[slapo]

Hey! I've had these for a few days now, these are my absolute first "proper" headphones and for the price I paid (50$) I'm absolutely amazed. At first I was terribly afraid that my pair came faulty with perceptible channel imbalance (Left speaker seemed 3-4dB louder). In the process of troubleshooting I learned that the fault lies in... my ears. Learned something new, I suppose.
As I said, I'm SHOCKED that the K245's are so under the radar. Basically nobody is talking about them online, there are very little reviews and this forum is just about the only sizeable discussion about them. It feels like such a steal to be able to buy these for such a low price. For the first few hours I was giddy with excitement, coming from KZ ZSN Pro X IEM's, these headphones blew me away.
I recently purchased a pair of Beyerdynamic DT990 Pro's for 120USD as the audiophile bug bit me good and I thought to myself: "Well, if this is what 50$ can sound like, what about 120$?". The DT990's are heavily recommended headphones everywhere online, so I expected to get wowed even more considering the investment!
Sadly, the DT990's came with an actual factual channel imbalance (as confirmed by multiple friends of mine), which is a bummer, but there was science to be conducted!
And so I began comparing the two. Considering the prices of both of these headphones, the K245's seem to punch far above their weight for music, media consumption and gaming. The DT990's are almost unlistenable to without EQ, and as such the K245's eat them for dinner if no EQ is available. The DT990's do put up a much better fight if they're EQ'd, and at that point... things get difficult for the K245's.
One thing I'm wondering about though; there are tens of guides as far as EQing the DT990's and more popular headphones. No such thing for the K245's. Does anybody have any recommendations as far as EQing them goes, then? I saw FR measurements for them earlier in this thread, but, alas, I'm a grasshopper when it comes to headphones and struggle to come up with something on my own.
Also, do the K245's represent the famous "AKG-sound" you hear about online so much? If yes, I'd say I rather like it, and am inclined to look for it in future headphones!

Bear in mind that both the K245 and K275 used to retail for around £120 before the K371 and K361 came out. The latter two are, frankly, pretty poorly built.
That aside, price isn't necessarily an indicator of sound quality (well, more often than not, it won't be).

If you'd like to equalise the K245, you'll probably want to start with a question to yourself - what is it you would like to change about their sound?
If there's nothing, the just enjoy them as they are. There's nothing wrong with that.

I'm not actually sure what the AKG house sound actually is, considering that earlier models were generally pretty lean on bass and (perhaps excessively) fairly abundant in highs.
Things somewhat changed with the K550 (mostly neutral with an earlier brighter and a somewhat less bright later tuning), K702 (decent bass, highs somewhat harsh), K712 (I haven't heard that one, but should have more bass than K702, but not necessarily be tidier in the treble).
Then K245 and K275 came and they were sort of like variations on the K550.
Now we're at K361 and K371, which are highly praised (and which I don't like).

Of course, they also had portable headphones which had a tuning of their own, from what I'd heard, but I haven't heard any of those.

 

[TheDunkPapa]

Bear in mind that both the K245 and K275 used to retail for around £120 before the K371 and K361 came out. The latter two are, frankly, pretty poorly built.
That aside, price isn't necessarily an indicator of sound quality (well, more often than not, it won't be).

If you'd like to equalise the K245, you'll probably want to start with a question to yourself - what is it you would like to change about their sound?
If there's nothing, the just enjoy them as they are. There's nothing wrong with that.

I'm not actually sure what the AKG house sound actually is, considering that earlier models were generally pretty lean on bass and (perhaps excessively) fairly abundant in highs.
Things somewhat changed with the K550 (mostly neutral with an earlier brighter and a somewhat less bright later tuning), K702 (decent bass, highs somewhat harsh), K712 (I haven't heard that one, but should have more bass than K702, but not necessarily be tidier in the treble).
Then K245 and K275 came and they were sort of like variations on the K550.
Now we're at K361 and K371, which are highly praised (and which I don't like).

Of course, they also had portable headphones which had a tuning of their own, from what I'd heard, but I haven't heard any of those.

I was not aware that the k361 and k371 came after these, it seems bizzare considering their poor materials and somewhat divisive opinions on their sound!

And yes, I'm well aware that sound isn't necessarily corelated with price, it just seems that there are many people in this hobby that like to dissuade anyone from even TRYING headsets under 150$. This was a major problem when I was shopping around, literally everybody kept recommending things barely touching the 150$ mark, often exceeding it, the space below 150 and especially below 100 is a difficult subject, apparently.

The thing I'm thinking of correcting with EQ on my pair is something I was made aware of about them only after listening to the DT990's. As I said, I think the K245's sound outrageously good for their current price, but now I wish they could maybe sound a tiny bit warmer and a tiny bit less... congested? I think this has to do with the bass on these, feels like it can mask some higher frequencies at times. Having browsed this threat many times already it got me thinking about swapping out the pads for something more spacious and further away from the drivers to alleviate this issue. But I also tried my hand at EQ with APO and... I don't know how to go about these headphones! It feels like it's extremely easy to screw up their sound with EQ. My cheap KZ's (which I still adore for introducing me to this hobby), you could basically do whatever to them in any EQ software and soon enough you'd find something that sounds better to your ear. Even the DT990's, which as I said in my previous post are nigh unusable without EQ (to me), are EQ friendly (especially with advice from reddit's oratory1990 subreddit) in comparison. The K245's just get screwed up in one way or another if I try to warm them up and give them more air.

Maybe I'm just trying to do something these headphones were not born for? These two pairs of headphones are different beasts and I still struggle to pick out a favorite between them. I'll have to decide in a few days time which one stays and which goes. I'm not looking forward to making that choice, haha!

To any future readers, if you're on the fence about buying the K245's, and they're for sale at around 50$, just get them and prepare to be surprised.

 

[Corti]

@TheDunkPapa
Glad to see another one liking this headphone. i've recently bought my 3rd unit
My first one broke after fiddling too much with the replacement pads because the foldable joints are not very strong. Fortunately, got it refunded.

So i bought a new one for 75€. As it happens i like these headphones a lot but knowing they are discontinued and are not of the best durability, i snatched another unit for 50€ as backup Guess i dont have to worry anymore now.

Now owning 2 units, i again swapped stock pads on one (like in my post some pages back) while leaving the other one untouched. I mainly did replace pads for better comfort since stocks are really shallow and touched my ears, also become sweaty after longer listening sessions due to inferior breathability. YMMV and you may get along with the stocks just fine though.

However the sound does change drastically with different pads. Bass becomes more modest, Treble and Uppermids increase with those specific perforated brainwavz pads, so EQing becomes mandatory.

Good thing for me is i like the process of learning how to EQ and change the way the headphone sounds depending on what genre and recording i listen to.
You also learn a lot about your music or sound in general while studying Frequency response (but since i am a scientist i kinda love to analyze a lot of things just because )

So if you want to use EQ to your advantage, @slapo has already pointed out that its a very subjective process where you have to find the frequencies that bother you
or which you want to be boosted, on your own. I use a visualizer that shows me which frequencies are played during sound playback, so that i can see where e.g. the unpleasant snares are located or what bassnotes sit in the sub or in the midbass region, so that i could equalize accordingly.
SOmetimes it's just the artist that mastered the music on a whole different system (or that was intended to be played in clubs on big speakers) that may never sound right on your headphones i think.

Youve mentioned oratory1990, i use a lot of his data to get a general idea on how the hp is tuned and he provides EQ settings to match the Harman curve..
You may want to start there and read about that, there is a lot of articles (if u havent already). fun fact is that the harman curve is almost exactly how the AKG K371 is tuned, which a lot of users find pleasant. I also own the 371 but rarely use em. While the tuning is fine, i dont 'feel' the music with those.. I plan on selling them in the near future.
On the other hand u complained the K245 occasionally has too much bass bleeding into the mids and sounding congested. As i found, this is typically the region of around 150-400 Hz which i also lower with most of my hp's. If overdone though the sound can become somewhat hollow, just try and see for yourself.

I think the K245 is not capable of becoming more airy or get better imaging by a good margin.. That is probably the compromise of their design. If that is what you are looking for, I'd strongly recommend you to look into Planar Magnetics, those do a terrific job on that matter. There are some cheaper entry models from Hifiman for instance. Ive owned the HE400i 2020 and they are great, but the +200$ for the Sundara is totally worth it.

Back to the AKG, I feel the K245 has great bass in general (the semi-open design is just awesome in this case). Also the treble is not too bright (still need to lower it a bit).
I think the weakest area with this hp is the midrange. This has also been mentioned before in this thread i believe.. With certain tracks it gets very shouty (probably 1k-4k Hz area) while it feels missing on others. So, no easy equalizing there..
Thus the K245 has become my main deep electronic music hp, since those genres do not focus/rely on midrange too much. It might also correlate to the just-mediocre timbre..
So yea, with the K245 i always keep my EQ app at hand to dynamically adjust FR to what im listening to. I still want to try the Schiit Lokius some time when it becomes available in the EU again.

Soo, quite a long post, sorry for that
Anyway have fun with your headphones and enjoy understanding how they work

Edit: i just remembered a very thorough tutorial for beginners on how to eq from this forums' sound science section:
https://www.head-fi.org/threads/the...ost-ideal-sound-for-non-professionals.796791/

 

[TheDunkPapa]

@TheDunkPapa
Glad to see another one liking this headphone. i've recently bought my 3rd unit
My first one broke after fiddling too much with the replacement pads because the foldable joints are not very strong. Fortunately, got it refunded.

So i bought a new one for 75€. As it happens i like these headphones a lot but knowing they are discontinued and are not of the best durability, i snatched another unit for 50€ as backup Guess i dont have to worry anymore now.

Now owning 2 units, i again swapped stock pads on one (like in my post some pages back) while leaving the other one untouched. I mainly did replace pads for better comfort since stocks are really shallow and touched my ears, also become sweaty after longer listening sessions due to inferior breathability. YMMV and you may get along with the stocks just fine though.

However the sound does change drastically with different pads. Bass becomes more modest, Treble and Uppermids increase with those specific perforated brainwavz pads, so EQing becomes mandatory.

Good thing for me is i like the process of learning how to EQ and change the way the headphone sounds depending on what genre and recording i listen to.
You also learn a lot about your music or sound in general while studying Frequency response (but since i am a scientist i kinda love to analyze a lot of things just because )

So if you want to use EQ to your advantage, @slapo has already pointed out that its a very subjective process where you have to find the frequencies that bother you
or which you want to be boosted, on your own. I use a visualizer that shows me which frequencies are played during sound playback, so that i can see where e.g. the unpleasant snares are located or what bassnotes sit in the sub or in the midbass region, so that i could equalize accordingly.
SOmetimes it's just the artist that mastered the music on a whole different system (or that was intended to be played in clubs on big speakers) that may never sound right on your headphones i think.

Youve mentioned oratory1990, i use a lot of his data to get a general idea on how the hp is tuned and he provides EQ settings to match the Harman curve..
You may want to start there and read about that, there is a lot of articles (if u havent already). fun fact is that the harman curve is almost exactly how the AKG K371 is tuned, which a lot of users find pleasant. I also own the 371 but rarely use em. While the tuning is fine, i dont 'feel' the music with those.. I plan on selling them in the near future.
On the other hand u complained the K245 occasionally has too much bass bleeding into the mids and sounding congested. As i found, this is typically the region of around 150-400 Hz which i also lower with most of my hp's. If overdone though the sound can become somewhat hollow, just try and see for yourself.

I think the K245 is not capable of becoming more airy or get better imaging by a good margin.. That is probably the compromise of their design. If that is what you are looking for, I'd strongly recommend you to look into Planar Magnetics, those do a terrific job on that matter. There are some cheaper entry models from Hifiman for instance. Ive owned the HE400i 2020 and they are great, but the +200$ for the Sundara is totally worth it.

Back to the AKG, I feel the K245 has great bass in general (the semi-open design is just awesome in this case). Also the treble is not too bright (still need to lower it a bit).
I think the weakest area with this hp is the midrange. This has also been mentioned before in this thread i believe.. With certain tracks it gets very shouty (probably 1k-4k Hz area) while it feels missing on others. So, no easy equalizing there..
Thus the K245 has become my main deep electronic music hp, since those genres do not focus/rely on midrange too much. It might also correlate to the just-mediocre timbre..
So yea, with the K245 i always keep my EQ app at hand to dynamically adjust FR to what im listening to. I still want to try the Schiit Lokius some time when it becomes available in the EU again.

Soo, quite a long post, sorry for that
Anyway have fun with your headphones and enjoy understanding how they work

Edit: i just remembered a very thorough tutorial for beginners on how to eq from this forums' sound science section:
https://www.head-fi.org/threads/the...ost-ideal-sound-for-non-professionals.796791/

Thank you for your advice and input!
Alas, after hours and hours of testing, I decided that I prefer the warmth and airiness of my DT990's instead of the colder sounding K245's.
One thing hasn't changed for me, I recommend the K245's for people to check out around the 50-60$ budget mark as I still think there are plenty of people who will fall in love with them just as I did at first listen, with the caveat that replacement earpads are definitely in order. Even a couple of millimeters of extra space helps the K245's sound much better, IMO. And well, maybe the cable, but who knows, I guess some people like coiled cables, I don't.
I have also slightly gotten better at EQing on my own over these few days, mainly by trial and error, haha, but I guess there's no beating just getting in the mud yourself!
I also think your observation of EQing the K245's depending on what you're listening to to be interesting, but probably correct. I have found this as well during my time with them, but having to adjust back and forth based on my music choice for the evening proved to be cumbersome.
If my budget were slightly larger and freer, I think I might just have kept both pairs, but money doesn't grow on trees, especially in this hobby! If anything, I'm grateful that I had them because of how much I have learned because of them, both in technical skills and what my own preferences are. The audio journey is a wonderful thing!

 

[Precaud]

Just FYI, it appears that AKG has either resurrected the K245, or has some inventory left to sell. They're available on the US AKG website for $89 with free shipping.
https://www.akg.com/professionalemployeesale/3405H00020.html

I have completed my K245 mods to address its issues without resorting to EQ, have been living with them for a week now, and am extremely pleased with the results. I will post them as soon as I can take photos and do the writeup.

 

[Corti]

Just FYI, it appears that AKG has either resurrected the K245, or has some inventory left to sell. They're available on the US AKG website for $89 with free shipping.
https://www.akg.com/professionalemployeesale/3405H00020.html

Thats curious indeed. Over here the last units disappeared about a half a year ago from regular merchants. German AKG site still says unavailable. Only listed on amazon since then, most of the time for >100€.
But in the end it's probably Samsung doing Samsung-things.

I have completed my K245 mods to address its issues without resorting to EQ, have been living with them for a week now, and am extremely pleased with the results. I will post them as soon as I can take photos and do the writeup.

Nice, looking forward to reading it.

 

[Precaud]

This is installment #1 detailing the mods I've made to the AKG K245's. I thought it useful to explain some of the reasoning behind the choices made, so it will be dribbled out in segments over the coming days.

We'll start with an overview of the various parts (which I call driver, baffle, cup, and earpad), and make note of important details and dimensions.

No doubt you've seen pics of modern dynamic headphone driver diaphragms. They consist of a small voice coil attached to plastic diaphragm with an equally-small center "dome", and a large spiral-corrugated-looking "surround". The surround's "spiral pleats" permit the diaphragm to have a low self-resonance frequency, and allow as much of the surround (which has the largest surface area) as possible to move linearly with the voice coil. This last point is important, because the edge of the surround attaches to the baffle, where it obviously does not move. So it is this transition from "pistonic motion" to "motionless edge termination" that causes the most problems in a given driver design.

The K245 driver diameter is about 48mm, or 1.9". From my years designing speaker systems, with so many variables in materials, geometry, etc., there are are no perfect formulas which precisely predict what resonant frequencies will be created. But these three are a good place to start:
SOS / PI * Radius = approximate upper piston frequency (4.5 kHz)
SOS / Circumference = edge-related breakup frequency (2.26 kHz)
SOS / Diameter = radiation pattern narrows, causing peaking (7.1 kHz)
where SOS = Speed Of Sound.

Remember these frequencies, because as you can see, they are right in the range where the K245 frequency response is the most nonlinear. And, if you look back at the impedance curve I posted a few months ago, it is where there are impedance irregularities as well, suggesting mechanical resonances.

The nine holes around the back of the driver are directly under the back side of the surround, and emit sound which is reverse polarity (out of phase) with the front of the driver. Some of this "backwave energy" escapes though the slots and vents in the cup, and some of it is used to tune the bass-mid frequency response at the ear, by letting some of it leak through ports in the baffle. In my opinion, this is one of the greatest strengths of the K245. The engineers did a superb job of tuning this headphone at low to mid frequencies, it is among the best I've heard in that respect (without the stock earpads).

 

[Precaud]

Here is the inside of the left cup, 60mm inside diameter. The right-side cup is the same except for the moulding that houses the min-XLR jack, and the small pcb. Most of the inside is reflective plastic material; there are only 3 smallish areas that allow the backwave to escape through the back of the cup to the outside.

The baffle attaches to the cup via the six posts near the perimeter. These posts stand the baffle off of the cup, maintaining a few mm gap all around. Some of the backwave escapes through this gap and exits through the holes you see along the edge of the cup rim. Some of it goes through the tuning ports on the baffle, to the ear. And some reflects back to the driver.

Reflected arrivals interact with the direct waveform and form a "comb filter"-like frequency response. I estimate the trip from the driver to the back of the cup and back is about 2.5", maybe a little more. This corresponds to a first peaking frequency at about 5.4 kHz. This creates yet another tonal anomaly right in the same lower-treble range as the driver resonances.

The first mod is to damp this reflection at high frequencies, but do so without interfering with the low-frequency tuning. You definitely do not want to use anything like loose fiberfill material that would interfere with the tuning. A fairly thick piece of polyester felt against the back of the cup works well. Arrange it inside the cup so it stays against the back wall and doesn't flop up and block the baffle ports or the baffle gap. Cut a 60mm circle and slice a slot in it for the wire to feed through. The notch cut out for the headband swivel area may not be necessary; I did it just to make sure the felt didn't block the gap up there.

We'll get into the different grades of felt later. This is the thickest stuff I've seen sold in sheets.

 

[Precaud]

Word has it that the K245 were designed in Austria before the Samsung acquisition. It basically says so right on the front of the baffle. I am convinced that those Austrian AKG engineers had nothing to do with choosing the earpads or the cable, both of which are horrible. Those were chosen after the fact by marketing types totally ignorant of the audible consequences.

The first pic below shows the front of the baffle. The rings of holes in the center are directly over the driver. The six sausage-shaped ones above and below the driver area go to the cup behind the driver. They are the "backwave tuning ports" I referred to earlier. Outside diameter of the ports is 59mm. Keeping these vents open and unobstructed is absolutely essential.

The second pic shows the stock earpads. The dense memory foam inside is nice. But the center hole is only 50mm! I used a heat gun and glass bottle trying to enlarge them and got another mm. But the basic fact is, the stock earpads completely block the tuning port holes. Idiots! (If I were one of the original design engineers, I'd be pissed that someone in marketing ruined my painstaking work by using these pads.) This is a major cause of the "bloated bass" these 'phones have.

What we need are 95mm diameter pads with 60mm inner hole and dense foam (often called "memory foam"). Unfortunately, such a thing does not appear to exist. I have contacted several earpad vendors on eBay and AliExpress and none of them have anything close. BrainWavz says all of their 100mm pads have 55mm holes. In order to get a 60mm inner hole from anyone, you need to buy a 110mm earpad. I did, and it doesn't stay on, it slips right off. But it let me hear the impact of keeping those vents unobstructed.

The closest I could find that work well are the velour velvet pads for the Senn HD560 sold by Wang YiFei on eBay:
https://www.ebay.com/itm/251387223412

They are 105mm OD and barely stay on. He said the ID was 58mm, but the pair I received were closer to 55 (see 3rd pic). Nevertheless, they sound MUCH better than the stock pads, though just a tad light on the bass, because the foam is not very dense. I wrapped a layer of cloth tape around the outside of the pads, which helped the bass.

So the second mod is, earpads with the correct inside diameter so the careful tuning is not ruined. The ideal candidate has not yet been found; I consider the Wang HD560 pads a temporary compromise. If you know of any pads that might fit the bill, please post it.

 

[Precaud]

A word about felt material. Most of the stuff sold in 9" x 12" sheets at hobby shops around here is very thin, maybe 1mm. Held up to a light, you can see right through it. This material is not suitable for these purposes.

The better stuff is more than twice that thickness and more dense. It comes in bolts and is sold by the yard at fabric shops. I got mine at our local JoAnn Fabrics store a few years ago when looking for material to blacken the interior of telescope tubes.

This pic shows three different grades of poly felt. Left to right; the typical thin 1mm felt, some medium thickness stuff, and the nice thick, dense material which is needed for inside the cup, and for the main event, which is yet to come...

 

[Precaud]

With a minimum impedance of 43.5 Ohms, the K245 is definitely a low-impedance headphone, and cable quality is important. I'm not talking about insignificant stuff like OFC (Oxygen-Free Copper), silver coating, or many other complete BS marketing hype. This is not the place for lengthy explanations as to why, but for low-impedance cans, the two most important parameters are 1) the series resistance of the signal and ground conductors, and 2) the loop inductance of a signal conductor with ground (made by connecting a signal wire to the ground lead on one end and measuring the inductance at the other end). Having separate ground leads for each headphone back to the amp would be nice, but the K245 uses a 3-wire, common ground scheme.

The stock K245 coil cable is the worst-measuring (and sounding) headphone cable I've ever experienced, with 0.75 Ohms series resistance for signal and ground leads, and a whopping 3.1uH loop inductance at 10kHz. The high inductance is largely caused by the coiled construction.

I bought an aftermarket 150cm cable with mini-XLR connector from a reputable source. It wasn't quite as bad but still very thin-sounding. Signal wire R was 0.72 Ohms, ground wire R was 0.55 Ohms, and loop inductance 750nH.

So I cut the ends off of the stock cable and spliced in 6 ft. of wire from an old, fairly flexible VGA cable (labeled Shielded Computer Cable) I had laying around. Inside it has six 26 ga conductors, with a foil shield and drain wire. I gave each conductor two wires, plus the ground lead also got the foil shield and drain wire. The difference was dramatic. Less "edgy", much more relaxed, open and dynamic. Signal wire R was 0.15 Ohms, ground wire R was 0.11 Ohms, and loop inductance was barely measurable at 50nH. These are very good numbers.

If you're buying a pre-made cable, ask the vendor for those parameters. If they don't know or can't measure them, they have no business selling cables. Buy somewhere else.

Here again is the K245 impedance measurement.

 

[megadud]

All fascinating so far, thanks @Precaud !

 

[130rne]

Agreed! @Precaud Keep it coming. This is my first pair of high end headphones and I'm learning a ton. Great job!

 

[Precaud]

Thanks guys, I'm glad you find it useful.

We can't really talk about cables without mentioning headphone amplifiers (HPA's). Nor can we dive deeply into HPA design here. But there is one thing that needs to be discussed that is particularly relevant to driving low-Z headphones (Z = shorthand for "impedance").

Low-Z phones have become popular because they can be "driven" from low-voltage, portable devices. "Driven" in this sense merely means, they can play loud enough and with quality that is "good enough" to be useful.

Most amps use some amount of negative feedback to lower distortion. Simply put, it does this by sensing the output and summing an inverted (negative) version of it with the input signal, thereby correcting any differences. From most POV's (mine included), the more negative feedback available, the merrier.

Regardless of the amplifier type (Class A, Class AB, tube, whatever), if it is a feedback amplifier, we want to protect that system so it senses the load accurately.

This is especially important because dynamic headphone loads are 1) complex i.e. reactive, not like simple resistors, and 2) varying, i.e. as the voice coil moves, the impedance changes. And when impedance changes, reactance i.e. phase also changes. And phase change translates directly into relative time delay changes.

(Rant): One of my beefs about the way most headphone-tech-related websites discuss this topic, is that they typically boil things down to one-dimensional terms. Like impedance, focusing on Ohmic values and not discussing the equally-important reactive i.e. phase angle aspect. Or damping factor, an abstraction which uses a single, non-complex number ratio to allegedly describe an amp's "ability to control the load". Why? Because it's easier conceptually. Easier to discuss. Easier to measure. You get the picture; it's "easier". But it is inaccurate and incomplete. (end Rant)

Personally, I think "controlling the load" is the wrong metaphor for this event. "Sensing and adapting" is more like it.

So, the question posed is, what is this one most-important thing that destroys an amp's ability to sense the load accurately? The answer is: Series Impedance. Any resistance or reactance that exists between the HPA and the headphone.

That is why we want headphone cables with low series resistance and loop inductance.
And HPA's with low output impedance.
AND, it is why we want an HPA that doesn't stupidly put a series resistor on the output, to "isolate it from the load". And many if not most of them do exactly that. "Isolating the amp from the load" = "not sensing it accurately", because series impedance linearizes any phase shift in the load, which IS most assuredly present and changing. And the lower the impedance the headphone is, the more difference this makes.

I can't tell you what HPA to go out and buy that does a good job of this, because I build my own amps. For headphones, I tend to prefer low-gain class A designs. They have an advantage of being able to be "optimized" for a given load impedance. But some Class AB amps, integrated opamps with current buffers, some chip amps, etc. can perform very well. Whatever it is, make sure it DOES NOT HAVE A RESISTOR ON THE OUTPUT!

There could be a clue in an HPA's specs that would indicate if it drives low-Z phones well. The distortion should be just as low into low-Z loads as it is into higher-Z loads. If the distortion increases into Low-Z loads, i.e. it is higher into, say, 32 Ohms than it is into 150 Ohms, then the amp does not drive low-Z phones well.

Next we'll get into the last of the mods. I have some more photos to take first.

 

[Precaud]

OK we can wrap this up tonight.

In the 1st post I made the point that, for dynamic headphone drivers (and speakers in general), the most problematic area is near the edge of the diaphragm, where it attaches to the baffle. The piston-like motion in the center becomes increasingly restricted, creating ragged frequency response in the range corresponding to it's dimensional wavelengths. In the K245, the edge effects start around 2kHz or so. The 2kHz to 5.5kHz region is where these HP's have issues. And it is primarily from the edge.

So why not try filtering them out? I experimented with blocking the large holes on the perimeter of the driver, the ones right above the driver's edge. First with thin felt and then with the same thicker stuff used in the cup. First a couple holes, then a few, then nearly all of them. Yes, it made a positive difference. But it wasn't sufficient. Then I remembered a "trick" used by a couple HP's, partly acoustic, partly psychoacoustic.

With two ears horizontally displaced, our hearing is quite horizontal in orientation. And our listen "habit" is to turn and position sound sources so they're directly in front of us. Much has been written about the tonal changes that happen as sound wraps around our foreface and enters the ear canal.

But the K245, and most headphones, present sound full-spectrum all around our ear. This tends to make "flat" HP's sound bright. And it delivers the full, extended high frequencies to our ear from behind, which is highly unnatural. We certainly are not accustomed to having objects that are in front of us do this. So, in more ways than one, it makes sense to limit or restrict the sound from the HP's that is behind the ear canal. The first time I saw this being done was on a pair of Stax SR-5's, and I remember liking it.

So I tried doing this with the K245, and voila, everything started falling into place. What surprised me is how much of the driver radiating area needed to be filtered to get it sounding neutral. The pic below is of the right earpiece. The area inside the green is what needs to be covered with the thick black felt.