[dhruvmeena96]

Hey everyone, about wires, everyone is using estron litz wire? like the one Soundlinks sells? I'm finding this wire too weak, they tend to break when I'm prototyping and moving parts a lot.. like assembling and disassembling the same driver/tubes multiple times. they don't seem to have a good resistance for that. If I'm mounting just one time and don't touch it anymore they work ok.

Dind't find any thicker litz wire, Any suggestions?

Take 2 wires, twist them and solder the ends...here you have a strong cable...

they are weak, but okay. Best for prototyping.

 

[DannyBouwhuis]

Hey everyone, about wires, everyone is using estron litz wire? like the one Soundlinks sells? I'm finding this wire too weak, they tend to break when I'm prototyping and moving parts a lot.. like assembling and disassembling the same driver/tubes multiple times. they don't seem to have a good resistance for that. If I'm mounting just one time and don't touch it anymore they work ok.

Dind't find any thicker litz wire, Any suggestions?

The solder pads are so weak, I've damaged so many drivers this way, I bought some thicker litz from tao a while ago, I'd guess like double the thickness of the soundlink wire and even that managed to pull of a solderpad from a swfk... Thinner wire is also a lot easier when there are lots of components in your iem, easier to move around, most importantly what forcem said: heat

 

[DannyBouwhuis]

Question, a few pages back dhruv concluded that clarity is the field of sound perfected in phase. And I read somewhere that soundstage is caused by the difference in phase between lower and higher frequencies for example, delay. So its either stage or clarity ?? Is this correct and is there a compromise ?

 

[Lucas Taboada]

Personally I would suggest you to keep using those estron wires especially when prototyping. yes they have a tendency fray and they are "weak".
BUT that weakness is an advantage. they will not rip the solder pads apart from your drivers hanging from them, usually the wire will break before detaching/damaging the pads.
thin wires means short soldering times thus less chances of damaging the driver and pads. just a simple tap is good to get them attached.

they are easier to solder to drivers with small pads like twfk/swfk and co

These are my 2 cents but they come with ton of experience and some damaged drivers along the way.

especially for prototyping I would use them.

Got it!! thinking this way it really make sense. I will continue to work with this wire then, thank you and everyone for answering!

 

[Furco]

My lord! This forum has accelerated out of the “middle” school level and straight into 4 year university knowledge and beyond. Very impressed with the last 250 pages of contributions.

Now for a dumb question: Anyone have opinions about the Knowles GV-32830 (quad)? I’ve got shells for my daughter sitting on my bench begging to get completed.

 

[eunice]

The Knowles GV is very good, definitively worth a build. It’s hard to beat.

 

[eunice]

Question, a few pages back dhruv concluded that clarity is the field of sound perfected in phase.

Clarity in the sense I posted it is just a fancy way of measuring impulse response (e.g. amplitude in first 80ms of a Signal vs amplitude Overall)

Imagine you had a room with a lot of reverb, your brain would have a hard time to isolate the signal from the reverb and you would loose information about where the sound is coming from. You would describe that as less clear.

But how do you measure ‚amount of reverb‘? The idea is that reverb builds up over time. You would first hear the original signal and then the reflections from the wall, adding volume to the original signal. If there were no reverb in the room, the signal would stay at the same volume. That’s why the clarity measurement is the relation of the first 80ms of signal (which should by dry) to the rest of the signal.

Phase coherence is also necessary for clarity as it helps you pinpoint the source of the signal and therefore separate the signal you focus on from the rest. Again in a room with a lot of reverb you will have to concentrate harder to isolate the signal you want to hear.

Of course an IEM would not introduce a lot of reverb, but it not react quickly to a new signal, causing the same issue (less sharp / less clear sounds)

And I read somewhere that soundstage is caused by the difference in phase between lower and higher frequencies for example, delay. So its either stage or clarity ?? Is this correct and is there a compromise ?

Soundstage is caused by the phase difference and timing. Exactly that is why IEM need to be able to accurately reproduce phase and impulses. So it’s not an either/or scenario, better impulse response and phase stability (=clarity) directly improves soundstage.

If you look at my Bellsing 6 measurements above SPL and distortion is almost the same with or without zobel. Yet the soundstage is absolutely improved by the Zobel. The only measurement that seems to show that is clarity.

The clarity graph by RoomEqWizard does not care about phase though, it is just a fancy way to display impulse response.

 

[dhruvmeena96]

Question, a few pages back dhruv concluded that clarity is the field of sound perfected in phase. And I read somewhere that soundstage is caused by the difference in phase between lower and higher frequencies for example, delay. So its either stage or clarity ?? Is this correct and is there a compromise ?

Soundstage information is already there in the music itself. No need to make it worse with phases of drivers etc.
Imaging increases the soundstage construction. Soundstage is delay of phase in time space domain, whereas clarity is correction of electrical phase, so that driver can work in equilibrium. Second, all info is recorded in music and anything earphone adds(except for FR) just destroy the naturalness of staging.

My lord! This forum has accelerated out of the “middle” school level and straight into 4 year university knowledge and beyond. Very impressed with the last 250 pages of contributions.

Now for a dumb question: Anyone have opinions about the Knowles GV-32830 (quad)? I’ve got shells for my daughter sitting on my bench begging to get completed.

Working on GV recently.
Finding a perfect point zobel to get the peaky TWFK dual impedance to normal.


Update(this is edited)
33uF capacitor and 35 ohm resistor for the zobel(still under test)

Yellow damper 8mm away from spout on HODVTEC
Green damper 4mm away from spout on TWFK

TWFK tubes are 10mm length 2mm ID
HODVTEC tube is 10mm length 2mm ID(can be experimented with lower ID, I will test)


Just make sure to open the HODVTEC vents open, so the driver internal pressurization can be released properly and bass can boost.(important)

Clarity in the sense I posted it is just a fancy way of measuring impulse response (e.g. amplitude in first 80ms of a Signal vs amplitude Overall)

Imagine you had a room with a lot of reverb, your brain would have a hard time to isolate the signal from the reverb and you would loose information about where the sound is coming from. You would describe that as less clear.

But how do you measure ‚amount of reverb‘? The idea is that reverb builds up over time. You would first hear the original signal and then the reflections from the wall, adding volume to the original signal. If there were no reverb in the room, the signal would stay at the same volume. That’s why the clarity measurement is the relation of the first 80ms of signal (which should by dry) to the rest of the signal.

Phase coherence is also necessary for clarity as it helps you pinpoint the source of the signal and therefore separate the signal you focus on from the rest. Again in a room with a lot of reverb you will have to concentrate harder to isolate the signal you want to hear.

Of course an IEM would not introduce a lot of reverb, but it not react quickly to a new signal, causing the same issue (less sharp / less clear sounds)



Soundstage is caused by the phase difference and timing. Exactly that is why IEM need to be able to accurately reproduce phase and impulses. So it’s not an either/or scenario, better impulse response and phase stability (=clarity) directly improves soundstage.

If you look at my Bellsing 6 measurements above SPL and distortion is almost the same with or without zobel. Yet the soundstage is absolutely improved by the Zobel. The only measurement that seems to show that is clarity.

The clarity graph by RoomEqWizard does not care about phase though, it is just a fancy way to display impulse response.

@eunice
Well explained bro...


@DannyBouwhuis
Since IEM is near the eardrum, reverb are not a chance, and I didn't say about it. Well, what does electrical phase really means.

Directional queues come from the first arrival response. We judge arrival direction in well under a millisecond. However, judging what we are hearing takes longer. To determine spectral balance, the ear/brain combination analyzes the incoming sound typically over a 5 to 30ms interval. This interval is called the Haas fusion zone. Within this interval we are not aware of reflected sounds as separate spatial events. All of the sound appears to come from the direction of the first arrival. Lateral reflections from adjacent walls help extend the soundstage beyond the physical span of the loudspeakers. The comb filtering action of the many early reflections arriving at the listening position with varying phases adds a sense of spaciousness to the sound.

So anything after 30ms will sound like an echo or reflection. Since room can do it way better, veterans prefer loudspeaker in a well damped room. IEM dont have it and most of the signal ends up within Haas fusion zone instead of getting separated. And since most musician still use studio speakers instead of iem, this problem creates mashes of signal data which when fused with irregular electrical phases of BA, becomes technically useless info, which our brain is trying to decode. Like a bad audio file or corrupted video somewhat running. Phase, is angle of signal in graph at respect to zero. Dips mean, its starting late and peaks mean it is starting early from standard zero. It is because of some frequency impedance peaks and dip. If a frequency needs more voltage compared to other, it will take time, before it play and vice versa(in layman), so anywhere you see a high impedance peak, it is always followed by a phase dip

There are only two uses of the zobel filter

1. Linear phase
2. Linear impedance(to prevent signature change due to swinging)

Both helps our brain as somehow the phase is correct, so brain is left to untangle the time mess for certain tracks. And now new musicians are using IEM, it becomes a way lesser load for our brain.
So, the brain, instead of wasting the efficiency on other task, now would only see lookup table for grammar and process data, and enjoy the ambience.



And remember, ear loves to listen late

That's why loudspeaker sounds awesome
Then headphones
And then iem


Custom iem are different as they play on eardrum as intended by engineers, but cannot beat the old loudspeaker in a good damped room. (Condition is until or unless you go all out to match the impulse)


Well, I was creating an driver efficiency model from diaphragm to eardrum.

RAB design mentioned above is 72% efficient (exact value)....which is territory of planar magnetic headphone speeds for transient response.

RAB quad by @Xymordos is sitting at 86%. That is 4% away the STAX effortlessness sound(electrostatic bliss).

Etymotic ER4S was 68% efficient




For anyone starting new in IEM diy

-Try to make FR close to standard to a target (increases learning curve, but also teaches a lot more)
-Make impedance flat(via crossover or zobel)
-Try having flat phase response


I follow Harman kardon in ear target and sometimes diffuse field.

 

[dhruvmeena96]

My lord! This forum has accelerated out of the “middle” school level and straight into 4 year university knowledge and beyond. Very impressed with the last 250 pages of contributions.

Now for a dumb question: Anyone have opinions about the Knowles GV-32830 (quad)? I’ve got shells for my daughter sitting on my bench begging to get completed.

Well it is going to go even more crazier though. We all just started.

I was thinking of introducing cascaded circuit, series crossover, true resonator(still trying to comprehend what FIBAE black has, man @piotrus-g is genius), time domain correction, Ear compensation circuit etc

 

[dhruvmeena96]

@Furco your setup at pg206 gave birth to MASM by the way.

@DannyBouwhuis

Continuing
There is some evidence that large phase errors at low frequencies soften bass drum strikes. Loudspeakers develop large phase shifts near and below their low-end cutoff frequencies. This, in turn, produces group delays on the order of 5 to 15ms in that frequency range. Bass drum fundamentals then lag their upper harmonic components by this amount, which may explain this phenomenon. Countering this effect would require compensating bass amplitude and phase response flat down to below 10Hz or lower. This is the reason for muddy bass in old crappy earphones.

phase data is made up of two components: minimum phase and excess phase. The minimum phase response is related to the dips, peaks, and ripples in frequency magnitude response by a mathematical operation called the Hilbert transform. Any phase shift beyond this minimum phase shift is called excess phase, which is a measure of loudspeaker time dispersion. In particular, excess group delay, the derivative of excess phase with respect to frequency, has the units of time and is a measure of that dispersion.
Treble phase delay is okay, but something where we are sensitive, like 100hz to 1kHz, 2kHz up to 7kHz...is really bad as it can also change tonality and structuring of soundstage on a detail aspect.



Well I think, this much theory is okay for now

Prerequisite for a real soundstage

1. Time space domain linearity
2. Phase linearity
3. Impedance linearity
4. Lower distortion(all kind)
5. Lower noise
6. Less obstruction in sound path
7. Target response which focuses on soundstage like Harman Kardon target

Then to increase soundstage improve the parameters till 6 or increase some treble or bass

 

[Furco]

@Furco your setup at pg206 gave birth to MASM by the way.

I’m truly honored and greatly appreciate you giving credit. With all these advanced topics you’ve been discussing, I think it may be time for a new volume in the “definitive” diy guide. I usually do that for selfish reasons; it allows me in to digest the topics being discussed. It sounds like you’re bursting at the seams with exciting ideas. Please stay motivated! This is going to be a fun ride.

 

[Xymordos]

Clarity in the sense I posted it is just a fancy way of measuring impulse response (e.g. amplitude in first 80ms of a Signal vs amplitude Overall)

Imagine you had a room with a lot of reverb, your brain would have a hard time to isolate the signal from the reverb and you would loose information about where the sound is coming from. You would describe that as less clear.

But how do you measure ‚amount of reverb‘? The idea is that reverb builds up over time. You would first hear the original signal and then the reflections from the wall, adding volume to the original signal. If there were no reverb in the room, the signal would stay at the same volume. That’s why the clarity measurement is the relation of the first 80ms of signal (which should by dry) to the rest of the signal.

Phase coherence is also necessary for clarity as it helps you pinpoint the source of the signal and therefore separate the signal you focus on from the rest. Again in a room with a lot of reverb you will have to concentrate harder to isolate the signal you want to hear.

Of course an IEM would not introduce a lot of reverb, but it not react quickly to a new signal, causing the same issue (less sharp / less clear sounds)



Soundstage is caused by the phase difference and timing. Exactly that is why IEM need to be able to accurately reproduce phase and impulses. So it’s not an either/or scenario, better impulse response and phase stability (=clarity) directly improves soundstage.

If you look at my Bellsing 6 measurements above SPL and distortion is almost the same with or without zobel. Yet the soundstage is absolutely improved by the Zobel. The only measurement that seems to show that is clarity.

The clarity graph by RoomEqWizard does not care about phase though, it is just a fancy way to display impulse response.

Would the results of this be similar to the waterfall graph?

 

[dhruvmeena96]

I’m truly honored and greatly appreciate you giving credit. With all these advanced topics you’ve been discussing, I think it may be time for a new volume in the “definitive” diy guide. I usually do that for selfish reasons; it allows me in to digest the topics being discussed. It sounds like you’re bursting at the seams with exciting ideas. Please stay motivated! This is going to be a fun ride.

Well I am trickling some hearing aid tech

Buffer horn(discussed)
Resonator
Cascaded circuit(advanced level)
Etc

I will be motivated as you guys taught me how to make hearing aids(even though you discussed IEM per se)

 

[dhruvmeena96]

Would the results of this be similar to the waterfall graph?

Naaaahhh.....
It is cumulative of many things which tell that the signal hitting the microphone is how much resolved. Waterfall is decay and energy distribution plotting and is a part of clarity testing. Impulse tells a lot of thing if we know how to read it.

For example the perfect impulse structure of Hifiman RE400 and 600.

Almost all BA iem in old time performed bad on impulse response(except TDK BA200 and FIBAE1)

The thing is, now companies are taking this seriously and we as homemade iem should also

 

[DannyBouwhuis]

@dhruvmeena96
@eunice

Thanks for the awesome explanation, so adding delay would only create a 'artificial soundstage' I'm starting to understand some of this.. You guys are great !