• Please note that Audiokarma will be offline briefly for updates early on Monday morning, September 28th.

Adding Zobel to existing crossover?

Bratwurst7s

Bratwurst7s

In The Frying Pan
Subscriber
I'm considering adding a zobel network to the midrange of my Speakerlab 7s. The new mids that I installed are almost a bit too bright and I was hoping that the zobel might help the crossover to keep unwanted high freqs away from it. Originally the crossover only has a zobel on the 2 woofers.

My concern/question is what if any detrimental effect that would have on the crossover point. Would it be shifted up or down?

Thanks,
James
 
Register to hide this ad
A Zobel is used to flatten the electrical impedance of a speaker, countering the natural rise due to inductance. If they designed the original crossover for a fixed impedance, the Zobel would help. I doubt they did, more likely they did a better analysis and maybe even listened to the thing during development! Given that, I wouldn't look at the Zobel so much as a Zobel, but just as additional response modifying components. It's certainly worth a try and might well do what you want. Or, you could have a look at the crossover and modify the values a bit. I wouldn't worry much about the crossover frequency, as it's somewhat of a loosely defined thing to begin with. FWIW, I think any decent speaker has had a certain amount of crossover tuning done by ear during final development. I'd rather have a speaker that sounds good, rather than one that tightly adheres to formulas.
 
ConradH said:
A Zobel is used to flatten the electrical impedance of a speaker, countering the natural rise due to inductance. If they designed the original crossover for a fixed impedance, the Zobel would help. I doubt they did, more likely they did a better analysis and maybe even listened to the thing during development! Given that, I wouldn't look at the Zobel so much as a Zobel, but just as additional response modifying components. It's certainly worth a try and might well do what you want. Or, you could have a look at the crossover and modify the values a bit. I wouldn't worry much about the crossover frequency, as it's somewhat of a loosely defined thing to begin with. FWIW, I think any decent speaker has had a certain amount of crossover tuning done by ear during final development. I'd rather have a speaker that sounds good, rather than one that tightly adheres to formulas.
Click to expand...

I got to this point because I had read that as the impedance curve rises it lets the crossover leak some high frequencys to the driver. This got me curious to add the zobel to the mid and see what happens. I was only worried as to if this would cause any shift of the overall crossover point (3800hz).

I've already ordered the recommended resistor and cap. I also ordered some extra resistors in several different values so I can tone down the mid a bit if I don't get what I'm hoping for with the zobel. I'm avoiding adding a L-pad because I don't want do deal with oxidation issues in the future.

James
 
highend64 said:
Well, I was thinking of a fix L-pad at a specific not requiring a variable one.
Click to expand...

Ah, sorry, I misunderstood. That's most likely where I'll end up. I'll try the zobel first and give it a month or so of listening before making an evaluation.

James
 
Long ago when I was active in speaker building, I'd often use an L-pad to zero in on the levels I wanted, then replace it with fixed resistors. IMO, the less adjustable stuff in circuits, the less to go wrong in the future. I even like soldered-in bias resistors in amplifiers!
 
ConradH said:
Long ago when I was active in speaker building, I'd often use an L-pad to zero in on the levels I wanted, then replace it with fixed resistors. IMO, the less adjustable stuff in circuits, the less to go wrong in the future. I even like soldered-in bias resistors in amplifiers!
Click to expand...

Someday when I'm no longer capable of using them these speakers will go to my son. I don't want to leave him some oxidized pots, and they always get that way.

James
 
I think you misunderstood what Conrad is saying. For a time saving add a adjustable L-pad and adjust it to the level that sounds right. Then measure the resistance across the L-pad and replace it with a resistor of the value you measured the off the L-pad.
 
Silverman said:
I think you misunderstood what Conrad is saying. For a time saving add a adjustable L-pad and adjust it to the level that sounds right. Then measure the resistance across the L-pad and replace it with a resistor of the value you measured the off the L-pad.
Click to expand...

:yes: Yes, I understood. But thank you, I understand now how my answer looked and that you are helping to clarify things.

It's getting late here and it's time for me to stop trying to think. But I don't know where I'm going to add a pot that's easy to adjust without penetrating the box, which I don't want to do. I don't really want to make a new hole in the box. If I have to pull a woofer each time that I want to make an adjustment it might be just as easy to swap out a resistor.

My parts won't be here till next week and there's time to do some more thinking and discussion before I do anything.

James
 
Well, this seems to have gone somewhat sideways. But I'm not as tired as the other day and perhaps tonight I can write more coherently, and hopefully I won't create so much confusion.

For the sake of simplicity lets say that:
1) I understand what a zobel is supposed to do and (I think) why one might want to use one, but not the theory behind how it works it's magic.
2) I understand that a resistor or pot can be used to throttle a driver that is too loud in relation to the other drivers in the system.
3) I understand that the people that built my speakers 30 years ago probably had some knowledge of what they are doing, but that perhaps they didn't take into account that 3 of the 4 drivers might have been changed out later.
4) I don't want to cut any extra holes in my boxes.
5) I'm still interested in finding the answer to my original question but perhaps I didn't define it clearly enough.

Here's the question:

If a person adds a zobel to a crossover that wasn't designed from the start to have one, or adds one to a portion of a crossover that didn't originally have one in that section of the crossover, my concern/question is, what, if any, detrimental effect would that have on the crossover point. Would it be shifted? Up or down?

Thanks,
James
 
What you will notice with the Zobel, if you could connect/disconnect it, is a reduction in highs from the applied driver while connected. It is not so much an adjustment of the crossover point as it is allowing the driver to reproduce only what it's designed to do.

Zobels are ideal for woofers where we don't want them to reproduce high frequencies. Mid ranges, however, are not so clearly cut and dried as they produce a broader range of freq's.

Any time you introduce a capacitor in parallel to a driver there is a small shift in phase. The resistor minimizes this, but it still exists. I've found this to change imaging to some degree. Sometimes good, sometimes not.

One way to find out. If you don't like it, pitch it. Be brave.....Spooker
 
Just changing the level alone will change the crossover point. If one is flattening the upper cutoff for a midrange, it seems like the crossover point would move down in frequency. I could be wrong, but if you just draw the response curves roughly on a piece of paper, then pull the mid down, the intersection with the next curve will move down. IMO, it's a red herring unless you're making really major changes.

What I really want to know is, how are you going to choose the values for a Zobel that matches the driver?
 
Last edited:
Spooker said:
What you will notice with the Zobel, if you could connect/disconnect it, is a reduction in highs from the applied driver while connected. It is not so much an adjustment of the crossover point as it is allowing the driver to reproduce only what it's designed to do.

Zobels are ideal for woofers where we don't want them to reproduce high frequencies. Mid ranges, however, are not so clearly cut and dried as they produce a broader range of freq's.

Any time you introduce a capacitor in parallel to a driver there is a small shift in phase. The resistor minimizes this, but it still exists. I've found this to change imaging to some degree. Sometimes good, sometimes not.

One way to find out. If you don't like it, pitch it. Be brave.....Spooker
Click to expand...

Thanks Spooker, this exactly the sort of info that I'm looking for, and also the effect that I'm hoping for.

A little more clarity: My (ears only) feeling is that while I'm very happy with my new mids in general, they seem to be getting almost a little shrill. I believe (again, ears only, I haven't bought a mic for testing yet) that they are a little louder than they should be. But before starting to throttle them I was/am interested in hearing if there is any benefit to be had with the zobel.

And I will also be buying a new 10" woofer in a few months and only want to take things one step at a time. I'm hoping that the new woof will be a better match for the system than the old new woofer that I bought 14 years ago, and don't want to do any padding before I can evaluate the new woofer. ( a Scan-Speak 25W/8565-00) It is very likely that later I will need to do as Conrad and others have suggested and pad the mids, but again, one step at a time, and I have a curiosity itch that needs scratching.

James
 
ConradH said:
Just changing the level alone will change the crossover point. If one is flattening the upper cutoff for a midrange, it seems like the crossover point would move down in frequency. I could be wrong, but if you just draw the response curves roughly on a piece of paper, then pull the mid down, the intersection with the next curve will move down. IMO, it's a red herring unless you're making really major changes.

What I really want to know is, how are you going to choose the values for a Zobel that matches the driver?
Click to expand...

Thanks very much Conrad. I was very tired when I answered your last answer and was way too curt. No disrespect intended, my bad.

This is very helpful and exactly the sort of info that I'm looking for. The answer to your Q is, the new mid is a Wavecor WF160WA01, and buried in their webpage is a chart listing the recommended cap and resistor for a Zobel, a 22uF cap and 3.9ohm resistor. At this point I've ordered some 10 watt 3.9 ohm MOX resistors and Mundorf NPE caps with the intention of installing them and evaluating the changes (if any) for a month or 2 before doing anything else.

I do intend to learn how to measure to determine how to select zobel components because I want to check if the existing zobel on the woofers is still valid, because my existing 10" woofer is non-original and I will be replacing it this summer when I have a bit more cash saved up. The only original driver in the entire box is the 12" woofer.

So far I've located a nice frequency generator at work, a Wavetek FG-5000. It has a digital freq meter and built in amp, but the amp only outputs 500mA, and if I understand correctly I need to test at 1v? We also have a Fluke Scopemeter that I can use, and I bought one of those little Lepai LP-2020A amps also if I need to test at a higher output than the FG-5000 supplies.

I just have to learn how to do the tests and am getting very confused, partly because everything that I've been reading is geared towards single woofers or 2 identical woofers in parallel, but my speakers are asymmetrical, having a 12" and a 10" woofer in parallel. I'm not understanding how to apply what I've been reading to my setup.

Lastly (sorry about being long winded), if you read my above answer to Spooker, I did pay attention to your (and others) advice about padding, even if I didn't acknowledge it very well. After I get the new 10"ers installed the next step is to buy a mic and do real spl testing, and hope to be able to afford to buy Dayton's DATS for testing also, I've found a place where I can order it for €99. One step at a time is all that I can afford, and seems best anyway.

James
 
Spooker said:
Any time you introduce a capacitor in parallel to a driver there is a small shift in phase. The resistor minimizes this, but it still exists. I've found this to change imaging to some degree. Sometimes good, sometimes not.
Click to expand...

How would I go about testing for phase? Do I understand correctly that the 2nd order crossover is already introducing some phase change? I know that the tweeters are installed with reverse phase, and have read that there is 180° phase change present, but have not understood why only the tweeter is inverted. Are the frequencies involved in the mid and woofer area too low for the phase changes to be noticeable?

And if there is an existing phase change between the woofer and mid then am I correct in understanding that you mean that the phase change added with the zobel can either add to or subtract from that, hence what you say, "Sometimes good, sometimes not"? How does one check this? My curiosity is deepening, I only hope that my brain can keep up.

James
 
I'm not a technician but here goes. If a woofer is wired first-order (no phase shift) and the tweeter is 2nd-order ( 180 degree shift) then the driver's surfaces (cones, domes, etc.) are moving in opposite directions when current is applied. One needs to be reversed, normally the tweeter, to provide that they are "pumping" in the same direction.

I primarily listen to open baffle speakers where there is a complex interaction of front and back waves radiating around the baffle or mounting board. Changes I noted are purely perceived by my 56-year-old ears and I don't have testing equipment to verify such. My perceptions are founded in published articles regarding phase shifts.

I wouldn't be so concerned about testing equipment as I would be about using my ears. I take a particular track, like Jim Keltner's Sheffield drum solo, become familiar with the sound stage and imaging created, then introduce a modification and listen for what happens.

There are those in this forum with much greater knowledge and experience than I, who perhaps will chime in. Hope this helps....Spooker
 
Spooker said:
I'm not a technician but here goes. If a woofer is wired first-order (no phase shift) and the tweeter is 2nd-order ( 180 degree shift) then the driver's surfaces (cones, domes, etc.) are moving in opposite directions when current is applied. One needs to be reversed, normally the tweeter, to provide that they are "pumping" in the same direction.

I primarily listen to open baffle speakers where there is a complex interaction of front and back waves radiating around the baffle or mounting board. Changes I noted are purely perceived by my 56-year-old ears and I don't have testing equipment to verify such. My perceptions are founded in published articles regarding phase shifts.

I wouldn't be so concerned about testing equipment as I would be about using my ears. I take a particular track, like Jim Keltner's Sheffield drum solo, become familiar with the sound stage and imaging created, then introduce a modification and listen for what happens.

There are those in this forum with much greater knowledge and experience than I, who perhaps will chime in. Hope this helps....Spooker
Click to expand...

The crossovers are 2nd order throughout, with the mid and high sections in series, and the woofer section in parallel with the rest.

My ears are a year older than yours, but as you say they are still useful (except when my wife starts talking about that new kitchen that she wants). And I've been listening to these same speakers for 30 years now, so I guess that my ears are somewhat in tune to them. Like you I have a list of songs that I use to test with, Chicago's version "I'm A Man" is one of my favirites to use, it has such a nice mix of bass, horns, and guitar. The desire to learn to do testing is as much for the fun of learning as anything else.

Also my listening environment is a difficult one. My living room is really too small, and the brick walls and laminated wood floor doesn't help any. So getting a mic and paying attention to what other AKers are sharing about room adjustment is becoming more and more important to me. One step at a time. :music:

Thanks for the info/help,
James
 
No need for a zobel network

James,

A Zobel network is designed to create a constant impedance (primarily resistive) load in a reactive network. Your woofers, with their large motors represent a highly inductive load - so the Zobel network was added to make the woofer appear as though it is a resistive load - and not reactive (inductive).

From the top; voltage and current in ANY electrical system are not always "in phase". This means that you can have high voltage and no current or high current and no voltage. This difference between the voltage and current phase is referred to as "power factor". These two values, voltage and current become out of phase when the electrical flow encounters reactive elements (capacitors and inductors). Your crossover network is by definition highly reactive.

To make the situation even more difficult the speaker motors themselves (big windings) are also reactive (primarily inductive). So given the large motor on your woofers the Zobel was added to adjust the power factor of the incoming audio signal so that whatever power is being pushed to your speaker has a unity power factor (voltage and current completely in phase). Thus making your crossover and motor network more efficient.

Therefore adding a Zobel to the mid-range elements of your crossover will NOT make the mid-range "less" pronounced - if anything the mid-range will get LOUDER - for the power factor should be, if you design your Zobel correctly, enhanced and therefore pass a more powerful signal to that driver.

Do not add a Zobel to your network. That will not serve your purpose (which seems to be to attenuate the mid-range signal). My suggestion is that the only way you can lower your mid-range signal is to add a "real", i.e. a resistor of some value (I would go higher than 10 watt capability by the way - if nothing else parallel the two 10 watt resistors). Look for around one or two ohms to start your experiment.

Or you can be cute - as I have done myself - and add a lower than sufficient resistor thereby creating a volume dependent load, i.e. the resistor will heat as you increase the signal and the resistance will therefore INCREASE - so you will load the signal more heavily at higher volumes. Works quite well with the one caveat that you may burn out the resistor at a ridiculously high volume (ceramic resistors simply crack when heated too much - and the resistance then goes to infinity or "open circuit").

Let me know if this helps or you need more advice.
 
There's lots of info on single and dual driver systems, but asymmetric drivers make my brain hurt. No idea on how they interact or how to calculate for them. A single driver system with an asymmetric passive radiator is common and presents no problems, but I'm thinking asymmetric sized drivers probably need their own separate chambers to work properly, and even then the crossover design might be complicated or not realizable due to the electrical interactions.

I've tested t&s parameters of lots of drivers and it's pretty easy. Common signal generators don't have the output capability, so you need any audio power amp between the generator and driver. The FG-5000 is uncommon here, but it appears to have the standard 50 ohm output. Just hook it to any amp or receiver. You can figure out Vas using either a quick test enclosure, or with the added mass method. Both work, but I prefer the enclosure method.
 
You must log in or register to reply here.
Back
Top Bottom