[IvanNOON]

@dhruvmeena96
Tried L- and T-pads today and they seem to work. Kinda. (probably because the values are not exact)
There is some sub loss but less than with just a resistor. General signature and dynamics of the driver are preserved, which is nice.

I'm currently experimenting with IR and RC low-pass filters.
IR seem a bit unpredictable/messy because inductors have notable resistance (14Ohms for 1mH) which needs to be accounted for.
But it could be an advantage because you can combine L-pad with 2nd order low-pass or T-pad with 3rd order low-pass in some configurations.

 

[Slater]

@dhruvmeena96
Tried L- and T-pads today and they seem to work. Kinda. (probably because the values are not exact)
There is some sub loss but less than with just a resistor. General signature and dynamics of the driver are preserved, which is nice.

I'm currently experimenting with IR and RC low-pass filters.
IR seem a bit unpredictable/messy because inductors have notable resistance (14Ohms for 1mH) which needs to be accounted for.
But it could be an advantage because you can combine L-pad with 2nd order low-pass or T-pad with 3rd order low-pass in some configurations.

I’m having good luck using a base plate of prefabulated amulite.

 

[dhruvmeena96]

@dhruvmeena96
Tried L- and T-pads today and they seem to work. Kinda. (probably because the values are not exact)
There is some sub loss but less than with just a resistor. General signature and dynamics of the driver are preserved, which is nice.

I'm currently experimenting with IR and RC low-pass filters.
IR seem a bit unpredictable/messy because inductors have notable resistance (14Ohms for 1mH) which needs to be accounted for.
But it could be an advantage because you can combine L-pad with 2nd order low-pass or T-pad with 3rd order low-pass in some configurations.

don't use inductor until or unless you are not making a series notch filter for your driver. and inductor has to be SMD air coil

series notch filter is for fixing the EMF(Electro-motive force) resonance which creates peak in Impedance graph and phase shift at specific frequency response response

the only driver which you can work on is RAF, ED and CI(Impedance graph available)

crossover is good in RC and is better after you make the reactive load purely resistive instead impeding.

If you use coil, it will become hard to control some factors.

and it is good if you know Theile small parameters of the BA before using inductor so that you can track the change in 2nd order harmonic due to inductor.

well let me explain

When we do a low pass
(RC crossover)
Resistor in series and capacitor in parallel

we add Voice coil resistance due to series resistance and also increase Moving Mass of BA(But the mass of diaphragm is fixed, right.....actually it is fixed, from what i mean to say, you slow down your driver for lower freq)

When we do a High pass

(RC crossover)
Capacitor in series and resistor in parallel

we first reduce the mechanical damping or electrical damping by parallel resistor, you can take it in any way(Electrical damping it is, but it is simple to calculate while taking value of mechanical damping). This make the driver have the ability to move faster and louder but can distort due to over excursion. then adding capacitor technically reduce the movement zone, which lead the driver to play loud on HF. You can say, adding capacitor in series reduce moving mass of diaphragm and resistance reduce mechanical damping


But when you do an inductor crossover
Things change as now you are controlling compliance ( compliance = z* 1/stiffness , where z is a constant of suspension).
This means you are controlling the voice coil directly instead of frequency as RC type does.

Here you can control the level of slope with more freedom and more precisely, but it changes the driver itself, which is okay on loudspeaker because they are loud, bigger and inefficient compared to super sensitive BA.

The bigger nature of loudspeaker prevents a drastic change in sound and voice coil of a loudspeaker have a low inductance compared to 7.8mH on RAF driver with 22ohm impedance vs Mark audio alpair7 21.30uH and 5.48ohms.

plus loudspeaker do need inductor to actually show some real change in sound profile while crossover by desire of ours.

BA higher impedance also means the EMF(Motive force) and resonance lies higher in 2kHz compared to dynamic driver low frequency resonance and a very high rise in treble impedance. Higher inductance of BA show variability of BA in phase nature and inductor crossover can do some changes which might not be good


untill or unless knowles dont release Qts, Qms, Qes and other parameter, dont try using inductor

well you can, but then you have to calculate and derive the things properly too(I can help for some drivers though and can give some formula)


plus one more thing.....SMD small core inductor may be tempting but never buy for BA purpose becasue that small inductor core may saturate, again showing some bad behavior in impedance. always get an aircoil and also tape your BA with MU metal tape(dont plug tape or vents and nozzle, only body)

Note:
well RC crossover can be parallel or series to any value here according to calculation in depth for perfect crossover but you can only see two inductor

1. Impedance
2. Springiness of coil or stick in BA(Compliance in parallel)

Impedance is okay

springiness needs to be stiffer and joining a series notch filter to Driver may load an inductor parallel to compliance, which may reduce the resonant frequency and rising impedance

but the cap and resistor are also that much important


One more thing, the ideal crossover cross is smooth and symmetrical but that doesn't really happen because the crossover component doesn't work as intended

Reason
1. We seriously don't know driver is working properly(Q factors and thiele parameters)
2. SMD component has more error than rated
3. Bigger component have more tolerance and less deviation



to be continued.....................

More complex part ahead

to cover:

how a capacitor, Resistor and Inductor affects the signal in line.
Impulse response
Burst Decay
CSD
Harmonics
Resonator and Formula's(I still dont get the FIBAE BLACK, so dont expect you will make a FIBAE BLACK)

I’m having good luck using a base plate of prefabulated amulite.

well.....hmmm i didnt get the joke

prefabulated amulite is fictional thing ryt(I know i destroyed/ruined the joke, but still the reference is out of my comprehension)

What lol


@piotrus-g

well why FIBAE 1 is my favourite till now

He got the custom driver according to his spec and he horned it to make a limited band a wide band



plus BA are fullranger due to resonance at 2kHz to 3kHz and if you remove the resonance for perfect phase and linear impedance will lead to true nature
(BA are limited by nature, sealed box so small can create the impedance so high in mids compared to dynamic resonance in bass)

Dynamics are real full range because it has been perfected over the period(DITA Fealty and truth, Hifiman RE2000) or this DIY driver

The driver i linked is legendary for diy purpose and even outclass DITA or Senn IE800s driver in everything

and it is true 48kHz extension with <1.4us impulse response. Faster than any BA which are <1.6 for SWFK(Fastest)

Note:
BA with flat impedance are faster. Again FIBAE 1 as example

BA were introduced very late according history but are technically faster for smaller voice coil movement and diaphragm. But iss

 

[Slater]

well.....hmmm i didnt get the joke

prefabulated amulite is fictional thing ryt(I know i destroyed/ruined the joke, but still the reference is out of my comprehension)

What lol

Ah, it’s an engineering joke (a light hearted reference to technobabble).

Spoiler: See this hehe

 

[IvanNOON]

don't use inductor until or unless you are not making a series notch filter for your driver. and inductor has to be SMD air coil

series notch filter is for fixing the EMF(Electro-motive force) resonance which creates peak in Impedance graph and phase shift at specific frequency response response

the only driver which you can work on is RAF, ED and CI(Impedance graph available)

crossover is good in RC and is better after you make the reactive load purely resistive instead impeding.

If you use coil, it will become hard to control some factors.

and it is good if you know Theile small parameters of the BA before using inductor so that you can track the change in 2nd order harmonic due to inductor.

well let me explain

When we do a low pass
(RC crossover)
Resistor in series and capacitor in parallel

we add Voice coil resistance due to series resistance and also increase Moving Mass of BA(But the mass of diaphragm is fixed, right.....actually it is fixed, from what i mean to say, you slow down your driver for lower freq)

When we do a High pass

(RC crossover)
Capacitor in series and resistor in parallel

we first reduce the mechanical damping or electrical damping by parallel resistor, you can take it in any way(Electrical damping it is, but it is simple to calculate while taking value of mechanical damping). This make the driver have the ability to move faster and louder but can distort due to over excursion. then adding capacitor technically reduce the movement zone, which lead the driver to play loud on HF. You can say, adding capacitor in series reduce moving mass of diaphragm and resistance reduce mechanical damping


But when you do an inductor crossover
Things change as now you are controlling compliance ( compliance = z* 1/stiffness , where z is a constant of suspension).
This means you are controlling the voice coil directly instead of frequency as RC type does.

Here you can control the level of slope with more freedom and more precisely, but it changes the driver itself, which is okay on loudspeaker because they are loud, bigger and inefficient compared to super sensitive BA.

The bigger nature of loudspeaker prevents a drastic change in sound and voice coil of a loudspeaker have a low inductance compared to 7.8mH on RAF driver with 22ohm impedance vs Mark audio alpair7 21.30uH and 5.48ohms.

plus loudspeaker do need inductor to actually show some real change in sound profile while crossover by desire of ours.

BA higher impedance also means the EMF(Motive force) and resonance lies higher in 2kHz compared to dynamic driver low frequency resonance and a very high rise in treble impedance. Higher inductance of BA show variability of BA in phase nature and inductor crossover can do some changes which might not be good


untill or unless knowles dont release Qts, Qms, Qes and other parameter, dont try using inductor

well you can, but then you have to calculate and derive the things properly too(I can help for some drivers though and can give some formula)







plus one more thing.....SMD small core inductor may be tempting but never buy for BA purpose becasue that small inductor core may saturate, again showing some bad behavior in impedance. always get an aircoil and also tape your BA with MU metal tape(dont plug tape or vents and nozzle, only body)



well.....hmmm i didnt get the joke

prefabulated amulite is fictional thing ryt(I know i destroyed/ruined the joke, but still the reference is out of my comprehension)

What lol

Thanks for the write-up, man! That cleared up a lot of things.
Do you have any more BA inductance figures? I wanna try Zobel sometime soon.

I’m having good luck using a base plate of prefabulated amulite.

Hah, after I re-read what I wrote I got the technobabble vibe too. But I assure you, my babble actually makes sense

 

[dhruvmeena96]

Ah, it’s an engineering joke (a light hearted reference to technobabble).

Spoiler: See this hehe

He is amazing...i know that joke

damm

but i tried writing things in simple manner as possible

thats why i didnt get the joke

lol

Thanks for the write-up, man! That cleared up a lot of things.
Do you have any more BA inductance figures? I wanna try Zobel sometime soon.



Hah, after I re-read what I wrote I got the technobabble vibe too. But I assure you, my babble actually makes sense

more coming up with even more specific examples







plus multi driver is not good as the more driver and more crossover creates more feedback, changing sound structure, destroying phase etc

i have U18t tzar and i feel they tried fixing it

by 8 driver bass, 8 driver mids , TWFK with its tweeter in tia arrangement


making it three way, so less feedback and simplistic impedance curve

but i feel i can enjoy it and not mix in it

dont know why

 

[dhruvmeena96]

well, there is one more thing

Voice coil saturation, a plague nobody can avoid

electrostatic and planar doesn't have it
dynamic and BA have it

well, you can see this diagram, Voice coil saturation happens at resonance frequency in dynamic and both in BA
one more BA design limitation

BA is technically a sealed/vented compression driver designed on telephone armature.

and the body is actually a shield body, which can make internal coil saturated, but is still better than core. This may not come when we benchmark but can effect real time usage etc

Ferrofluid may show better sonic and chart response but makes it more susceptible to saturation of signal, thus again limiting extension on both side

so please try Zobel and series notch filter as it solves many issues like this(some hard work, but helps saturation and distortion level)

calculators are available online

if series notch filter need explaination, pm me

 

[dhruvmeena96]

The Fletcher–Munson curves are one of many sets of equal-loudness contours for the human ear, determined experimentally by Harvey Fletcher and Wilden A. Munson, and reported in a 1933 paper entitled "Loudness, its definition, measurement and calculation" in the Journal of the Acoustical Society of America.


the first research on the topic of how the ear hears different frequencies at different levels was conducted by Fletcher and Munson in 1933. Until recently, it was common to see the term Fletcher–Munson used to refer to equal-loudness contours generally, even though a re-determination was carried out by Robinson and Dadson in 1956, which became the basis for an ISO 226 standard.

It is now better to use the generic term equal-loudness contours, of which the Fletcher-Munsons curves are now a sub-set, and especially since a 2003 survey by ISO redefined the curves in a new standard.

According to the ISO report, the Robinson–Dadson results were the odd one out, differing more from the current standard than did the Fletcher Munson curves. The report states that it is fortunate that the 40-phon Fletcher–Munson curve on which the A-weighting standard was based turns out to have been in agreement with modern determinations.

The article also comments on the large differences apparent in the low-frequency region, which remain unexplained. Possible explanations are:

• The equipment used was not properly calibrated.
• The criteria used for judging equal loudness at different frequencies had differed.
• Subjects were not properly rested for days in advance, or were exposed to loud noise in traveling to the tests which tensed the tensor tympani and stapedius muscles controlling low-frequency mechanical coupling.

So, please follow Robinson–Dadson loudness curve for volume matching instead of Fletcher-Munson curves(many audio company still do old one,as it is easier to do)

Robinson–Dadson article stated

"The article also comments on the large differences apparent in the low-frequency region, which remain unexplained. Possible explanations are:

• The equipment used was not properly calibrated.
• The criteria used for judging equal loudness (which is tricky) differed.
• Different races actually vary greatly in this respect (possible, and most recent determinations were by the Japanese).
• Subjects were not properly rested for days in advance, or were exposed to loud noise in travelling to the tests which tensed the tensor timpani and stapedius muscles controlling low-frequency mechanical coupling.

"



so this is what we call Chi-Fi, japanese tuning or some other tuning from either extremely new company like kumitate labs, Custom art(well i think, he never thought of this but achieved it....but i can be wrong, because according to my dad, peter is crazy freak perfectionist.....lol....i am still laughing @piotrus-g )

Note: by the way, my dad, me and my sis love FIBAE 1 and my younger brother love sony XBA z5(The grand bass)

or from Sony in house sound


Germans still follow from mic based Fletcher-munson

well making a Robinson–Dadson is easy.

Tune your earphone loudness curve first to Fletcher-munson and then ear test it so that you get real volume attenuation instead of slight acoustic change. measure that and repeat with 10 guys with different ear shapes and eardrums. Take an average and here you have it.


by the way, it is curve for loudness and not standard curve for sound

 

[IvanNOON]

so please try Zobel and series notch filter as it solves many issues like this(some hard work, but helps saturation and distortion level)

I only have experience with parallel notch as it is widely used in speaker building. Why do you recommend series notch?

 

[dhruvmeena96]

I only have experience with parallel notch as it is widely used in speaker building. Why do you recommend series notch?

you can see speaker diagram

How adding components effect speaker internal performance
it is easy to monitor performance of notch and acoustic change

you can do a parallel notch but check how it effect the internal circuit parameter

and for parallel notch, you really need Q factors for precise effect(am i right? you can correct me if i am wrong)

on a parallel notch, it is a series cap with resistor and inductor in parallel

which then you have to calculate inductor resistance and parallel resistor, and how to get precise value.(Reason 1)
Adding a cap in series make it an high pass, which is bad for full range(Reason 2)
parallel inductor only reduces the voice coil inductance and not the compliance directly, it may seem same effect, but you need more value of it(Reason 3)
Parallel resistor reduces the voice coil resistance directly, making it an high pass(Reason 4)

well, it is general definition according to speaker equivalent circuit




it is just for taking the calculation
it just do the same things like a Series notch filter

but series notch filter actually helps in focusing a specific peak resonance when we dont have proper data on driver

if i have all the theile parameter, i will do a parallel notch any day instead of series notch(it seams easy with parameters)

Series Notch filter

parallel notch filter






well the best part is you can swap RLC to anyplace in parallel notch filter, so i cannot determine most of the value without all given datasheet in driver

as i can be wrong on above part about parallel notch but what i know is that RLC works in tandem compared to discreet in series notch




well now i think, you can use a parallel notch filter(i think it is easier to do compared to series)

here is the calculator for parallel
http://www.carstereo.com/help/Articles.cfm?id=19

For checking of impedance on reference
http://www.mh-audio.nl/parallelnotchfilter.asp
check the impedance value change here


series(I have some other way, which is proprietary of mine for series)
http://www.carstereo.com/help/Articles.cfm?id=23

but let me tell you, you would have to do hit and trail for exact damping profile



L and C in such circuit control the Q factor(kurtosis or the area of frequency affected) whereas the R effects the damping

if we add only L and C, we can control the peak rising and falling "area" making the peak narrow or broad


see, this is the control we can have with specific circuitry

Where did i get motivation of posting this thing

i saw a guy using orthodox t notch circuit on SR driver here, on this thread and accidentally from my point of view damped a lot.

and people adding simple resistor for damping profile, which is wrong because it control specific property and doesn't not work like we expected

If we are going single driver, we still have a lot of space for smd components which can actually help us tune a lot of thing, and even to the core of how it is producing sound

single driver has a lot of possibility, you know

 

[dhruvmeena96]

Well, should i cover more circuit and the properties of it, or should i jump in structural things like resonator

Well, i was working with back wave from vent and it can also produce some drastic effect

if i can load the back-wave in front wave with specific damping and with same phase as front nozzle, i can take the bass on amazing level. Actually up to 5Hz with no roll off on a RAF

will tell in coming stuff i am going to writing.....lol

By the way

RAF is amazing piece of driver....

In my theory and some of my calculation, it is performing better than ED and TWFK on mostly everything.

Actually better than Sonion 1723, if I am able to tune properly.

And with back loaded vent, I can attach a tube on vent pretty easily

 

[dhruvmeena96]

One more thing, before notch equalising, please add a zobel circuit first, it helps for proper calculation result

 

[Mart.oporto]

Hello friends!
I want to say hello to everyone first of all and thank you for the valuable information that every day they teach for free for the love of technology.
I'm dabbling slowly in DIY iems, but now I have to buy headphones from another brand.
I have saved $ 1000 and would like to know which CIEM model they recommend. I could stretch my budget to $ 1500 if it really deserves it.
I hope I'm not disrespecting you with this question.
I send my love from Argentina!

 

[dhruvmeena96]

Hello friends!
I want to say hello to everyone first of all and thank you for the valuable information that every day they teach for free for the love of technology.
I'm dabbling slowly in DIY iems, but now I have to buy headphones from another brand.
I have saved $ 1000 and would like to know which CIEM model they recommend. I could stretch my budget to $ 1500 if it really deserves it.
I hope I'm not disrespecting you with this question.
I send my love from Argentina!

make one yourself or go chi-fi. Chinese nowadays are nailing it. get a BGVP DM6 custom in ear monitor which will save you a lot of money and will get a near flagship sound(300$ 5 driver) or Rose BR7(800$ for 7driver) or Tansio Mirai Star 7 custom on taobao(1000$ for 7 driver).

But if you really are into pure acoustic mastery, Go for FIBAE 1(My reference) or FIBAE BLACK(What people are calling the best stage, bass a single driver IEM has produced till date). If you can spend upto thousand and want a spiral ear SE5 Ultimate sorta sound, get a Warbler Audio Prelude(1000$). It is softer and less bassier but mids are actually better than spiral ears.

well if you are Stage performer, Musician or an audiophile, go for FIBAE Black. If you are in mixing, get FIBAE 1. If you are in driver count, go chi-fi(Less price for more driver). If you want pure audiophile, get a Warbler Prelude

all lies within $1000

but wait, invest and save money more, because this thread can burn your money Caution
but the work really pays back, and no matter what iem you gonna buy, you will love your own creation more(Because its yours).

well more info is going to come, and subscribe(on top right) and keep yourself updated
DIY creation here help us to improve our skill, tuning. And this thread is actually one of the best threads on head-fi to learn(@bluemonkeyflyer T50RP thread, @Bill-P headphone mods and @Slater and me destroying Philips SHL3300 thread by moding the heaccckkk out of the 40$ Headphone(Well,see his work), @CoiL IEM gallery(His woodwork level is nearing Ocharaku level after the iBasso it01 wood work he did on back vent of dynamic)
This thread is even crazier but fun. You are welcome here bro @Mart.oporto . By the way, you gave us the whole country love(damm, thats big). ROFL

by the way, if you ever start making an IEM
1. @Furco compiled cookbook(amazing work)
2. Learn RC circuit crossover[/SIZE]
3. Learn Tubing and Damper from someone who can explain in more depth here. I also have problem there because those are very variable for me to understand

That all you need to know before designing an IEM on paper

all others would gladly teach you all the way through

 

[IvanNOON]

Series Notch filter

parallel notch filter

Ah, nevermind then, I mixed the them up. (I was thinking parallel with the driver)
Series FTW