Topping L30 II Review (Headphone Amp)

[daftcombo]

I hard a tiny static in one channel in one song but could not replicate it. That usually indicates clipping and that track had super deep bass. And I was at max volume.

Which track is it, please?

 

[JohnYang1997]

Which track is it, please?

If it was at high gain max volume then it really should be clipping into lower impedance load. Unlike old L30, you actually can get the amplifier's full output swing at maximum volume with 2V input.

 

[tifune]

Which track is it, please?

Max Richter - Cradle to the Grave

I actually have no idea, but that track caused my A90 to pull around 28W from the wall on DCA Stealth per my cheap-o wattmeter. nothing else has ever come close

 

[Atanasi]

The high gain sets the gain for the amplifier circuit. The mid gain essentially has no extra attenuation and no extra gain in the amplifier circuit. Low gain has an extra attenuation resistor to attenuate the signal before the amplifier circuit.

Is the implementation of the original A90 different in this regard? Amir's review of A90 shows that the low gain followed more or less the same curve as the medium gain, while the high gain had a higher noise floor.

It is noteworthy that the graphs indicate THD+N with respect to a particular output power, so the low gain would have hotter inputs to produce the same power.

 

[H-713]

@JohnYang1997 Any chance you'd be willing to discuss the topology used here a little? I'm genuinely curious how you've gotten the distortion this low.

 

[JohnYang1997]

Is the implementation of the original A90 different in this regard? Amir's review of A90 shows that the low gain followed more or less the same curve as the medium gain, while the high gain had a higher noise floor.

It is noteworthy that the graphs indicate THD+N with respect to a particular output power, so the low gain would have hotter inputs to produce the same power.

The attenuation ratio is different so the noise with -9db is lower than -14db when volume is at max position.

Also the balanced input of A90 which has a differential to single ended conversion shows a bit more noise than the single ended input of A90 and L30 II.

A combination of these two factors lead to that noise is much higher in comparison in -14dB position with L30 II.

When in real world use, pot position at 9-12 o'clock, the noise from either the attenuation resistor or the differential to single ended conversion is attenuated, resulting both low and consistent noise level.

 

[KEFCarver]

My power sweeps vs impedance push the amp well into clipping and keeps it there for good 2 to 4 seconds. The protection circuit must do its job or it gets damaged.

I do like to develop some kind of stress test fixture in the future to do some quick testing for reliability. But have not had time to do so.

Thanks for replying- that makes sense. Understand about trying to making time for "things"

 

[audiofun]

@JohnYang1997 Any chance you'd be willing to discuss the topology used here a little? I'm genuinely curious how you've gotten the distortion this low.

L30 is just a TPA6120A2 buffered version of JDS Atom with good ground layout. see this post and this post.
L30 II is just a bootstrapped version of L30.

 

[Rja4000]

@JohnYang1997
What is the input impedance ?

You always give the output impedance in specs, but never the input impedance...

 

[Jimster480]

@JohnYang1997
What is the input impedance ?

You always give the output impedance in specs, but never the input impedance...

True. I feel like every brand should start including this

 

[iamsms]

I wish these came with laptop style power supply - the plug that goes into the power strip is small and doesn't take multiple slots like this one. Something like this would be nice too.

 

[jasonhanjk]

Low gain is measured worse here because the volume position is set at maximum. This is OK as a test but in real world use you wouldn't do that as you would switch to higher gain settings already.

The reason for the higher noise is because when the pot is set a maximum the noise is set by the high attenuation resistor and the high value of pot. When you use at say less than 12 o'clock the noise will be largely diminished to 1/4th or lower. And that's also why the noise floor is the same when the pot is set to nil for both low and mid gain settings.

The best result is to set pot at max volume and gain at mid, change the volume by the DAC.
Using pot is a guarantee way to add noise.

 

[Labjr]

I wish these came with laptop style power supply - the plug that goes into the power strip is small and doesn't take multiple slots like this one. Something like this would be nice too.

 

[Jimster480]

I wish these came with laptop style power supply - the plug that goes into the power strip is small and doesn't take multiple slots like this one. Something like this would be nice too.

View attachment 220807

The A30 comes with a brick like that, the L30 doesn't?

 

[JohnYang1997]

L30 is just a TPA6120A2 buffered version of JDS Atom with good ground layout. see this post and this post.
L30 II is just a bootstrapped version of L30.

You were basically correct. But learning from is very different from copying. I also learned these circuits from Bob Cordell's book and Douglas Self's book on TMC and TPC. I didn't make a big fuss about how new the circuit was but the actual performance it was achieving. That type of circuit was already experimented long long time ago by many people. JDS also must have learnt it from somewhere. And that's absolutely ok. And in the course of the time I was on ASR, I respected JDS very much and recommended Atom to members all the time.

However the said company in China not only copied the exact circuit of L30(including the not working dc protection circuit), but also filed patent exactly describing circuit of A90 in China. This is the completely different behavior than what I was doing. And now they are claiming the circuit is independently developed(which they apparently couldn't do in the past decade?) while the L30/A90 circuit was the first design I put out. I only joined Topping in December 2019.

 

[Jimster480]

You were basically correct. But learning from is very different from copying. I also learned these circuits from Bob Cordell's book and Douglas Self's book on TMC and TPC. I didn't make a big fuss about how new the circuit was but the actual performance it was achieving. That type of circuit was already experimented long long time ago by many people. JDS also must have learnt it from somewhere. And that's absolutely ok. And in the course of the time I was on ASR, I respected JDS very much and recommended Atom to members all the time.

However the said company in China not only copied the exact circuit of L30(including the not working dc protection circuit), but also filed patent exactly describing circuit of A90 in China. This is the completely different behavior than what I was doing. And now they are claiming the circuit is independently developed(which they apparently couldn't do in the past decade?) while the L30/A90 circuit was the first design I put out. I only joined Topping in December 2019.

So you have other chinese companies 100% ripping off your designs?

I thought there were no patents in china for these sorts of things? That is a real disaster.

 

[JohnYang1997]

So you have other chinese companies 100% ripping off your designs?

I thought there were no patents in china for these sorts of things? That is a real disaster.

 

[Jimster480]

So what are you to do about this problem? Can you go to court over it ? Will the other company start selling knockoff Topping stuff or put it into another brand and claim it to be theirs (sounds like it is the 2nd one from your above post).

 

[FrantzM]

Hi

What else to say, it has become customarily: Topping tops it.... again.
Congratulations.

There is, already an, of course, excellent 8-channel DAC... Now Can we dream of an AVR or Pre/Pro? With balanced outputs, DRC and ..the usual AV codecs and, superlative performances? I am certain not to be alone.

Peace.

 

[FrantzM]

Yep, amazing performance for $149... at least it's not $1499 or worse $14999!

Thanks for your time @amirm to perform these tests and measurements.


JSmith

Don't give the High End Audio, ideas. Some solid, brushed aluminum case, a power supply for 200 wpc in another but separate solid aluminum chassis and ... $19,999,00 later you have the new HEA "best preamp ever", a true bargain



Peace.

P.S. Just to make sure that I own a Topping, I just purchased a D10 Balanced, to replace a perfectly good , working but single-ended Khadas Tone Board... There!!