Sennheiser HD 490 Pro Headphone Review

[IdleTalk]

Funny... Amir and staticV3 are arguing the exact same point to each other but from a different perspective.

Basically amplifiers are voltage and current limited (staticV3 sometimes posts nice plots about this aspect).
Drivers can have different impedances, different impedance curves (over a frequency range) and different sensitivity.
Effectivity is determined by the sensitivity and impedance (which can differ per frequency band).

Most amplifiers have increased distortion in lower impedances. So... with the same sensitivity (but different effectivity) the higher impedance headphones are usually in the advantage. Also these will draw less power which, when playing really loud, could improve service life of a battery fed device on 1 charge.

When the output is voltage limited only a higher sensitivity headphone can play louder. Often these are also low impedance so there's that.

I don't think so.

It's only the ohms law what [staticV3 & me] and [Amir & Zolall] are agreed on,
but that does not make we are arguing the exact same point to each other.

What [staticV3 & me] are talking about and trying to apply ohms law is:
TOPIC A [effect of different impedance at different frequency of a same headphone+amp setup].

What [Amir & Zolall] is talking about is:
TOPIC B [driving different headphones with different amps].

The impedance curve of a specific headphone, the original question, obviously relates to TOPIC A.

The problematized fallacy is caused by confusing TOPIC A and TOPIC B.

But the TOPIC B, which is totally irrelevant, is constantly brought up,
either by still being confused, or to confuse readers to cover up the fallacy (don't want think that way though).

Again, please remind that the impedance curve of a specific headphone relates to TOPIC A, not B
I don't think I can do any better to make it clearer.

 

[ZolaIII]

@IdleTalk observe self amplifier impedance impact to a headphone xyz with it's driver type A, B, C and it's dependance in that regard to your defined topic A. As that's actually related to your defined topic B and well documented and measured from many sources it comes to A & B are correlated and it's a topic C entirely on.
Seriously universe is much more complicated than we individually want to admit and in that regard I retire from this debate further on, have a nice time and carry on if that's what make you happy.

 

[solderdude]

There is just confusion about the impedance peak, efficiency and sensitivity as well as loudness and some misconceptions about current and voltage limits and the usage of the word 'power'.

What Amir meant with 'most power is the lows' had nothing to do with Ohm's law but was about the amplitude of lows in music.
A different 'power' concept.
Amir and staticV3, as well as IdleTalk and I have a good grasp on Ohm's law so we are not talking past each other or don't understand Ohm's law.
The confusion is about the word 'power' and the addition of the word 'rising impedance'.

To get this back on the HD490 Pro... it does not require much power to play loud. It does require a bit higher than phone output voltage (0.5V for E.U and 1V for the rest) when you want to play really loud. A higher impedance just means less power is drawn than when an equal sensitivity headphone with a lower impedance (so different efficiency) headphone were used.
The HD490 only draws 100mW when pumping out 120dB SPL. On a E.U. phone it can only reach 103dB SPL and on a phone with 1V out 109dBSPL

To put in perspective:

For the HD490Pro to reach 110dB SPL at 1kHz 1V is needed = 7.7mW, for 120dB SPL = 0.077W
For the HD490Pro to reach 110dB SPL at 80Hz 0.9V is needed = 3.9mW, for 120dB SPL = 0.039W

 

[Graph Feppar]

I get that there's a relationship there, but so that I understand you well, are you suggesting that IMD can be directly predicted by harmonic distortion measurements ?

Yes! This is exactly what I am trying to explain, every non-linear thing in universe that can produce harmonic distortion can also produce intermodulation distortion. Every single amplifier, every headphone, every speaker ever made. You dont need to make IMD measurements, you can calculate and predict IMD if you know the THD.

Fun fact = human ear drum is non-linear and can make audible inter modulation tones.

 

[IdleTalk]

Seriously universe is much more complicated than we individually want to admit and in that regard

The very same complication of universe makes us to worship pure silver cables and flying spaghettis.
I have nothing against being happy with that because I admit I myself will probably never be completely free from that.

But measuring, theorizing and calling it "science" is to deconstruct that complication, not admitting it as a whole.
So please let's not bring ad hocs and third topics to blur the point.

Being scientific just before encountering some scientific objections and then move on to a rather holistic approach, doesn't seem honest.
I'd rather indulge to be honestly unscientific than to be selectively scientific for personal happiness.

That means, if you wish me to stop, I'll stop for communities sake.
There's no need for you to retire.

 

[IdleTalk]

What Amir meant with 'most power is the lows' had nothing to do with Ohm's law but was about the amplitude of lows in music.

That was the most favorable (mis-)interpretation possible but (un-)fortunately was ruled out,

1) by his post #105, where he brings out driving different headphones with different amps

2) by the statement in his T1V2 review, where he said, "The T1 is stated to have 600 ohm impedance which is already high. Turns out that the real impedance can be much worse:"

P.S. You seem to know what is right and what is wrong (scientifically). So the problem is narrowed down to human communication: why should his obvious fallacy be so well hidden and defended?

 

[staticV3]

What Amir meant with 'most power is the lows' had nothing to do with Ohm's law but was about the amplitude of lows in music.

I also gave Amir the benefit of the doubt, but later on it became clear that he misinterpreted the 490 Pro's impedance peak as a negative aspect that hampers driveability, when in fact the opposite is true.

So I joined @IdleTalk in trying to explain why.

 

[ZolaIII]

@IdleTalk I am not retiring because you are right or bringed experimental facts to light but because I have smarter things to do. It's funny to observe how you (not you individually but, wider present crowd in this discussion) hypotheses things and then try to call it case on your internal doxa introspection or try to put them in some absurd groups again based on what you thought other people had in mind and if it isn't there that worse for the facts (Hegelian closed deductive system). I have more knowledge in logic and methodology and enough to know better and not argue about it for those closed to the facts. When asked to pick a source of measurements by your likings about self impedance of an amplifier and it's impact on FR of different headphones with different driver types (dynamic, armature, planar, electrostatic...) you're answer whose about magic Mike, flying circus and alchemy. Blaming it on me as a complete different and fictional subject even I played on your little closed classification system only to show you it's a nonsense by its own. Cases that stand out of expected in scientific research are called anomalies and delth with individually. Either explaining why their deviate or bringing down fundamental starting hypotheses as wrong. Not either or. We are far from statistically significant number of samples in audio studies and even more from classification and dealing with anomalies.

 

[IdleTalk]

I have more knowledge in logic and methodology and enough to know better and not argue about it

Very impressive way to speak "I'm done playing with you".
Thank you.

 

[peniku8]

Sensitivity as it is, impedance x2 = W/2 = - 3 dB. Now read solderdude note and take a look at impedance plots. You you need to hit it harder.

Voltage Sensitivity is a function of SPL over voltage. Impedance is not a part of this function, hence it is entirely irrelevant from this metric's point of view. Maybe you're confused because the sensitivity figure is often given referenced to a nominal impedance figure, for example replacing 2.83Vrms for a 1W@8R figure, which is technically incorrect, since "99dB/W" is a metric for efficiency, which as mentioned before, sensitivity is not. If the voltage sensitivity is 100dB/V, then the device will produce 100dB with a given input voltage of 1V, no matter if it's a 2 Ohm subwoofer or a 600 Ohm headphone (and as long as your amp will be able to supply the right about of current).
Regarding Amir's and static's discussion, I think this is more about semantics rather than fundamental understanding, since I've seen Amir misuse the term 'power' in the given context, when he really meant 'voltage'. I get the sentiment tho that headphone amps that can supply a large voltage swing could be considered 'powerful' even if their actual power isn't that great. Amps like this would typically fare really well in high impedance tests but fall behind in low impedance tests. I'm sure @staticV3 can think of such am amp off the top of his head
I don't know how much of a topic power compression (and thus distortion) is in headphones, but this could be why the HD490 seem to be able to deliver cleaner bass than the HD650 for example. Higher efficiency=lower power compression.
It would be quite cool to have a direct apples to apples comparison between the HD490 and HD650 aka both EQ'd to the same target and then comparing compression and the resulting distortion of the two. I don't know enough about headphones to interpret whether the occuring compression would be power compression or due to the driver running out of linear travel, but I'm sure somebody here is knowledgable enough to make this call.

 

[solderdude]

With such a high impedance, you are voltage limited. At that impedance, you get far less power out of your source if it doesn't have high voltage drive.

The easiest to drive headphones are the ones that are low impedance and have high sensitivity. Then even a dongle with low output voltage can drive them. Boost the impedance up high and the same source will struggle to deliver power.

That was the most favorable (mis-)interpretation possible but (un-)fortunately was ruled out,

It depends on how you read the quote above.
The only error Amir makes is what I bolded.

What I got from Amir's reply is that low impedance and high sensitivity do not require powerful amps. (easy to drive)
Higher impedance headphones usually are lower sensitivity and require a higher drive voltage (usually this also means a higher power rating of the amp itself, aside from some OTLs.

From a technical p.o.v., as I explained earlier already, higher impedance headphones put less strain (current wise) on the amps and thus usually exhibit lower distortion as well.

So it depends on whether you look at it from a technical p.o.v. (Ohm's law) or from a practical p.o.v. (consumers and how they view the word 'power')

P.S. You seem to know what is right and what is wrong (scientifically). So the problem is narrowed down to human communication: why should his obvious fallacy be so well hidden and defended?

Yep, it is a matter of communication and viewpoints. This was my point in stating Amir and staticV3 were arguing about the same thing from different p.o.v.
I can easily see both viewpoints and know that Amir has at least basic understanding of Ohm's law.

 

[staticV3]

Yep, it is a matter of communication and viewpoints. This was my point in stating Amir and staticV3 were arguing about the same thing from different p.o.v.
I can easily see both viewpoints and know that Amir has at least basic understanding of Ohm's law.

Amir's train of thought seems to be:

You need power to produce SPL->An impedance peak impedes power-> Therefore you need an even more powerful Amp to overcome that.

And in that light of constant power, an impedance peak could indeed be seen as a negative aspect that makes a headphone harder to drive.

But that premise is fundamentally wrong.

Amir's frequency response graph was captured at constant voltage, not constant power.

An Amp playing music puts out a diffuse field-ish voltage spectrum, not power spectrum.

The actual power spectrum is a secondary attribute that just follows the Amp's output voltage response and the headphone's impedance response.

70-80Hz looks to be the 490 Pro's Eigenfrequenz where it is the most efficient, so it barely needs any power to play that frequency.
Less, not more.

 

[Chocomel]

It's of a hd600 but the premise is the same

 

[IdleTalk]

It depends on how you read the quote above.

Yes. The quote is a part of a context, and there should be a right/wrong way to read according to the context. The matter of communication and viewpoint has caused the secondary problem and obscures the primary one.
Let's see.

Impedance is middle of the road but rises in bass to over 200 ohm (where you need most power)

This sentence is right above the impedance response graph in this review and describes the impedance response of a certain headphone: HD490pro. Therefore it's about the TOPIC A [effect of different impedance at different frequency of a same headphone+amp setup]. This is what I questioned.

With such a high impedance, you are voltage limited. At that impedance, you get far less power out of your source if it doesn't have high voltage drive.

The easiest to drive headphones are the ones that are low impedance and have high sensitivity. Then even a dongle with low output voltage can drive them. Boost the impedance up high and the same source will struggle to deliver power.

This is Amir's answer to my question. And this is very "the quote above" you mentioned, which needs to be read correctly.

But you can see that this is not an answer for TOPIC A that I asked. It is another statement about TOPIC B [driving different headphones with different amps]. Amir has not yet answered anything about TOPIC A.

More problematic is that some of you seem to be covering this up with certain communication skills like "both are right", because his statement about TOPIC B is right. Now I look like an asocial interrogator who keeps asking Amir's stance on TOPIC A.

In fact there's no need to "interrogate" anyone about anything, because his stance on TOPIC A is clearly stated in his T1V2 review. A very erroneous one though.

I was just shocked that, after all that years, that error still persists and no one corrected it in this big community of "audio science".

Mr. Solderdude, you've shown you know what is technically right about TOPIC A.
That means you can also communicatively discern TOPIC A from B.

But I don't understand why you choose to be communicatively wrong, implying that there would be nothing right/wrong in the world of communication and viewpoints, while the technic itself is in fact nothing more than a particular type of communication.

Such communicational failure, intended or not, can be worse than any technical misunderstanding.

----
For your convenience,
TOPIC A = effect of different impedance at different frequency of a same headphone+amp setup = the primary problem
TOPIC B = different difficulties of driving different headphones with different amps
the secondary problem = communicational failure = confusion of TOPIC A & B

 

[IdleTalk]

It's of a hd600 but the premise is the same
View attachment 365601

very nice!

dB/mW and dB/mA goes parallel because voltage is constant.
dB/V is not parallel with other two because impedance is varies in low frequency.
Same voltage input into 100Hz and 1kHz produces almost same SPL.
Same wattage & current input into 100Hz and 1kHz causes higher SPL at 100Hz than 1kHz, because there lies the impedance peak.

 

[Robbo99999]

"Impedance is middle of the road but rises in bass to over 200 ohm (where you need most power)"

No, please. An impedance peak is where least power is needed.

The frequency response is a response in SPL against AC in varying frequency and constant voltage.
Since the input voltage is constant, current becomes lower where impedance gets higher, resulting less power consumption.

And it is quite shocking that no one (except @thewas in comment #4) has questioned about the channel imbalance with producing pads, which is not seen with mixing pads. What was the culprit? The drivers? pads? or placement? This should be easy to find out if the reviewer tried swapping L/R of producing pads. I hope to see some scientific reasoning like such than procrustean job.

Pads can definitely cause channel imbalance if the left & right pads are not totally identical (manufacturing tolerance & additionally potential installation problems), I've noticed this when measuring headphones & replacing the pads on them (measured using my miniDSP EARS rig). Placement variation of the headphone on the rig can also cause some channel imbalance - the most accurate way to get rid of that is through multiple reseats of the headphone on the measurement rig and then averaging the measurements to remove some of the placement variation variable. Depending on the measurement rig and the headphone being measured though sometimes you can still get an accurate channel balance measurement from just one measurement - if you have a rig & headphone that doesn't vary much with reseats. Regarding the HD490 Pro measured here I'm expecting the drivers are matched quite well given that the mixer pads are showing good channel matching, so it does tend to point to the producer pads being maybe not totally identical between left & right, or it could just be a measurement problem given Amir's only showing one measurement, but he tends to try to show a representative measurement, but you know.

 

[Makahl]

the most accurate way to get rid of that is through multiple reseats of the headphone on the measurement rig and then averaging the measurements to remove some of the placement variation variable

Isn't that the usual procedure for any operator? Regarding positional variance, Aregina (also using a GRAS 45CA) exhibited a clear difference that could explain the imbalance if the position on the test fixture differed slightly from a single measurement. It'd be nice if Amir swapped the earpads and repeated the measurements to figure out the real culprit.

 

[Robbo99999]

Isn't that the usual procedure for any operator? Regarding positional variance, Aregina (also using a GRAS 45CA) exhibited a clear difference that could explain the imbalance if the position on the test fixture differed slightly from a single measurement. It'd be nice if Amir swapped the earpads and repeated the measurements to figure out the real culprit.

View attachment 365649

Well, it could be that, but it depends on how the rig & the headphone interact - sometimes it can be very stable & very easy to measure from reseat to reseat and most of my headphones that I've measured have been super easy to measure on my miniDSP EARS rig - in which case a single measurement would be really quite representative of the real channel matching of my headphones, but it will depend on how easy it is to get stable measurements on Amir's flat cheek GRAS rig. You're right from the pic you show that placing the headphone to the extremes of the possible positions around the artificial ear will show some variation in most headphones, albeit some headphones will differ more than others when placed to the extremes. My own experience with headphone measurement is that placing roughly central is best, and I try to look down at the artificial ears as I'm placing the headphone to try to avoid misshaping the ears - because afterall when we wear headphones ourselves you will generally move the headphone around so that there is the least touching against your ear. Amir is pretty unkeen on the whole multiple reseat & measurement front, and he's more about probably doing a quick bit of placement experimentation to see what is a normal measurement for the headphone after throwing it on the rig, and then he'll display that measurement. I think in general a person could feel quite confident in HD490 channel matching given what's been shown with the mixer pads in Amir's measurement, and solderdude over on diyaudioheaven saw good channel matching with the HD490 too:

HD490 PRO

back to Sennheiserback to measurementshome published: Jan-27-2024, edited Mar-04-2024 NO SMOOTHING is applied to the shown plots. Most measurement sites have some smoothing applied which ‘irons fla…

diyaudioheaven.wordpress.com

And solderdude's measurements aren't averaged either, so they're singular & I guess he tries to make sure it's a representative measurement rather than an outlier. Maybe there is gonna be more variance in the Producer Pads just by their design, that's a maybe. (It's one of the main reasons I bought my miniDSP EARS measurement rig - so I could make sure channel matching was perfect through per channel EQ, or just be choosing headphones that measured with good channel matching - it's always a bit of an unknown otherwise, and there's enough uncertainty surrounding headphones, lol!)

 

[solderdude]

The weird part here is that Amir measured a level difference with producer pads (which remained not investigated)
If it were positional issues then it would likely not affect the entire frequency range but rather what is shown in the plot above.
When it is a seal issue I would expect only the lower frequencies to drop.
The height difference between the producer pads could be a reason and 0.5dB or so would be about 1mm difference in height which might not be easy to see.
Below the difference when the HD490 producer pad is compressed:

Pad thickness would be the most likely reason for the channel imbalance, swapping pads would be a useful and easy test.
Chances are Amir already sent the headphone to its owner...

I did not find a L-R balance issue with both pads in overall level.

 

[usern]

Which are more comfortable - 490 pro or HD800S?