[Kelowna]

This is a good thing and much better than what I’m doing now.

It also should be noted that using passive components to create the BSC like the stock interface does also results in an uneven phase response. So the stock interface box does not have a flat phase response even when no EQ is applied. So with my interface combined with the FIR BSC, folks will finally be able to hear how these ribbons sound with a perfectly flat phase response. Something you certainly want from single driver systems. A flat phase response is one of the big reasons folks like full range driver speakers.

 

[tunes]

I wonder if the Schiit LOKI HIGH QUALITY TONE CONTROL will enable bass improvement when running the SR1a off the JOT-R. Will it work if routed between the output of my DAVE/HMS and the amp? Can I use the volume controller on the DAVE to control head speaker volume instead of the JOT-R? If the LOKI is in the chain, with all EQ flat, will it introduce any noise or coloration to the signal?

 

[Kelowna]

I just discovered that HEDD already thought about this:

https://www.hedd.audio/en/hedd-lineariser/

Only problem is VST plugins are mediocre and have their own set of problems. This is why Roon doesn't accept VST plugins.

 

[Pale Rider]

I just discovered that HEDD already thought about this:

https://www.hedd.audio/en/hedd-lineariser/

Only problem is VST plugins are mediocre and have their own set of problems. This is why Roon doesn't accept VST plugins.

Yeah, I learned to dislike and stay away from VST plugins years ago.

 

[Kelowna]

Yeah, I learned to dislike and stay away from VST plugins years ago.

At least it shows folks that phase response is a real problem with headphones. The great part about using Audiolense is it does it right. And uses 64 bit floating point precision. But it's extremely important to use a very good mic, along with mic pre and ADC when taking the measurement sweeps. Because the quality of the filters is limited by the quality of the mic, mic pre, and ADC. I'm using the Merging Anubis mic pre/ADC:

https://www.merging.com/products/interfaces/merging+anubis

Speaking with the best studio mastering engineers out there that own dozens of ADC's, the ADC in the Anubis is the best there is at any cost! Quite a feat in such a small package and the cost! For a mic I'm using an Earthworks M23:

https://earthworksaudio.com/products/microphones/measurement-series/m23/

From what I hear just as good as their M30 for this application.

 

[FLTWS]

@Ciggavelli There's a pdf on this thread pages back.

 

[ctemkin]

Do people with the SR1a who have heard the TOTL STAX with best dedicated amps think that the bass quantity and quality are comparable?

I have Stax 009 headphones and a BHSE with vintage tubes that, in my impression, produce a more satisfying bass than the stock tubes. I have tried comparing this combination against an SR1a out of a Jot R.

A few up-front disclaimers. I can match volume only approximately, because I have only an inexpensive SPL meter and because, with the SR1a, it is hard to know where to position the meter, since the "wings" are a small distance from your ears. Also, the bass frequencies of an SR1a are most affected by positioning of the wings; the bass increasing as the wings are located closer to the ears. Finally, to make any sense of the differences, I have to try to listen to passages of maybe 10-15 seconds, or I just start getting lost in the music, and it is never long before I just wear out.

That said, I find that the SR1a produces the same or somewhat more bass than the Stax gear. I tend to use a Loki with the Stax much more often than with the SR1a. I also find that the SR1a has somewhat better resolution at all frequencies than the Stax gear. Finally, I prefer the effect of the sound coming from a source that is a slight distance from my ears rather than right on top of them, and it is also a lot more comfortable physically, especially because the room I listen in tends to be warmer than the rest of the house. On this last point, the Jot R is barely warm to the touch, while both the BHSE's power supply and the BHSE itself put out a lot of heat.

 

[Ciggavelli]

So, I have a question about fitting. The metal part highlighted below, is it supposed to be loose like that and bend a bit?

 

[Aleksandar R.]

It also should be noted that using passive components to create the BSC like the stock interface does also results in an uneven phase response. So the stock interface box does not have a flat phase response even when no EQ is applied. So with my interface combined with the FIR BSC, folks will finally be able to hear how these ribbons sound with a perfectly flat phase response. Something you certainly want from single driver systems. A flat phase response is one of the big reasons folks like full range driver speakers.

Kelowna, it is naive to think that I would let the very basic things like phase response to be in error.
"Uneven" phase response is what is needed for this.
Let's not forget that the reason for using EQ is physical and natural, that transducers are minimum-phase-systems, thus, whatever happens that will skew the frequency response, will also skew phase along with it.
What needs to be additionally compensated for is the dipole rolloff that is not completely compensated by proximity effect. Basically, phase is also skewed and in order to recreate the flat response and flat phase, the filter must turn the phase response the other way.
So, if you use FIR with flat phase, you would actually create the unnatural response, being flat, but with skewed phase (as you haven't compensated the original skew)
You don't need to use FIR for this, but if you prefer it, have it synthesised to recreate the phase response of the stock interface box.

 

[Aleksandar R.]

So, I have a question about fitting. The metal part highlighted below, is it supposed to be loose like that and bend a bit?

Yes. That reduces the pressure compared to the pressure of upper pads at your temples.
Bend it outwards to relieve the pressure to soft tissue in front of your ears, if you will. Do the opposite to increase it.
Twist it so the angle of the headphone follows the shape of your head, bottom going inwards or outwards, as desired.
Be bold, nothing will break.

 

[Kelowna]

Kelowna, it is naive to think that I would let the very basic things like phase response to be in error.
"Uneven" phase response is what is needed for this.
Let's not forget that the reason for using EQ is physical and natural, that transducers are minimum-phase-systems, thus, whatever happens that will skew the frequency response, will also skew phase along with it.
What needs to be additionally compensated for is the dipole rolloff that is not completely compensated by proximity effect. Basically, phase is also skewed and in order to recreate the flat response and flat phase, the filter must turn the phase response the other way.
So, if you use FIR with flat phase, you would actually create the unnatural response, being flat, but with skewed phase (as you haven't compensated the original skew)
You don't need to use FIR for this, but if you prefer it, have it synthesised to recreate the phase response of the stock interface box.

The beauty of these FIR filters is they'll be free, and have the ability to be enabled and disabled with a click of a button in Roon. So folks can form their own opinions on whether or not they're the superior way to do the BSC with these headphones.

The phase is fairly flat already.

But you're also forgetting that a big part of the improvement will be removing the physical passive components from the signal path. If one believes that IIR filters such as the passive filter in the stock interface box is the superior way to do things, they can always just use a PEQ to create an IIR based BSC instead. All these are experiments that are free to do, and take minutes to compare.

When I say "BSC" I'm referring to "Baffle step compensation". the same thing you're referring to when you say "Dipole rolloff". Just different terminology. Perhaps Dipole rolloff" is a more accurate term to describe the effect. However correcting it is the same procedure as compensating for baffle step losses in loudspeakers.

 

[Ciggavelli]

Yes. That reduces the pressure compared to the pressure of upper pads at your temples.
Bend it outwards to relieve the pressure to soft tissue in front of your ears, if you will. Do the opposite to increase it.
Twist it so the angle of the headphone follows the shape of your head, bottom going inwards or outwards, as desired.
Be bold, nothing will break.

Thanks so much. I had been feeling pressure in my temples, and bending that piece really helped. I love these earfield monitors!

 

[Kelowna]

I thought of something I could probably do once I have one built, and all 6 of my friends agree that it sounds better than the stock interface. Send one on a U.S tour between all RAAL SR!-a owners with high end speaker amp setups. Then they could all share their experience. And based on this large pool of subjective data, and measured objective results I also have the means to provide, others can decide whether or not it's worth the $700 in parts to build their own.

 

[Aleksandar R.]

Baffle step is a completely different thing. It's about increased response intensity because the radiation turns from 4Pi sterradians to 2pi sterradians, from sphere to half-sphere, at frequency that depends on the size of the closed speaker front baffle. Basically, reducing the radiation angle does not involve phase change.
Dipole roll-off does, as you have the decrease in intensity due to opposite phase waves cancellation...
Basically, it's wrong to correct for that with phase linear filtering.

I'm not really forgetting the presence of components in the signal chain, you are welcome to kick them out, but not for the reason you mentioned, the phase skew, just to be very clear on that.

These two things that you are lumping together, BSC and open baffle rolloff, are very different in origin and DO NOT require the same compensation method.

Take my advice, or not, but do not invent wrongs on my part to sell-up your misconceptions. Just to be clear on that, too, as I won't sit quietly and you let market your ideas on those basis.

 

[Kelowna]

Baffle step is a completely different thing. It's about increased response intensity because the radiation turns from 4Pi sterradians to 2pi sterradians, from sphere to half-sphere, at frequency that depends on the size of the closed speaker front baffle. Basically, reducing the radiation angle does not involve phase change.
Dipole roll-off does, as you have the decrease in intensity due to opposite phase waves cancellation...
Basically, it's wrong to correct for that with phase linear filtering.

I'm not really forgetting the presence of components in the signal chain, you are welcome to kick them out, but not for the reason you mentioned, the phase skew, just to be very clear on that.

These two things that you are lumping together, BSC and open baffle rolloff, are very different in origin and DO NOT require the same compensation method.

Take my advice, or not, but do not invent wrongs on my part to sell-up your misconceptions. Just to be clear on that, too, as I won't sit quietly and you let market your ideas on those basis.

I have no doubts you understand all the fundamentals of the design, and terminologies to best describe the phenomenon that needs compensated for. So I won't argue with you on this. However I'm going to attempt to correct for whatever you want to call it that will result in a flat phase and frequency response at the ear position.

And there's no point in debating any claims. Because if this project becomes a flop, it will reflect in the end results. And the results will say all we need to know.

And I never once said you did anything wrong. That would be like telling a loudspeaker manufacturer they are doing it wrong because they're building passive speakers. No they aren't doing it wrong, they're just not utilizing the best of all available technologies to get the job done.