched ab it and found an image of the PCB which shows the output to be LMH66 series:

1577563773627.png


Which comes in dual package. :)
 

Bob-23

Senior Member
Joined
Jul 3, 2019
Messages
425
Likes
379
Location
Berlin, Germany
If the input cap is followed by a resistor to ground - and there has to be one, for the functioning of the opamp (bias current) - of, say, 100k, we have a highpass-filter, with a -3db-point of 1,59 hz. So even with only 10k to ground we would have a bass-roll-off beyond audibility. Thus there must be other reasons for the bass-roll-off of the amp.

As a rule of thumb: you prefer film caps in the audio path, but if you use electrolytics take 10x the value you would calculate for film caps so that there is only a very small voltage drop across them - hence distortion remains inaudible. You would take a low-ESR cap with high temperature rating, as they don't age that quickly, but all electrolytics age and change their electric behaviour during that process...

If you put an electrolytic cap on the output of the amp as a protection against a dc-accident (failing opamp e.g.) you indeed have to take into account the time-constant: the bigger the cap, the longer it takes until the cap is 'loaded' with dc, and only then it fully blocks dc. Compromise is necessary: Avoiding unwanted highpass-filtering (cap+headphone form a highpass) affords a high electrolytic cap, the protective function affords a low value... But anyway, that's probably to slow, dc-accident-protection is better done otherwise: by a muting circuit withe relays e.g.

Positioning input caps after the vol pot shields the pot against the opamps bias current.
 
Last edited:
Top Bottom