Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Hissing from Topping NX2 (Read 5968 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Hissing from Topping NX2

Bought one of these recently to solve the issue of an incredibly noisy output from my chromebook.

The device seems to have an audible, albeit quiet, hiss when turned on; which does not vary with potentiometer positioning.  I'm currently using AKG K550, which are only 32Ω, but within the manufacturer's impedance specifications.

Is this just a result of the low impedance of my headphones and something I'll just have to deal with for any NX2, or is my device a dud?

Any clues as to how I may fix this?

Thanks in advance.

Re: Hissing from Topping NX2

Reply #1 –
Since the thing is rechargeable, do you hear hiss while running off batteries with USB disconnected?    It's possible that the noise is getting in through the USB power.

Re: Hissing from Topping NX2

Reply #2 –
I'm afraid I've never used USB power and the noise is present with and without the USB data connection and analogue audio input.

Re: Hissing from Topping NX2

Reply #3 –
Well, seems you're hearing the actual noise from the output buffer. As you don't notice changes when moving the potentiometer. And it isn't surprising honestly. xd  For example the Xduoo X2 supposedly has an SNR of 107 dB(no reference) with a max output voltage of 1 Vrms. So if we use that 1 Vrms as reference, the noise is 4.4 uVrms. Which is very good. For example that would produce an inaudible SPL in your K550 (Which are sensitive headphones) of 8 dB.

But it's been measured here: https://headphoniaks.com/blog/en/comparativa-xduoo-x2-fiio-x1-sansa-clip/2/

And turns out its noise is -87 dBu (u=0.775 Vrms), That is 33 uVrms. In that case that would be 25 dBSPL for the AKG, Which is a really really faint hiss. (For me, anything under 21 dBSPL is pure silence). But this is just an example, I haven't found anything about the Topping NX2.

Re: Hissing from Topping NX2

Reply #4 –
Topping claims sub 10µV noise on its website; which, from what you've written, I would infer to be silent (though I wouldn't know how to make the calculations myself). Does this mean my device is a substantial deviation from the usual spec, and hence a dud? Or is it more likely that the manufacturer is exaggerating; as in the case you mentioned? 

Re: Hissing from Topping NX2

Reply #5 –
Quote from: pixelwitch on
Topping claims sub 10µV noise on its website; which, from what you've written, I would infer to be silent (though I wouldn't know how to make the calculations myself). Does this mean my device is a substantial deviation from the usual spec, and hence a dud? Or is it more likely that the manufacturer is exaggerating; as in the case you mentioned? 

More likely that the manufacturer is exaggerating, like the case of the Xduoo. And here are the calculations explained http://nwavguy.blogspot.com/2011/09/noise-dynamic-range.html

(According to innerfidelity) K550s require 0.055 Vrms to reach 90 dBSPL. So to know the SPL of "X" noise, use this: 90 dB - 20log(Vheadphone for 90 dB/Vnoise).

Also, there's no info about the chipamp "TP9260", but anyway, try to ground its case (Join the ground pin of the output jack to the case). That might reduce hiss a bit.

Re: Hissing from Topping NX2

Reply #6 –
Ah, thanks a lot. I'll certainly give grounding a go (or at least ask someone more competent than myself to do so). Very interesting link too.

Re: Hissing from Topping NX2

Reply #7 –
Quote from: pixelwitch on
Bought one of these recently to solve the issue of an incredibly noisy output from my chromebook.

It is reasonable to expect some improvement, but not total inaudibility of noise.

Quote
The device seems to have an audible, albeit quiet, hiss when turned on; which does not vary with potentiometer positioning.  I'm currently using AKG K550, which are only 32Ω, but within the manufacturer's impedance specifications.

This may not be a problem.

There is no necessary connection between earphone sensitivity and impedance. 

Quote
Is this just a result of the low impedance of my headphones and something I'll just have to deal with for any NX2, or is my device a dud?

Since there is no necessary connection between earphone sensitivity and impedance. the relatively low impedance of your headphones is not the problem.  Two other potential is that the headphones may be too sensitive for your applicaiton, and that your apparent exectation of no audible noise at all my be unreasonable.

Quote
Any clues as to how I may fix this?

Find headphones that are not, as the K550s  are, exceptionally sensitive.

Find a headphone amplifier/DAC that has less residual noise.  Noise level or SNR are more relevant specs.

Adjust your expectations. Is the hiss interfering with your actual enjoyment of music?

Re: Hissing from Topping NX2

Reply #8 –
I too am experiencing a hiss in the NX2 and it is very audible with IEMs like the Rock Zircons. It is not audible with PX100, HD598 and other "less" sensitive HP.

I'd be willing to open the NX2 up and reduce gain so I can use it as a DAC with sensitive iems.

Re: Hissing from Topping NX2

Reply #9 –
Quote from: korewell on
I too am experiencing a hiss in the NX2 and it is very audible with IEMs like the Rock Zircons. It is not audible with PX100, HD598 and other "less" sensitive HP.

I'd be willing to open the NX2 up and reduce gain so I can use it as a DAC with sensitive iems.

Might not help much at all - as the noise floor may not vary with the basic gain of the amplifier.

Options that may work would include  to get less sensitive headphones, a different amp, or re-engineer the amp.

Re: Hissing from Topping NX2

Reply #10 –
You could pad down the output of the amplifier so you don't hear the hiss, and then turn up the gain to make the music loud enough.  But you might just make the hiss louder by turning up the gain.  Sounds like a lousy design to me.

Re: Hissing from Topping NX2

Reply #11 –
Well, the hiss is there with no input. It does not change with volume. You can turn the volume knob to 10 and the hiss level is the same.

I tried to find some info about the "TP9260" op-amp but there's nothing. My guess is that it's a custom chip made for topping.

Anyway, I opened it up and located what seem to be the resistors for the regulation of the amp. It's looks like a simple fix and it could even be "easy" to put a hi/lo gain switch in the NX2. I'd be happier to have a datasheet and more background tough.

Andrew, why do you think that the noise floor is not related to the gain ?

Re: Hissing from Topping NX2

Reply #12 –
Quote from: korewell on
Well, the hiss is there with no input. It does not change with volume. You can turn the volume knob to 10 and the hiss level is the same.

I tried to find some info about the "TP9260" op-amp but there's nothing. My guess is that it's a custom chip made for topping.

Anyway, I opened it up and located what seem to be the resistors for the regulation of the amp. It's looks like a simple fix and it could even be "easy" to put a hi/lo gain switch in the NX2. I'd be happier to have a datasheet and more background tough.

Andrew, why do you think that the noise floor is not related to the gain ?

Because the source of the noise is located after the potentiometer. It means that noise is either from the opamp itself, badly designed DC-DC converter, or amplified Johnson noise due to the presence of >100K resistors with the opamp. You could try to fix it.

But honestly, if you're willing to go this far, why don't you build a headphone amp that's shown to have a good noise performance?
http://nwavguy.blogspot.com/2011/07/cmoy-with-gain.html

That design can be further improved by the way.

Re: Hissing from Topping NX2

Reply #13 –
It's not just the analog amp that attracts me. I like the idea of using the NX2 for its dac with my aging airbook and also for on-the-go spotify with an android phone. I could probably add a dac inside the cmoy but it will be more complicated.

There are a couple of 39K resistors around the audio path and the TP9260 but nothing major. R9-R10 and R11-R12 are the two audio channels. The section around R6-R7 and R4-R5 are the two audio channels going out of the DAC.

On the second photo, I think pins 12 and 14 are the outputs that go straight to the audio out jack. R9-R11 and R10-R12 seem the resistors that control the gain.

Maybe I should just try out the empirical way :)

Re: Hissing from Topping NX2

Reply #14 –
Quote from: korewell on
It's not just the analog amp that attracts me. I like the idea of using the NX2 for its dac with my aging airbook and also for on-the-go spotify with an android phone. I could probably add a dac inside the cmoy but it will be more complicated.

There are a couple of 39K resistors around the audio path and the TP9260 but nothing major. R9-R10 and R11-R12 are the two audio channels. The section around R6-R7 and R4-R5 are the two audio channels going out of the DAC.

On the second photo, I think pins 12 and 14 are the outputs that go straight to the audio out jack. R9-R11 and R10-R12 seem the resistors that control the gain.

Maybe I should just try out the empirical way :)

And why don't you use your phone and airbook as DAC?

So, you're saying that R9 to R12 are all 39K ohm?

If those are the gain resistors, that means the chipamp has a gain of 2. This is the lowest gain that someone would find in a amp. Try feeding 0.2 Vrms at line in and then measure the output voltage. you should have 0.4 Vrms at the output, that way you can confirm that those are the gain resistors.

Once confirmed, just remove the two 39K resistor that are connected to ground, and you'll have a gain of 1, hopefully the chipamp is unity gain stable  ;D.

But seriously, by the number of pins, seems there's a DC-DC converter integrated in that chipamp. Its more likely that's the source of hiss, but without datasheet of the chipamp, there isn't much left to say.

Re: Hissing from Topping NX2

Reply #15 –
Phoenix, I was under the impression that these were 2x18 and 2x39K resistors. The marking are 393 and 180 (looks like "18C")
This would make it a unity gain amp (!?!).

I am missing something for sure. I need to draw the schematic of this thing.

Thank you for the hint on the DC/DC possibly being integrated. I'll try to do a proper scan of the PCB with the battery removed.

ADD: I don't use my phone/comp as a dac as they hiss even more than the nx2.

Re: Hissing from Topping NX2

Reply #16 –
Quote from: korewell on
Phoenix, I was under the impression that these were 2x18 and 2x39K resistors. The marking are 393 and 180 (looks like "18C")
This would make it a unity gain amp (!?!).

I am missing something for sure. I need to draw the schematic of this thing.

Thank you for the hint on the DC/DC possibly being integrated. I'll try to do a proper scan of the PCB with the battery removed.

ADD: I don't use my phone/comp as a dac as they hiss even more than the nx2.

No way 18 Ohm if they're the gain resistors. The thing would instantly clip any signal. I would first put a 0.2 Vrms signal at the line in and measure the output to know the gain of the amplifier. Then since they're no longer the same resistors in the feedback divider, you would have to find out if the opamp is set as an inverting or non inverting amplifier, to do that you'll need to do a loopback test to check if the absolute phase is inverted or not. Only after that I would start to mess with the circuity in the amp.

And by the way, not all those chipamps that have a integrated DC-DC converter let you change the gain, some have the feedback resistors integrated in the IC.

Another solution is to add 10 ohm resistors in series with the output, that would attenuate hiss 4.2 dB for a pair of 16 ohm headphones. (that method has its side effects). Unfortunately there's basically no useful info about your IEM's to know the scale of the side effects. Even the official specs are so bad that no way those headphones have a 93 dB mW sensitivity, that would be 0.09 Vrms for 90 dB SPL, while the PX100 need 0.05 Vrms to reach that SPL!

Wrt "ADD", that shouldn't be a problem. If you have a amp where you can't hear hiss, remember that the hiss at the output of your source is attenuated by the potentiometer of the amplifier.

If I only knew of some useful specs about those IEM's I could approximate the amplitude of the noise. Do you have any other headphone in which you hear hiss as well?

By the way, what is the model of your phone?

Re: Hissing from Topping NX2

Reply #17 –
Quote from: korewell on
Phoenix, I was under the impression that these were 2x18 and 2x39K resistors. The marking are 393 and 180 (looks like "18C")
This would make it a unity gain amp (!?!).

I am missing something for sure. I need to draw the schematic of this thing.

Thank you for the hint on the DC/DC possibly being integrated. I'll try to do a proper scan of the PCB with the battery removed.

ADD: I don't use my phone/comp as a dac as they hiss even more than the nx2.

The official specs mention a gain of 6 dB (2X).
http://www.tpdz.net/en/products/nx2/



Re: Hissing from Topping NX2

Reply #18 –
I did some work. R11 and R12 (marked 18C) are actually 15K resistors. The others are 39K. The configuration on the pcb looks like a simple inverting amplifier with 2x gain. The TP9620 is powered with 4V. It seems to generate +/- 4V and +/- 2V outputs via external capacitors.

I'm starting to question if lowering the gain will do any good. Have not hooked it to a scope yet.

The only other in-ears that I can hear the hiss are some UiiSii HM7 which also have bad specs. They are 16Ohm with 90dB sensitivity which cannot be correct. Actually they hiss more than the RZ.

I might not be the normal type of music consumer. I tend to listen at very low volume levels which makes the hiss a real issue. I also cannot stand pops and clicks from digital devices either. The phones I tried are used only as transports (tried Redmi 1S and a Oneplus One) but also use my notebook from time to time. Having a good dac/amp becomes mandatory. I used to have a X3-ii for those tasks and it was quite acceptable. Might have to just throw the NX2 in the bin and buy a new X3-ii.

Re: Hissing from Topping NX2

Reply #19 –
Quote from: korewell on
I did some work. R11 and R12 (marked 18C) are actually 15K resistors. The others are 39K. The configuration on the pcb looks like a simple inverting amplifier with 2x gain. The TP9620 is powered with 4V. It seems to generate +/- 4V and +/- 2V outputs via external capacitors.

I'm starting to question if lowering the gain will do any good. Have not hooked it to a scope yet.

The only other in-ears that I can hear the hiss are some UiiSii HM7 which also have bad specs. They are 16Ohm with 90dB sensitivity which cannot be correct. Actually they hiss more than the RZ.

I might not be the normal type of music consumer. I tend to listen at very low volume levels which makes the hiss a real issue. I also cannot stand pops and clicks from digital devices either. The phones I tried are used only as transports (tried Redmi 1S and a Oneplus One) but also use my notebook from time to time. Having a good dac/amp becomes mandatory. I used to have a X3-ii for those tasks and it was quite acceptable. Might have to just throw the NX2 in the bin and buy a new X3-ii.

Since you have an scope, I would guess you also have a breadboard. Again, you really should try building a buffer based on the 4556 before buying anything else.

I just did a quick calculation, and a fixed hiss of 50 uV (severe value) from a DAC with a max output voltage of 1 Vrms would be 5 uV if you attenuate the signal with the potentiometer of the buffer to get 0.1 Vrms. 0.1 Vrms is 104 dBSPL peak for a pair of hypotetic IEMs that have a 0.02V/90dB sensitivity (8 dB easier to drive than the PX100). 5 uV would produce an SPL of 18 dB. (For me, anything under 21 dB SPL is pure silence). (And this attenuation also applies to the clicks and pops)

Re: Hissing from Topping NX2

Reply #20 –
Phoenix, I have nothing fancy. Just a ds1000e.

The 4556 will require an external dc/dc or converter and the space inside the NX2 is at a premium, specially the vertical space. Sure I can put it in a cmoy tin and make it a NX2/Cmoy frankenstein :) I could actually just build a JDS cmoy in place of the TP9620 and call it a day.  Might be an idea because I'll gain a dac and the battery charging circuit. I'm not sure if it's a good route as this all started as a simple resistor mod project and is getting out of control already.

The problem imho is what you pointed out: The noise comes from the dc/dc converter inside even with shorted input. This is a dirt cheap dac/amp that works well with low sensitivity headphones. Pretending to make it a hi-fi all-round beast will require to rebuild half of it and at that point I'm better off starting with a proper dac and using proper design with a separate amp and buffer/filters.

One OT question: I don't think those sensitivity numbers on the RZ of UiiSii mean anything, 93 dB and 90dB (it they were spl/mw) would be too little with respect to the PX100 which is 112 dB SPL. So... what you think those dB levels in the headphone specs are related to ? 1 Volt ?