- Joined
- Oct 25, 2019
- Messages
- 11,111
- Likes
- 14,774
No problem!I didn't realize there was clipping. My bad.
No problem!I didn't realize there was clipping. My bad.
Hello,
I wonder if anyone saw these measurements?
https://goldensound.audio/2021/07/22/ifi-zen-can-measurements/
They have some curious differences to Amir's ones...
First, Amir's FFT's show significant mains harmonics, which are absend in the tests by Goldensound.
One wonders why this is is? HasAmir's test setup a hidden ground loop that adds around 20dB of 60/180 Hz noise.
This noise cannot be the result of poor power supply design (it would be 120/240 Hz in this case).
And if those two peaks at 160 & 180 Hz did not exist, could the SNR be significantly than Amir measured?
I often see these 60/180Hz peaks in Amir's measurements, if they are bogus, and their absence would improve measured SNR, perhaps many of Amir's measurements need-redoing after he resolves the problem that causes these, to be fair to manufacturers?
Another fun part is the shape of the bass boost:
View attachment 149027
Down to 50Hz the curves are the same, but they strongly deviate afterwards, this seems odd. Maybe Amir was using a very high signal level and clipping the Amplifier?
Looking at crosstalk with that infamous "3D" effect seems to again have curious differences:
Again, one wonders, is it just the scaling? And why select this kind of scaling? There also seems to be a 6dB in 320kHz measured crosstalk, why?
M.I.
The mains harmonics are coming from the leakage current of the power supply. It's low enough to not affect SINAD. The bandwidth is not captured on Goldensound measurements. Amir has been using 22khz instead of 20khz. The noise is no big deal in the context of the overall performance of the device.Hello,
I wonder if anyone saw these measurements?
https://goldensound.audio/2021/07/22/ifi-zen-can-measurements/
They have some curious differences to Amir's ones...
First, Amir's FFT's show significant mains harmonics, which are absend in the tests by Goldensound.
One wonders why this is is? HasAmir's test setup a hidden ground loop that adds around 20dB of 60/180 Hz noise.
This noise cannot be the result of poor power supply design (it would be 120/240 Hz in this case).
And if those two peaks at 160 & 180 Hz did not exist, could the SNR be significantly than Amir measured?
I often see these 60/180Hz peaks in Amir's measurements, if they are bogus, and their absence would improve measured SNR, perhaps many of Amir's measurements need-redoing after he resolves the problem that causes these, to be fair to manufacturers?
Another fun part is the shape of the bass boost:
View attachment 149027
Down to 50Hz the curves are the same, but they strongly deviate afterwards, this seems odd. Maybe Amir was using a very high signal level and clipping the Amplifier?
Looking at crosstalk with that infamous "3D" effect seems to again have curious differences:
Again, one wonders, is it just the scaling? And why select this kind of scaling? There also seems to be a 6dB in 320kHz measured crosstalk, why?
M.I.
The mains harmonics are coming from the leakage current of the power supply. It's low enough to not affect SINAD.
The noise is no big deal in the context of the overall performance of the device.
It's likely that goldensound connected the earth connection directly to the device under test to eliminate the noise.
Amir's measurements are fair. You do not want to create ideal test environment for devices that won't likely be used in.
The issue is that goldensound doesn't mention the test condition like what input what output being used.
There are also many devices don't have the mains noise being measured here. So it's all in the design of the devices.
We also use APx555/B in our lab. It's the absolute absence of mains noise that's suspicious not the measurement that shows it.
One possibility is that his 4.4mm cable has the ground wired to the XLR shield. This is not necessary good or bad. But this introduce a earth current return path directly at the output eliminating leakage current effects.
This is not representative to real world usage since headphones aren't earthed. So the return path is not present when users listen to music.
Dont know about the mains frequencies. Sure others can offer insight.The bass boost profile looks similar to me but @amirm shows down to 10hz so you see it roll off.
I dont know is how/ why Goldensound has the "non" 3d measurement so much lower- suspect different approach to measuring/ output level.
Both measurers tell you the key takeaways though- this is a powerful but far from clean amp
If input is unbalanced, connect AP's unbalanced output ground to earth does not have benefits or very little. But connect earth connection directly to the output will eliminate all the effect of leakage current.Hi,
SINAD - why use an obscure measure that is actually just THD&N but needs complex conversions to make sense and to be compared to other measurements and specifications? Just BTB.
And I did not ask if this noise might affect SINAD (which like THD&N) is not a very useful measure, but if it might affect SNR. So please answer questions asked, instead of trying to deflect from the issues with verbose verbiage that fails to address issues raised..
Is that really so? I see mains related noise 20dB above the FFT noisefloor. That's a lot to just brush off with "not a big deal".
You mean goldensound follow best practice, in order to avoid know issues with the Audio Precision input section's limited common mode rejection ratio?
But this mains noise would not exist if (for arguments sake) we used a the device under test plugged into a headphone which in turn is placed on a KEMAR etc. This noise is not generated by the device being tested, but by the test set-up and a failure to employ best practice.
So this measurement shown noise that will not exist in any use case for this device (which is to drive headphones)
Actually, it is mentioned, if you were to bother to read all of what is presented.
More precisely, it is in the design of the power supplies used with these devices and limitations in the Audio Precision Test Set's design. And the fact that many vendors from the far east ship products that would fail various agency compliance requirements if they were imported formally into the USA or EU, compared being grey or personal imports, which is what you get when you buy on Amazon from a vendor in Asia.
Really, you mean if people who know how to use test equipment get results that differ from your you suspect that other people's results are faked, instead of suspecting your own work first? How curious.
What you mean to say, the presence of a ground connection would in effect apply an Earth to an otherwise unearthed system and thus remove the common mode leakage that throws off the AP's test results, if the AP inputs are left floating?
Ou conraire, mes ami.
It precisely represents real world usage, as real headphones have in effect an infinite common mode rejection ratio and thus fail to show any response to common mode leakage, whereas the AP shows significant response. So it actually would correctly represent the user listening to music, as the user listens to the output from the headphone and not the FFT graph from an AP testest.
Please look again. The goldesound tests clearly show the same bass boost profile as with other iFi devices. It shows 9dB boost at 20Hz and smooth curve.
Amir shows a distinct "step" in the curve at around 43Hz (and remarks on it) and 12 dB boost at 20Hz. Clearly the two measurements show something very different.
Crosstalk should not be affected by signal levels, if the test equipment is used correctly, as it is a relative measure, not absolute.
I am not sure what "clean" means. I would suggest that 0.0009% THD&N for an amplifier seems sufficient, if Amir declares 0.009% THD&N "State of the Art" (for a headphone: Dan Clark Stealth Review (State of the Art Headphone).
What is more, the distortion from the Amplifier tested would be considerably lower than 0.0009% if tested at the output level at which the Headphone being discussed here produces it's 0.009% THD.
This level is 0.564V into 23 Ohm, or around 13mW, at this level noise might dominate the Amplifier's THD&N measurements, especially if ten times (20dB) more mains noise is measured that would actually be presented to the headphone.
So it would seem to me that the distortion of this amplifier is entirely inconsequential and will not degrade the distortion of a state of the art Headphone, much less one that is merely "good".
The crosstalk performance may very well be intentional, but again seems not a real problem when listening to music as the "crosstalk" in recorded music is much higher.
Signal to noise ratio seems adequate as well, but may be improved if the mains noise created by the test setup were removed from the measurement.
What more could you require for music listening?
True, other amplifiers may have lower distortion and greater channel separation (maybe greater SNR, hard to say with compromised measurements), but driving real headphones, in the real world I cannot see how anything in the objective performance would suggest an amplifier "far from clean" for the task of driving headphones.
FWIW, I own this Amp and find both the EQ function (I have the HD6xx version to match my Headphones) and the "3D" gizmo very useful.
I'm not interested in defending a purchase (I had this for a good while, since drop shipped) as there is no other Amp out there that includes these specific features that allow me to enjoy my HD6xx way more than a friend's HD800 + Sennheiser Amp that sell way more expensive. There is no need for me to defend my set-up (my friend got another Amp from Drop after he heard his HD800 on my 6xx Zen Can), it has zero altenatives for me.
But given the influence of measurements presented on this site on what people buy, I do think there is a duty incumbent on the operators of this site to ensure their measurements truly reflect the performance of devices tested and to adjust their practice to conform with best practice (using industry standard ways to present data would also be good, to allow test results to be compared to those of others).
I also feel that more consideration should be given to agency compliance both in terms of EMC and Electrical safety, as purchase recommendations of devices that are not tested and shown to be compliant with local regulations may either leave the buyer open to legal problems if the equipment emits too much EMI or even cause damage to health or property, if the electrical safety requirements are not passed.
I am sure we all remember the fake iPhone Chargers from Asia that would electrocute users, despite having all safety markings etc. applied in crass violations of relevant laws.
M.I.
Not much of the help here but saw some sources insinuating that the included psu is kind of crap, the ones sold by ifi overpriced, you can get third party with higher current (Amp) capacity to get better performance.I bought the sundara and ifi zen can. I have a pioneer u-05 already and got the zen can in case the pioneer wouldnt have enough power to drive the sundara.
Anyways I cant really tell a difference between the two except bass boost makes the sundara sound much more to my liking. Would 180mw be sufficent to drive the Sundara? I'm completely lost on electrical jargon regarding this.
Did anyone of you compare zen can with the cheap psu to the expensive xpower?
Not much of the help here but saw some sources insinuating that the included psu is kind of crap, the ones sold by ifi overpriced, you can get third party with higher current (Amp) capacity to get better performance.
I have no doubt in my case amp powered by internal AC power supply with class A up to 500mw (Asgard 3) beat up modest battery powered ifi Hip-Dac and it probably would be same for zen DAC. I use it at low gain up to 8-9 oclock plenty of headroom.
I wasn't crazy about it at first, but when I had it in hand, it's actually pretty cool.Al this hate regarding the design.....come on. It's not the same boring box design that's been around for eons. How often do you sit there and stare are your damn amp/DAC anyway? Shouldn't you audiophiles be listening to the music in any case? It's built really well and for the price, you simply cannot complain
Who knows, just use it as long as you need to.How many hours in a row can I run this amp without it being a detriment to its endurance?
I daily drive a combination amp with IEMs for gaming, meetings and yt, and use this unit to up the power when I want to listen to music with cans, but recently i've been pondering getting a balanced dac and daily driving the zen can, but im afraid that using it for several hours, sometimes days in a row might not be ideal for a class A amp that doesnt look to be using the chassis as a heatsink.
And also, i've become interested both in the aeon rt and the verum one headphones as openback and closed back main cans, but both units have very low impedance (12,5ohm and 8ohm respectively), has anyone used low impedance headphoes with the zen can and noticed anything bad in the sound quality? I know what Amirs analysis says about it but I would like to know about the "on ears" experience.
likewise, I combine SMSL M500 V2 with the Zen Can V2 and I am very happy with how it performs.Happy Friday all
As promised, here is my observation when pairing the SMLSL SU-9 with the Ifi Zen Can.
- Firstly, as mentioned many times, Zen Cans are very powerful and IMO can probably drive any HP or IEM easily.
- I did not observe any noise or distortion while playing a range of music at low and high volumes on my IEM and HP
- Like, the Topping EX5, the SMSL SU-9 DAC is very good. Clean and clear delivery of music. SU-9 has more thump and punch in its delivery than the EX5
- I did not get any ground loop buzzing on my Fiio FH7 IEMs like I did with the Bluesound Node 2i -- see my previous post this
- The FH7 sound very good with and without the XBass. With XBass they absolutely rocked and you could really feel the sub-bass and energy from some music.
- it is very easy to drive the FH7's from the Zen Can. I leave it on 0db Gain
- The DCA AEON RT Closed headphones also sound very good, full of energy and life (again XBass is enabled).
- I think the Zen Can compliments the SU-9 DAC really well - a good combination in my opinion.
I am not using any software or hardware EQ to tweak the sound of the AEON RT Closed as suggested by a few folks to sort out the lack of the sub-bass.
Overall, I am very happy with the Zen Can and will be keeping them to power my IEM and Headphone.
So, here is a question folks...... there are headphone amps that measured much better than the Zen Can like the Topping L30, Singer SA-1, SMSL SH-9, etc. If I pair any of these with the SU-9 will I see a night-and-day difference in music delivery?
The main differentiator for me using the Zen Can vs the other HP amps is the XBass feature. With this I don't need to worry about EQ'ing. I just turn it on and get amazing bass boost. I also know that the signal from the source is pure and unedited. Especially good for delivering pure MQA audio to the DAC (sorry if I have offended some folks) and to my ears.
Cheers
View attachment 145771
No, you're not the first to mention noise going away once RCAs are connected. I fear many or all Zen Cans have improper grounding plane. You connect 4.4mm which is balanced and does have ground pins, but you still end up with ground loop using a balanced DAC.The static noise stops the moment I either unplug the pentacon cable from the amp or if I plug either of the RCA channels to a single ended DAC (it doesnt even need to be on).
Also, if the pentacon cable is plugged in conecting to the d10 balanced, but the DAC is off i get lots of static noise on channel 3 (the balanced one), but, if i have a SE source plugged in thrugh RCA, if the source is off the channel is completly silent. Could this mean that the pentacon cable i have is shorted in some way? Is the how balanced connections work when theres no signal on them?