PDA

View Full Version : Zobel for 515B's



quindecima
11-12-2006, 11:30 AM
I would like to calculate a Zobel for my 8 ohm Altec 515b's but do not have nor can find the Le value for the driver. I think I can use a 8 0hm 10 watt resistor but need a cap value. Does anybody have that info? :blink:

Hoerninger
11-12-2006, 12:32 PM
Do you know this?
http://www.thielesmall.com/
(515B: 1 mH)
____________
Peter

quindecima
11-12-2006, 12:58 PM
Good site! I see that the L is listed as 0.001 mh. Would that be millihenries or micro henries. I am not real familiar with the usual listing of henries. :) thanks for the info!

Hoerninger
11-12-2006, 02:28 PM
Input error I suppose: 1mH = 0,001 H.
0,001 mH is impossible for a speaker.
____________
Peter

Tom Loizeaux
11-15-2006, 05:48 PM
My limited understanding is that woofers really don't need Zobel networks.
Midrange drivers and tweeters usually benifit from these networks.

Tom

4313B
11-15-2006, 06:08 PM
Too generalist. It all depends on what you wish to accomplish.

One can generally get away with not using a conjugate if they use higher order filters.

Sometimes the conjugate is purposely left off a high frequency transducer to allow for high frequency extension.

One can also use a conjugate to tweek the rolloff of the filter.

There is no hard and fast rule that one has to use the "Zobel" formula to come up with a certain set of values. In reality one can make the conjugate any way they wish to accomplish the effect they desire/require.

Too Tall clist
04-27-2007, 08:11 AM
Too generalist. It all depends on what you wish to accomplish.

One can generally get away with not using a conjugate if they use higher order filters.

Sometimes the conjugate is purposely left off a high frequency transducer to allow for high frequency extension.

One can also use a conjugate to tweek the roll off of the filter.

There is no hard and fast rule that one has to use the "Zobel" formula to come up with a certain set of values. In reality one can make the conjugate any way they wish to accomplish the effect they desire/require.

You need a zobel if the woofer (or mid) has a substantial voice coil causing a large Le (voice coil inductance).
Inductance causes impedance to rise as the frequency increases.

You can also tweak it, as you point out, to change the roll of the high pass, though this is difficult without modeling software.
You can also adjust the low pass to make up for it if it is not too severe. If the Le is large depending on how high the xover point (the higher the worse) you have no choice. If you do not use a zobel you get a very nasty peak.
If you were to measure the impedance of the driver a couple of octaves either side of the prospective low pass crossover point you look to see if the impedance is changing at a high rate.
The problem is that passive filters are effected by the impedance they see. So for example the xover is supposed to be at 1kHz. If the impedance at 1kHz is 8 ohms, but by the time you get to 1200Hz its is 16 0hms the high pass filter will not operate correctly since it was designed for a 8 ohm load THROUGH OUT THE XOVER RANGE.

4313B
04-27-2007, 08:22 AM
What are you quoting me for?

And we've already been over all this in virtually every version of this forum.

General crossover design is beyond the scope of this website as has been stated several times. If you have a specific project you are working on that relates to Altec or JBL then I will see if and when I can fit you in but don't hold your breath. My backlog is literally measured in years at this point.

Too Tall clist
04-27-2007, 09:09 AM
I would like to calculate a Zobel for my 8 ohm Altec 515b's but do not have nor can find the Le value for the driver. I think I can use a 8 0hm 10 watt resistor but need a cap value. Does anybody have that info? :blink:

If you go strictly by formula the DC Re of the driver is the resistor value. Then you need the Le to calculate the cap.
If you want to spend time charting the impedance point by point with a Multi meter, a 1% resistor and tones (could be from a CD or generator) I can model and tell you the perfect cap.

Here is a picture of a 12” woofer and a 1” horn.
It shows the raw impedance of both drivers and what changed when I added the zobel on the woofer and the series impedance shunt on the 1”.
This woofer is very efficient and therefore can not reach very low which often means the Le is not very large.
The xover was at 1200Hz so it is possible I could have gotten away without using one.
At 600Hz it was about 8 ohms and at 2400Hz it is about 12 ohms. In the end I used it to do some “tweaking” also.
You should be able to figure things out from the legend on the graph.
Too Tall

4313B
04-27-2007, 09:15 AM
Right. What we are really looking for is someone who has a 515B to run an impedance curve and then we can go from there. :)


If you go strictly by formula the DC Re of the driver is the resistor value.You can make it whatever you need it to be. Oftentimes it is Re * 1.1 but certainly not always. Anyway... if anyone has a 515B they can measure...

415A for this too - http://www.audioheritage.org/vbulletin/showpost.php?p=165449 since that thread revived this one.

Storm
04-27-2007, 09:21 AM
Exuse me for asking, but was is a Zobel?

thanks.

-Storm.

4313B
04-27-2007, 09:24 AM
Exuse me for asking, but was is a Zobel?

thanks.

-Storm.
Google zobel network and boucherot cell. JBL has traditionally referred to them as conjugate networks.

Too Tall clist
04-27-2007, 09:25 AM
My limited understanding is that woofers really don't need Zobel networks.
Midrange drivers and tweeters usually benifit from these networks.

Tom


Hi Tom,
With dome tweeter the resonant peak is often down at 1kHz and below and the xover often above 2kHz. Above the resonant peak the rise is often gentle so no zobel required.

A mid woofer that can reach up high usually has a small Le and flat impedance (except for resonant peak). No zobel required though a series resonance filter might help.

If it helps you can look at a zobel as the cousin of a shelf EQ and a series resonance filter as a cousin to a parametric EQ notch.
One deals in impedance and the “cousin” in frequency SPL.

On a horn the impedance peak is often right where your xover is. Even if the xover is an octave below it the notch filter you need if it is a CD horn runs square into the impedance peak. I have seen notch filters that did NOTHING till I fixed the impedance peak. They did not effect the frequency response at all.

Another case is a Large ESL. If you try to use a passive xover in the 300Hz to 600Hz area you will find a very high impedance peak making it impossible to use a passive xover.
Too Tall

4313B
04-27-2007, 09:27 AM
Hi Tom,
With dome tweeter the resonant peak is often down at 1kHz and below and the xover often above 2kHz. Above the resonant peak the rise is often gentle so no zobel required.

A mid woofer that can reach up high usually has a small Le and flat impedance (except for resonant peak). No zobel required though a series resonance filter might help.

If it helps you can look at a zobel as the cousin of a shelf EQ and a series resonance filter as a cousin to a parametric EQ notch.
One deals in impedance and the “cousin” in frequency SPL.

On a horn the impedance peak is often right where your xover is. Even if the xover is an octave below it the notch filter you need if it is a CD horn runs square into the impedance peak. I have seen notch filters that did NOTHING till I fixed the impedance peak. They did not effect the frequency response at all.

Another case is a Large ESL. If you try to use a passive xover in the 300Hz to 600Hz area you will find a very high impedance peak making it impossible to use a passive xover.
Too TallTom is dead. He was a great guy and he and I and G.T. already had these conversations with respect to his 4343's. Rather than quoting me and then ignoring me and then continuing to show off just drop it.

grumpy
04-27-2007, 09:36 AM
TTc,

Although it was kind enough for you to respond to Tom's message, don't
wait up for a thank you... although I'm sure he would have, kind gentleman
that he was.

http://audioheritage.org/vbulletin/showthread.php?t=14123&highlight=tom+loizeaux

-grumpy

P.S. ... appears Giskard captured part of what I wanted to say as well.

Ian Mackenzie
04-27-2007, 03:17 PM
What are you quoting me for?

And we've already been over all this in virtually every version of this forum.

General crossover design is beyond the scope of this website as has been stated several times. If you have a specific project you are working on that relates to Altec or JBL then I will see if and when I can fit you in but don't hold your breath. My backlog is literally measured in years at this point.

I dont think there is anything called general crossover design is there?....LOL.



Ian

4313B
04-27-2007, 03:22 PM
There's general and major, just not colonel.

Ian Mackenzie
04-27-2007, 03:28 PM
...LOL.......I get it :blink: ..Minor and Major:)

Too Tall clist
04-28-2007, 09:03 AM
Tom is dead. He was a great guy and he and I and G.T. already had these conversations with respect to his 4343's.

Rather than quoting me and then ignoring me and then continuing to show off just drop it.

My apologies.
I tried to do too much at once and stepped on accepted Internet etiquette.
I might not reply to your post promptly because of other chores at home that demanded attention.
Or I might have then answered someone else first after failing to note your response.

Please note that replies are often written to a forum audience at large and phrased to help people who might need simple background information and not to slight the knowledge of the person’s post that was quoted.

4313B
04-28-2007, 09:16 AM
My apologies.Thank you! :)

I tried to do too much at once.Understood. That was exactly the way we saw it. We look forward to further posts from you so thank you for signing up.