[logscool]

 

My graph doesn't start at 0 (my mistake) - so it is only a 7 db fall and not 20 db.

Oh of course in that case it is quite a different story. Still I can not help but notice that the relative peaks and valleys are the same between your graph and Tyll's it is simply their differences in magnitude which seems to have changed by a constant factor at all frequencies. 
 
Could this potentially be due to the test being conducted at a different overall dB level which might affect the amount of change?

 

[hans030390]

I think lokesen's measurements match up with Tyll's, more or less (with tight seal, not loose). Taking into consideration that lokesen's mics don't work well below 100Hz, you'll see the same dip that reaches a low point at 200Hz. From there, you see the hump that peaks around 1Khz and reaches another low point around 2-3KHz. You then have the spike in treble around 7-10KHz. Tyll's graphs seem to be set up in such a way that changes in the sound look more pronounced than what lokesen used to measure them. I would also assume that Tyll has a much more sophisticated setup, leading to a more accurate reading in relation to the human ear. Still, the measurements do look very similar from 200Hz and up.
 
I'd be surprised if the bass port is what is causing this dip at 200Hz. I'm not saying it couldn't be, or that it doesn't play some factor in it. I suppose given the design of the headphone, the placement and size of the bass port in combination with that could cause it. I'm no expert in this by any means, but I'd guess the bass port isn't the main issue.
 
Lokesen, have you tried incorporating some of the mods people do to the stock T50RP into your KODA headphones? I'm wondering how they would sound if you:
 
1. Removed all the current damping material in the cups and...
2. Lined the cups with Paxmate or Silverstone acoustic foam (on both the interior of the cups and the back side of the driver baffle)
3. Stuffed the cups with some teased cotton balls. I do 6 per cup with the stock T50RP, and it does take a very slight bit of effort to get the cups back on.
4. Experimented with different materials on the back of the driver. As I've mentioned, this is the material I used for my mod. I did it the same way you did with one square hole in the middle. The material is 75% wool and 25% fiber. Sound absorption and vibration deadening is one of the usages it mentions specifically, so I'm assuming it's a good felt for this situation. Sounds great to me, at least! 
5. Lined the entire front side/ear side of the driver baffle in the acoustic foam mentioned above
6. Tuned the bass port in relation to this
7. Made sure that everything is as air-tight as can be, except the bass port (obviously).
 
From how you've described it, the cup and baffle material you're using sounds like it doesn't need additional dampening with something like Dynamat or plasticine. It would be extremely unfortunate if the cup design itself wasn't suited for the driver, but I highly doubt that is the case! The suggestions I listed above are more or less what I've done with my stock T50RPs.
 
You may also want to look into other forms of damping and acoustic diffusion behind the driver and in the cups. Like I've said, I am no expert in this area...but the 200Hz dip might be due to a standing wave, sound reflections, etc. The suggestions above should help eliminate those type of issues. I could be wrong though...I'm only going off the tiny bit of knowledge I have on acoustics.

 

[logscool]

 

I think lokesen's measurements match up with Tyll's, more or less (with tight seal, not loose). Taking into consideration that lokesen's mics don't work well below 100Hz, you'll see the same dip that reaches a low point at 200Hz. From there, you see the hump that peaks around 1Khz and reaches another low point around 2-3KHz. You then have the spike in treble around 7-10KHz. Tyll's graphs seem to be set up in such a way that changes in the sound look more pronounced than what lokesen used to measure them. I would also assume that Tyll has a much more sophisticated setup, leading to a more accurate reading in relation to the human ear. Still, the measurements do look very similar from 200Hz and up.

 
^This^

 

[lokesen]

Just got around to another test. This time I found out how to eliminate audio feedback, so I could try at proper volumes - I know, I should have known this conducting the first test - but now i know This made it possible to get the lower frequencies too.
 
Red: Right cup
Blue: Left cup
Green: Loose seal
 

 
This time i tried about 10 times on each chanel and this is just about avarage result. The loose fit take a deeper or less deep plunge depending om how loose the fit is.

 

[liamstrain]

It would be unusual of Tyl to provide a chart where the seal wasn't good on the tested cans, but not impossible. Might be worth asking if he has the time to double check. 
 
That said - Tyls test rig, using a standardized test head, with ears - is pretty well calibrated for it's purpose - and while it may have some short comings in ultra fine resolution (sub .3db changes), is quite good, and he is an experienced reviewer and tester - I'd be more inclined to say his readings are the good ones, than Lokesen's - no offense.

 

[lokesen]

Quote:

It would be unusual of Tyl to provide a chart where the seal wasn't good on the tested cans, but not impossible. Might be worth asking if he has the time to double check. 
 
That said - Tyls test rig, using a standardized test head, with ears - is pretty well calibrated for it's purpose - and while it may have some short comings in ultra fine resolution (sub .3db changes), is quite good, and he is an experienced reviewer and tester - I'd be more inclined to say his readings are the good ones, than Lokesen's - no offense.

I agree - but it's the green curve that make me wonder - it's just so similar to his readings, and that is very very stange.
 

 
 

 

[hans030390]

That is strange that the loose seal exhibits the same 200Hz dip that Tyll measured, but the tight seal does not. 

 

[lokesen]

Quote:

That is strange that the loose seal exhibits the same 200Hz dip that Tyll measured, but the tight seal does not. 

 
 
   
 
 
 
Red: Right cup     Blue:[size=9.8pt] [/size]Left cup    Green: Bad/loose

And even more so, that my reading with proper seal shows pretty good readings - it's unbelievable frustrating. I will not modify my design when they sound just right to my ears and to my frequency response test. It feels a bit like accepting something is blue when you see it as red. 
 
Anyway - until somebody can make an alternative frequency response test, that will either confirm Tyll readings or my own, I will not sell or make any more KODA^2 headphones. My goal was never to sell many of these, but to make a beautiful and very good sounding headphone - because I enjoy making them.  
 
 

 

[Tyll Hertsens]

Interesting results.  It seemed like I was getting a good seal ... the isolation graph looks like it's isolating pretty well.  Also odd that I got a peak in the THD+noise at the same place.
 
But ... there's only one way to know for sure, and that's to measure them again, which I'd be happy to do.
 
Lemme know.

 

[MohawkUS]

So.. how do they sound? As useful as frequency graphs are they can't match a real review. Frequency charts don't show things like 'forward', 'laid-back' '3-D sounding' or any of that. The charts would leave me to believe that I would like the Senn 598, but I found it horribly boring.

 

[lokesen]

Quote:

Interesting results.  It seemed like I was getting a good seal ... the isolation graph looks like it's isolating pretty well.  Also odd that I got a peak in the THD+noise at the same place.
 
But ... there's only one way to know for sure, and that's to measure them again, which I'd be happy to do.
 
Lemme know.

That sounds great Tyll - It would be fantastic to get some confirmation before changing too much on my design. When I get my test rig up and running, we can compare the measurements, and see if they share similarities.
 
I'll have them forwarded to you when possible.
 
Thanks

 

[WhiteCrow]

Hmmm so very little bass slam, otherwise it looks good. Would be a warm sounding headphone but I would close the bass port a bit.

 

[lokesen]

Quote:

Hmmm so very little bass slam, otherwise it looks good. Would be a warm sounding headphone but I would close the bass port a bit.

There is actually much bass-slam in these headphones. I will get back with new readings when I get some proper microphones that picks up the lower frequencies properly. The microphone I use in my temporary test rig can't really detect fequencies below 75 hz properly. But the headphones has much bass slam for sure. 
 
I think people that tried them will confirm this. 
 
Tyll will also do some new measurements to either confirm his own readings - or my readings. We have to wait and see either way

 

[cn11]

Even my pair of 'almost Rastapants 2' modified T50RP's have decent bass slam... I can't wait to hear the KODA^2's.. 
 
If you look at Tyll's freq. response measurement, even that is only around -3 to -4 at 20hz. Not too bad. And many times freq. response graphs don't tell the whole sound signature story, as we all know.

 

[Tyll Hertsens]

Quote:

So.. how do they sound?

I thought I clearly heard the loss of low mids in listening.