Hi all,
Just looking for good advice.
Is it good form to XO @ a drivers natural rise of impedance if not using a zobel, or @ new impedance with zobel installed, or is it of no consequence?
Thanx, Mick.🙂
Just looking for good advice.
Is it good form to XO @ a drivers natural rise of impedance if not using a zobel, or @ new impedance with zobel installed, or is it of no consequence?
Thanx, Mick.🙂
If you listen carefully you can hear when a zobel is right, or wrong
You need to try many combinations
To just mount any one and think its ok doesnt work well
You need to try many combinations
To just mount any one and think its ok doesnt work well
Thanx Tinitus,
I'm not sure on this, however, does trying different zoble values shift xo point? Then is it a case of 'ballpark' both, (the xo and zobel), then work one or the other slightly one way or the other until it sings?
Cheers, Mick.🙂
I'm not sure on this, however, does trying different zoble values shift xo point? Then is it a case of 'ballpark' both, (the xo and zobel), then work one or the other slightly one way or the other until it sings?
Cheers, Mick.🙂
tinitus,
If you would, please hint at what you listen for to determine it is right. I don't mean this question as a flame, I am curious what your opinion is.
If you would, please hint at what you listen for to determine it is right. I don't mean this question as a flame, I am curious what your opinion is.
On the negative effects
I listen fore negative artefacts, mostly high frequency pitch
And blurred midrange
Small phase issues
And especially poor bass timing
Positive effects
Clarity
Good phase
Smooth natural sound
Easy on the ears
Good bass timing
Good definition
The last judgement is to have a relaxed feeling after long hours of listening
You should want to keep listening to music
Like if you simply cant get enough music
If you begin to feel stressed in any way, and want to shut it down, then its not good
Sometimes its really tiny tiny changes that makes the difference
Funny thing is that when the negative effects are avoided, the positive effects almost comes natural
Its kind of almost like it happens by itself, when it sound better WITH the zobel than without
The trick is to achieve improved clarity, and at the same time avoid any tendency to "tweetersound"
Voices should sound "full" with body, but not muffled and blurred
Almost like theres less distortion
When it works you will be able to listen to the most crappy music
And you will even enjoy listen to the news in radio, any kind of voices will sound like magic
My present speaker is made in a hurry, few days
Drivers just stand on the floor, on their magnets
15"/10"/2", just random speakers I had
You might call it omni-dipole
its not fantastic, but still quite enjoyable
Strange thing is that I cant play loud, but loose things close to the drivers still rattles a lot, when I try to play loud
Filter isnt perfect, but it works ok with the drivers
I listen fore negative artefacts, mostly high frequency pitch
And blurred midrange
Small phase issues
And especially poor bass timing
Positive effects
Clarity
Good phase
Smooth natural sound
Easy on the ears
Good bass timing
Good definition
The last judgement is to have a relaxed feeling after long hours of listening
You should want to keep listening to music
Like if you simply cant get enough music
If you begin to feel stressed in any way, and want to shut it down, then its not good
Sometimes its really tiny tiny changes that makes the difference
Funny thing is that when the negative effects are avoided, the positive effects almost comes natural
Its kind of almost like it happens by itself, when it sound better WITH the zobel than without
The trick is to achieve improved clarity, and at the same time avoid any tendency to "tweetersound"
Voices should sound "full" with body, but not muffled and blurred
Almost like theres less distortion
When it works you will be able to listen to the most crappy music
And you will even enjoy listen to the news in radio, any kind of voices will sound like magic
My present speaker is made in a hurry, few days
Drivers just stand on the floor, on their magnets
15"/10"/2", just random speakers I had
You might call it omni-dipole
its not fantastic, but still quite enjoyable
Strange thing is that I cant play loud, but loose things close to the drivers still rattles a lot, when I try to play loud
Filter isnt perfect, but it works ok with the drivers
Attachments
Hi tinitus,
Have you ever measured the impedance curves of your completed [zobel + driver combinations]? Are they flat?
Have you ever measured the impedance curves of your completed [zobel + driver combinations]? Are they flat?
I really dont know, but it would be interesting to see measurements
I have this thing with measurements, I dont trust them
Its the same with carburator adjustments
My bike runs better when I quit measureing, and adjust by using senses, and a minimum of "logic" approach, like monitor the heat, listen to how it sounds, and feel how it rides
Not quite the same as speakers, but its close
It has to give you that pleasent feeling, wanting more, without stress
Its like meditation
And like with any other thing, you need to practice
But I have to admit, my amps drives Infinity Kappa9 with ease
They are 2ohm at low frequency, and known to shut down most amps
Nice knowing amps can take whatever comes
Good amps are important to do any of it
Well, maybe I took this too far, fore a simple question
But maybe the answer just cant be simple
But checking with measurements?
Sure, that would be fun
Learn how it looks like, and the throw it way again😛
I have this thing with measurements, I dont trust them
Its the same with carburator adjustments
My bike runs better when I quit measureing, and adjust by using senses, and a minimum of "logic" approach, like monitor the heat, listen to how it sounds, and feel how it rides
Not quite the same as speakers, but its close
It has to give you that pleasent feeling, wanting more, without stress
Its like meditation
And like with any other thing, you need to practice
But I have to admit, my amps drives Infinity Kappa9 with ease
They are 2ohm at low frequency, and known to shut down most amps
Nice knowing amps can take whatever comes
Good amps are important to do any of it
Well, maybe I took this too far, fore a simple question
But maybe the answer just cant be simple
But checking with measurements?
Sure, that would be fun
Learn how it looks like, and the throw it way again😛
I have heard that 'magic' your refering to, (Krix Lyrix, believe it or not, (don't LOL @ that) nothing I've ever heard comes close!).
Just not sure how to get it, y'see, the driver i'm working with is vifa's p22.
As far as I can work out it has similar response charactaristics to the p21wo-39-08, Although I don't have the testing gear to confirm this, I find that if i work to the p21s impedance curve - 40%, I get the best results.
My question is should the xo be worked out @ drivers nom impe if the zobel is designed properly?
Sorry 4 the dumb questions!😱
Thanx, Mick.
Just not sure how to get it, y'see, the driver i'm working with is vifa's p22.
As far as I can work out it has similar response charactaristics to the p21wo-39-08, Although I don't have the testing gear to confirm this, I find that if i work to the p21s impedance curve - 40%, I get the best results.
My question is should the xo be worked out @ drivers nom impe if the zobel is designed properly?
Sorry 4 the dumb questions!😱
Thanx, Mick.
Prickears,
You should start at the nominal impedance of the xover band. If your zobel brings you to the nominal impedance of the speaker the use the nominal impedance. It is much easier with a zobel since the impedance is flatter.
I will agree with tinitus, you will end up tweeking to find the sweet spot. What order crossover are you planning?
Testing gear is cheap (free if you have a okay soundcard) for impedance. Speaker Workshop, an amp, and a sound card. If you buy a mic (fully calibrated $100US, 50$uncalibrated) and a mic preamp(30$US) you can do freq measurements to help guide your decisions. Regardless how you do it it is fun trying to find the magic.
You should start at the nominal impedance of the xover band. If your zobel brings you to the nominal impedance of the speaker the use the nominal impedance. It is much easier with a zobel since the impedance is flatter.
I will agree with tinitus, you will end up tweeking to find the sweet spot. What order crossover are you planning?
Testing gear is cheap (free if you have a okay soundcard) for impedance. Speaker Workshop, an amp, and a sound card. If you buy a mic (fully calibrated $100US, 50$uncalibrated) and a mic preamp(30$US) you can do freq measurements to help guide your decisions. Regardless how you do it it is fun trying to find the magic.
Last edited:
If you mean "nominal imp", it doesnt really exist
Its just an estimate and pretty useless
Funny thing is that when a zobel is mounted on say a woofer, some change might be heard on the midrange driver as well, if 3way
A zobel couples to ground, but other than that Im not sure what really happens
But zero effect on midrange driver from woofer zobel, may be achieved with careful componet adjustment
But you cannot calculate these things at all
Flat straight impedance is only possible with measurements
thanx 4 your help here guys things are a little clearer now.
I have built the xo 2nd order L/R 5 ohm tweeter (seas 27tff) and 2nd L/W 8 ohm (vifa P22).
The P22 is a 4 ohm nom driver but seems to be quite close 2 8 ohm @ the 3.3k xo point used here (high due to tweeter limitations/compromise).
The response of the P22 seems quite flat despite the rising imp, so I crossed it accordingly. Is that what we aim to achieve with a zobel (flatter response)?
or easier intergration with other drivers? or both/other?
I get much cleaner sound without a zobel, (lack of knowledge on the subject being the main problem here I think), but also increased listening fatigue, if that makes any sense whatsoever.
Thanx guys, Mick.🙂
I have built the xo 2nd order L/R 5 ohm tweeter (seas 27tff) and 2nd L/W 8 ohm (vifa P22).
The P22 is a 4 ohm nom driver but seems to be quite close 2 8 ohm @ the 3.3k xo point used here (high due to tweeter limitations/compromise).
The response of the P22 seems quite flat despite the rising imp, so I crossed it accordingly. Is that what we aim to achieve with a zobel (flatter response)?
or easier intergration with other drivers? or both/other?
I get much cleaner sound without a zobel, (lack of knowledge on the subject being the main problem here I think), but also increased listening fatigue, if that makes any sense whatsoever.
Thanx guys, Mick.🙂
BTW, does anyone have a pair of Gedlee's speakers?
Just going by reviews, They sound like a nice bit of gear. I would be keen to hear some before i die!🙂
Just going by reviews, They sound like a nice bit of gear. I would be keen to hear some before i die!🙂
Hi tinitus,
Is your 15" a nominal 4 Ohm driver?
If the R in the RC (of Zobel) is approaching towards small value, then the whole thing will tend to become a 2nd order filter with a very rounded shoulder (low Q curve).
That's a very good tuning tool, anyway. No matter how you see it. I myself like to play around this, too. Also on the added serial resistor for inductor in a 2nd order HP filter.
Is your 15" a nominal 4 Ohm driver?
If the R in the RC (of Zobel) is approaching towards small value, then the whole thing will tend to become a 2nd order filter with a very rounded shoulder (low Q curve).
That's a very good tuning tool, anyway. No matter how you see it. I myself like to play around this, too. Also on the added serial resistor for inductor in a 2nd order HP filter.
If you use a crossover that was designed without the Zobel, then adding one will not be a positive thing. It will change (and possibly/probably not for the better) the crossover's transfer function.
Contrarian view:
Zobels have been useful in times past because a flatter driver impedance is easier to cross over if you are using textbook methods. These days, almost no-one designs a crossover without CAD, and it is often (maybe even "usually") possible to hit the target function without Zobels or other compensation networks (though a series RC is still often useful, its values will be no-where near the Zobel values). It's usually better to think in terms of target transfer functions, and optimize the crossover topology and values with that in mind- it's not 1975 anymore.
And, of course, with the simplicity of digital crossovers and the availability of inexpensive, high quality amplification, active crossovers and multiamping make the whole impedance compensation approach moot.
Contrarian view:
Zobels have been useful in times past because a flatter driver impedance is easier to cross over if you are using textbook methods. These days, almost no-one designs a crossover without CAD, and it is often (maybe even "usually") possible to hit the target function without Zobels or other compensation networks (though a series RC is still often useful, its values will be no-where near the Zobel values). It's usually better to think in terms of target transfer functions, and optimize the crossover topology and values with that in mind- it's not 1975 anymore.
And, of course, with the simplicity of digital crossovers and the availability of inexpensive, high quality amplification, active crossovers and multiamping make the whole impedance compensation approach moot.
WRT, RC, (zobel) logic tells me C chooses freq, and R chooses constant impedance one wishes to adhere to.
Again WRT BSC, inductor chooses freq, R lets a certain amount of signal through unhindered (somewhat) by by the inductor.
Is this right or am I way off the mark?
Mick.
Again WRT BSC, inductor chooses freq, R lets a certain amount of signal through unhindered (somewhat) by by the inductor.
Is this right or am I way off the mark?
Mick.
SY,
Yes, I'm hearing negative effects when I add a zobel after plotting xo.
Again logic tells me that the xo point would move down if its done in this order, but I don't know enough to figure this is the actual case. It certainly sounds like it, which supports my hunch, (midrange move to far back........not as forward sounding).
Also I figure the less corrective circuits the better, in general.
Mick.🙂
Yes, I'm hearing negative effects when I add a zobel after plotting xo.
Again logic tells me that the xo point would move down if its done in this order, but I don't know enough to figure this is the actual case. It certainly sounds like it, which supports my hunch, (midrange move to far back........not as forward sounding).
Also I figure the less corrective circuits the better, in general.
Mick.🙂
Not sure what a BSC is refering to.
In the RC the R is in series with the C, one after the other. The R provides impedance at frequencies were the capacitor passes signal. Kind of a limiter, without it at high frequencies it would shunt the whole signal to a very low impedance. Like a short. As the frequency lowers the impedance of the capacitor increases ultimatley becoming much greater than R typically. R in parallel with the R of the voice coil will set the impedance of the circuit. More accuratley, the impedance of the RC circuit and the voice coil in parallel willset the overall impedance.
In regards to above, less is typically better but it becomes more compicated to tweek. The computer doen't listen to your speakers for enjoyment. ;-)
In the RC the R is in series with the C, one after the other. The R provides impedance at frequencies were the capacitor passes signal. Kind of a limiter, without it at high frequencies it would shunt the whole signal to a very low impedance. Like a short. As the frequency lowers the impedance of the capacitor increases ultimatley becoming much greater than R typically. R in parallel with the R of the voice coil will set the impedance of the circuit. More accuratley, the impedance of the RC circuit and the voice coil in parallel willset the overall impedance.
In regards to above, less is typically better but it becomes more compicated to tweek. The computer doen't listen to your speakers for enjoyment. ;-)
Last edited:
Makes perfect sense
Especially if clean sound is bright sound
If it sound better without a zobel, it doesnt mean it doesnt work with a zobel at all
It may just mean that zobel is wrong values