AKG K245 Review

[isostasy]

If you want to stick to plug-in power with stereo mics, I've used both the Zoom H1n and the Rode AI-micro (with a splitter like this, as the latter can't power a stereo mic from one input only : https://www.thomann.de/fr/soundman_splitter_adapter.htm).

Ah yes this is probably the best option. I think @IAtaman wants to take measurements with REW on a computer, in which case you'd need one of the Zoom models with a USB interface like the H4N.

I've always found your posts extremely informative, especially your explanations of research. Do you have your own measurement equipment, and any interest in trying out the K245 headphones?

They have a bit of a give if you like, but I'd not call them loose. Will take a photo tomorrow.

I really enjoy them. They are the top pair in my rotation by "hours on head" metric since I got the new pads . I'll be leaving for vacation soon and these will be coming with me as well.



I have no idea. I am literate in electronics but not in acoustics. All measurement I see are stereo so I thought that is what's needed. Plus, maybe my left pinna is a bit bigger or a bit flatter or a bit different then the right one that might be significant in some way or another, so get a pair while you are at it I figured . You gents think a pair will not be necessary?

My plan is to measure HD600 on my head a few times and make that my reference to compare it to other headphones, starting with K245s with Brainwavz perforated pads. I have a few more headphones so I thought maybe I might even be able to build myself a larger set of reference measurements to at least have something to refer to when adjusting tuning.

Those Primo EM25s look good to me. I will order a few and play around with them. On the link you shared, it says Primo suggests the EM258 as a possible replacement for the Panasonic WM-61A. I found Panasomic WM-61A online as well. Shall I get those two, are they better in any way, or maybe more standard?

Dayton Audio also has this iMM-6 calibrated mics. They are like $20. Maybe I can get that and use it with the calibration file it comes with which might make life easier?

Ok I'll measure my ones (both physical dimensions and FR).

You only need a pair if you want to take consistent measurements of both ears. Otherwise you will be constantly swapping a single mic between your ears and you won't be able to replace it in the same position each time.

I say do what you find interesting, when the aim isn't to create an industry standard database but instead experiment and share I don't think it matters too much.

You will have to look at spec sheets to decide, though I honestly don't think it matters too much (Primo, WM-61A).

I don't know about the imm-6, besides, how would you get that in your ear?

 

[IAtaman]

I say do what you find interesting, when the aim isn't to create an industry standard database but instead experiment and share I don't think it matters too much.

No no, I have no intention of getting involved in the dicey business of measuring audio equipment, especially electromechanical ones. I don't even plan to share beyond getting suggestions and advice on how to measure better. I just want to play around with them to learn a bit about acoustics, and make sure my lovely K245s sound as good as they can

You will have to look at spec sheets to decide, though I honestly don't think it matters too much (Primo, WM-61A).

I don't know about the imm-6, besides, how would you get that in your ear?

I was thinking surgical disassembly

 

[MayaTlab]

All measurement I see are stereo so I thought that is what's needed. Plus, maybe my left pinna is a bit bigger or a bit flatter or a bit different then the right one that might be significant in some way or another, so get a pair while you are at it I figured . You gents think a pair will not be necessary?

I think that it's important to capture both left and right ears as the behaviour of the headphones can be different between them once they're on your ears indeed. As @isostasy mentioned, it's still possible with a mono microphone to measure both sides simply by swapping the mic, but I also find it more convenient to use stereo mics. On the other hand with a mono mic you won't have to compensate the left and right mics to measure similarly.

If you feel good with soldering electrets maybe start small with a cheap DIY mono mic and then move on to other setups later on if you want to ? Personally I just repurpose stereo binaural mics, but it's a waste of money if you can solder your own mics.

Ah yes this is probably the best option. I think @IAtaman wants to take measurements with REW on a computer, in which case you'd need one of the Zoom models with a USB interface like the H4N.

Both the Zoom H1n and AI-micro have a USB interface and can be used with REW . There could still be another cheap option available, but I am not aware of it.

Do you have your own measurement equipment, and any interest in trying out the K245 headphones?

Only a whole bunch of DIY in-ear mics and a few clone couplers. I have never worn the K245 but if the occasion presents itself why not ?

 

[isostasy]

No no, I have no intention of getting involved in the dicey business of measuring audio equipment, especially electromechanical ones. I don't even plan to share beyond getting suggestions and advice on how to measure better. I just want to play around with them to learn a bit about acoustics, and make sure my lovely K245s sound as good as they can


I was thinking surgical disassembly

I wouldn't get the imm-6 just to disassemble, rather I think the advice from MayaTlab is best.

btw I'm fairly sure my perforated pads aren't brainwavz, they are more like 110mm diameter and the lip is extremely thin so they basically don't fit at all. Measure terribly:

n.b. mic is new and I measured the HD6XX with new pads hence why slightly difference to previous measurements.

I think that it's important to capture both left and right ears as the behaviour of the headphones can be different between them once they're on your ears indeed. As @isostasy mentioned, it's still possible with a mono microphone to measure both sides simply by swapping the mic, but I also find it more convenient to use stereo mics. On the other hand with a mono mic you won't have to compensate the left and right mics to measure similarly.

If you feel good with soldering electrets maybe start small with a cheap DIY mono mic and then move on to other setups later on if you want to ? Personally I just repurpose stereo binaural mics, but it's a waste of money if you can solder your own mics.



Both the Zoom H1n and AI-micro have a USB interface and can be used with REW . There could still be another cheap option available, but I am not aware of it.



Only a whole bunch of DIY in-ear mics and a few clone couplers. I have never worn the K245 but if the occasion presents itself why not ?

Oh no that's tempting me to get a H1n now... an all in one field recorder and USB interface for under £100 ...

 

[dxmurky66]

I found a cheap listing for these as parts/not working so bought them as a base for installing Peerless 50mm drivers. But to my surprise (pleasant or disappointed, depending on your viewpoint), they work perfectly! What could I do but give them a listen and take some measurements?

AKG says 'The new K245 represents the culmination of decades of refinement—from the transducers, to the porting, to the fit and finish', that they 'offer the best-in-class frequency response', and that the 50mm drivers offer 'greater low-frequency extension and detailed bass reproduction'. This sounds like a good amount of effort going into either R&D or copywriting. RRP is $89 USD but they are on sale for less than that most places I looked.

(from headphonecheck)

I quite like the design, it's reminiscent of a few different AKG models put together. There is a folding mechanism which is unique in combination with the suspension strap.

Unfortunately this does not fit small heads like mine very well, though I have found this to be the case with other suspension strap designs. I put some foam under the strap and held it in place with one of those wool headband wraps and this solved the problem. Sticking a HD650 foam pad on with double sided tape would probably work well too.

The pads, conversely, do not feel very roomy. They're 53mm inside which is plenty front-to-back but I have to give them a wiggle each time I put them on to ensure the tops and bottoms of my ear are tucked in rather than pressed against my head by the pad. Even then my lobes feel a bit squished against the inside of the pad.


Measurements

My new approach to measurements is to take both in-ear and flat-plate coupler measurements. This is to avoid being mislead by some deficiencies of the FPC, and it also provides confidence when I can see agreement between each system. I make new reference measurements with my HD6XX every time I do in-ear measurements because it is almost impossible to place the mic in the same position each time.

Channel matching was excellent so following measurements just use R for easy readability.

View attachment 297028

This looks pretty good. Bass response is great! Now to check with an in-ear measurement:

View attachment 297034

This is promising as features seem broadly consistent across both measurement systems. The deep 8kHz dip on the HD6XX and tall ~8.5kHz peak on the K245 both correspond to a phase shift so I'm not sure about EQing that. The FPC showed the K245 was elevated in the whole region 6-11kHz above the HD6XX however, so will need listening to see if there are any problems here.


Listening

These are not bad at all. The 1-2kHz dip, combined with the already warm tilt, makes it sound a bit closed in and lacking clarity in the mids. This makes some songs a bit artificial sounding.

With the following EQ I think it sounds really good and fixes those issues completely:

PK Fc 1350 Hz Gain 3.70 dB Q 2.200
PK Fc 200.0 Hz Gain -2.00 dB Q 1.200

Just the first 1350Hz filter makes a big improvement, and the second is if you want to reduce the apparent warmth a bit more (this is the same filter I use for my HD6XX).

View attachment 297039


I tried multiple ways to see if I could bring up 3-5kHz but it always sounded worse.

I had a go at reducing the broad treble elevation as well but I don't think it's bothersome enough to require any EQ. I think a proper GRAS measurement would be the only way to confirm if it's 'real' or not anyway. Without EQ some cymbals appear slightly pronounced which is nice. With EQ a sort of shimmer is removed from some songs which may appear like some coloration is being removed. It's a wash really.

This is a starting point if you are bothered:

PK Fc 7400 Hz Gain -4.50 dB Q 5.000
PK Fc 8680 Hz Gain -2.00 dB Q 5.000


Conclusion

I really like these! They need at least the 1350Hz filter to sound their best, then it's just a matter of adjusting bass and treble to taste. I can't think what would be better for the price. I would really like to see someone with an industry standard rig measure them still.

My main reservation is availability of pads. AKG don't appear to offer replacements at all which is disappointing. There are options from Geekria in pleather and velour but I'm wary of the changes to FR these might bring.

Thanks

Really like your review and enthusiasm, along with the subsequent contributions and links.
Makes me feel better about my own K245s, which I got 2-3 years ago on discount - £159 down to £110 (now only £45 on Amazon! ).
I think these were just before Samsung took them over, iirc.
My fav headphones, even with my big ears and 57yr old lugs!

 

[IAtaman]

I have been searching for Primo mics for a while now and I can't seem to able to source them where I live unfortunately.

I found these online CUI Devices mics online, and ordered a bunch. What do you think of these mics?

@isostasy @MayaTlab

Full datasheet is here:

https://www.cuidevices.com/product/resource/cma-4544pf-w.pdf

 

[AnalogSteph]

That's pretty meh sensitivity and noise for a 10 mm capsule. Only a 2-terminal capsule either, so expect mediocre distortion without a Linkwitz-style mod.

 

[IAtaman]

That's pretty meh sensitivity and noise for a 10 mm capsule. Only a 2-terminal capsule either, so expect mediocre distortion without a Linkwitz-style mod.

Thanks for the comment. This is the mod you are referring to I reckon?

Measurement circuit in the datasheet has an amp symbol instead of a FET - is that significant?

 

[IAtaman]

So I finally built myself a measurement device with the mics I mentioned above and a Røde A1 Micro. I soldered the mics to a pair of 3.5mm to 3.5mm cables I bought online and cut at one end; attached them to a pair of tips from an IEM and plugged them into my ears. And voila - I have a blocked ear canal measurement device now I guess?

Anyway, as every sensible person who builds a headphone measurement device should do, I measured by HD600s first. I did a bunch of measurements (all mono at this stage) and they all landed on top of each other so I am not sure how precise this is but at least it is consistent (enough?)

Then I measured the K245 with Brainwavz 100mm earpads I linked above. It had a big hump at 3K and the upper mids were all wonkey, so I put in a filters by hand to bring the FQ as close to that of HD600 as possible and this is what I got now.

I have no idea what I am doing so a few pointers to make sure I have something useful on my hands for the purpose of EQing my own headphones would be much appreciated.

1/6 octave smoothing.

 

[MayaTlab]

I have no idea what I am doing so a few pointers to make sure I have something useful on my hands for the purpose of EQing my own headphones would be much appreciated.

Let's face it me neither . I would continue to experiment with various mic insertion depth and in-situ mic designs, and see whether or not the results you obtain are repeatable. I've applied a reversed "variables control" logic to my in situ measurements, whereby I try to see whether or not the results are consistent when as many variables as possible are altered (which then makes it plausible - although not certain - that these variables are not causing any effect and increases my confidence in the results I obtain) - as controlling for each variable but one is practically impossible with in situ measurements.

I measured by HD600s first. It was find. I did a bunch of measurements (all mono at this stage) and they all landed on top of each other so I am not sure how precise this is but at least it is consistent (enough?)

This is just my personal experience as well, but I found the HD600 to be quite repeatable indeed, on my head, from seating to seating.

 

[IAtaman]

Let's face it me neither . I would continue to experiment with various mic insertion depth and in-situ mic designs, and see whether or not the results you obtain are repeatable. I've applied a reversed "variables control" logic to my in situ measurements, whereby I try to see whether or not the results are consistent when as many variables as possible are altered (which then makes it plausible - although not certain - that these variables are not causing any effect and increases my confidence in the results I obtain) - as controlling for each variable but one is practically impossible with in situ measurements.

I have feeling we are at very different levels of not knowing what we are doing

I am not sure what are the variables I need to control for. I can think of the following:

• Insertion depth - I am hoping this will be less of an issue as the mic is fixed to the ear tip so physically it can only be more or less in the same depth
• Different mics / mic unit variation
• The angle of the mic in my ear? Mic is fixed to an ear tip so it does not move in or out much, but I guess it can roll and yaw a bit inside my ear - would that make a difference?
• Different headphones - some headphones touch my ear and physically change the shape of my pinna so I suppose that would make a difference? As you alluded, different seating might make a difference and different headphones would have different sensitivities to seating / leakage as well. I
• Headphone unit variation - this I don't care about as long as I can reference all measurements to my own HD600, correct?

What am I am missing?

I came back and did another measurement, both for HD600 and K245, and results were very close the the original measurements for both headphones. That is a good sign for consistency I suppose - or it is too early to tell?

And finally, I am struggling with level matching a bit - what would be the proper way to do that, do an SPL measurement at a normalization frequency and match SPL at that F before running the sweep - would that be right?

 

[MayaTlab]

What am I am missing?

Have you read a few articles on the subject of in situ measurements ? May I suggest trying to read as many articles as possible from Hammershøi, Møller et al. ? Quite a few of them are freely available on the web.

For now I'd stick to centred (not off centre) blocked ear canal entrance electrets, making sure that they are at least flush with the ear canal's entrance if not a little bit deeper, and that they can be positioned repeatedly in the exact same position each and every time you put them back in. Some people have 3D printed housings for the capsule to that effect instead of relying only on IEM eartips - but I have no experience in that area.
Perhaps experiment later with open ear canal entrance mics and compare open vs closed results.

 

[IAtaman]

May I suggest trying to read as many articles as possible from Hammershøi, Møller et al. ? Quite a few of them are freely available on the web.

Alright, good suggetion, thank you.

These are the ones I was able to find, for those who might be interested in them as well

Head-related transfer functions of human subjects

https://vbn.aau.dk/ws/files/227875164/1995_M_ller_et_al_AES_Journal_c.pdf

Transfer characteristics of headphones measured on human ears

https://vbn.aau.dk/ws/files/227875204/1995_M_ller_et_al_AES_Journal_b.pdf

(I tried changing letters from a to f in the link, but ended up with the same articles. Not sure what is Aalborg Uni's naming convention for publications)

For now I'd stick to centred (not off centre) blocked ear canal entrance electrets, making sure that they are at least flush with the ear canal's entrance if not a little bit deeper, and that they can be positioned repeatedly in the exact same position each and every time you put them back in. Some people have 3D printed housings for the capsule to that effect instead of relying only on IEM eartips - but I have no experience in that area.

I did the following:

• Remove most of the nozzle from inside the IEM ear tips to create space for the mic, but leave some behind for rigidity
• Cover the back of the mic with enough blue tag (which came in white color actually) so that it fills the ear tip fully, keeps the mic centered in place and inside my ear safely
• Cut the bottom part of the ear tip to allow for the cables to pass without creating a leak

I did multiple measurement with multiple headphones. Up to 3K Hz, consistency is quite good. Above that, there is a bit of variation indeed. Will read the articles and try to figure out what is going on. Thanks a lot for the tips and the advise on the direction.

(I tried putting my hand 10cm away from the grill for a few headphones to see if reflections from my hand would make any difference. It didn't - always wanted to try that

 

[IAtaman]

I did some measurements using my new measurement mics. With more comfortable brainwavz pads, EQ settings below matches the K245 to HD600 tonality more or less, based on my totally inaccurate and non-scientific on head measurements.

I have been using these EQ settings for a month now almost and am very happy with it, in case someone is interested.

Preamp: -5dB
Filter 1: ON LSC Fc 70 Hz Gain 3 dB Q 1.41
Filter 2: ON PK Fc 200 Hz Gain 2 dB Q 0.9
Filter 3: ON PK Fc 1000 Hz Gain 4 dB Q 0.4
Filter 4: ON PK Fc 3000 Hz Gain -8 dB Q 1.41

 

[isostasy]

Yep, elevated and sloping bass, the familiar 2kHz dip (looks at Susvara) and fierce treble peak at 7kHz (sharpness)

This was not a studio headphone though but targeted at music consumers.

Hi, replying here as it's not relevant to the Susvara where I saw your post.

I wouldn't call it a "fierce treble peak". Remember this is only comparative to the HD6XX measurement, the microphone is not calibrated and there is no target. Therefore I would bear in mind the shortfall around 8kHz that is often measured in the HD6XX, especially compared to Harman on a GRAS coupler.

I think in practice the HD6XX does sound a bit deficient here so I EQ it up, almost exactly the same as the filter Amir uses in his HD650 review in fact. The K245 in comparison sounds a little bright as I mentioned in my review but not excessively so. The fact that removing the apparent peak makes the sounds muffled and less detailed suggests to me it might be better left alone and may be more true to neutral (or Harman, at least) here. In any case, not a "fierce treble peak".

Additionally, there is no dip at 2kHz. It is lower (I used a filter at 1350Hz to lift it) and not that severe. In fact the dip looks similar to that of the S5X so you may like the K245 too.

Finally, the bass doesn't sound elevated. There may be a bit too much around 200Hz but this is the same issue people find in the HD6XX. Doesn't bother me most of the time. Below this I would call it correct rather than elevated. Probably close to Harman.

I would like to see a review of the K245 from you, @solderdude . It is very affordable, sometimes incredibly cheap, especially second hand/open box. Not sure it's targeted at music consumers, the copy doesn't read that way:

You can take headphones out of the studio…​

…but you can’t take the studio out of these headphones. Now it’s even more convenient to take studio-quality sound with you. With the K245, it’s almost as if you were sitting in front of a pair of great reference monitors—transparent, defined, and uncolored. All three headphone models in the Foldable Studio Headphones Line (K175, K245, K275) offer the best-in-class frequency response, delivering unrivaled clarity and detail, making it easy to hear subtle detail and to create mixes that translate well across different playback systems. Their superior sensitivity and low impedance offer maximum output before distortion, so you can use them even when playing music on lower-powered mobile devices.

 

[solderdude]

K245 is no longer available and have not been able to find one in the usual 2nd hand market.
Maybe someone sends one in or puts one up for sale....

 

[IAtaman]

K245 is no longer available and have not been able to find one in the usual 2nd hand market.
Maybe someone sends one in or puts one up for sale....

Amazon.co.uk has one in stock. Used like new.

 

[solderdude]

That would become a € 100.- headphone once taxes etc have been added.

 

[IAtaman]

That would become a € 100.- headphone once taxes etc have been added.

Damn you Brexit! Italia has it in stock as well at EUR84.

PS. Add it to cart and see what is the total. When I added, it became GBP42.57 + shipping. If you are not gonna get the one in the UK, let me know, I will

 

[solderdude]

It ships from the UK seller though so think it is the same one.