DiyMobileAudio.com Car Stereo Forum banner

61 - 80 of 96 Posts

·
Registered
Joined
·
86 Posts
There aren't many tweeters I would suggest using a 6 db network on unless you're crossing over VERY high. And then, for a 2-way, you're going to have serious problems.

5.6khz is too low. That can be dangerous for a tweeter is you push it hard.

Because Dynaudio does it, doesn't mean that everybody can do it. The Dynaudio tweeter is a very good tweeter.
What's your opinion of the TBI Sound tweeter that comes with a "6.8 UF FILM CAPACITOR (3.5KHZ @4 ohms - 6db/oct)" (from the website: http://www.tbisound.com/dsp_products_auto_tweet1.asp)?

If you consider 5.6 khz too low, then 3.5 khz seems extremely dangerous, no? Would that not allow alot of "low" (i.e., tweeter-destroying frequencies) in? Or is this a robust tweeter as well...seems sketchy...

But it would be a nice and simple tweeter to add to a driver with natural roll-off (mach5 mli-6 or the seas model???). Is there something I'm missing?

Thanks again for a very informative thread...but it leaves me with more questions. Such as lobing (something Zaph talks about when selecting drivers and optimum crossover points) and how driver spacing plays a part in all of this...

chris
 

·
Registered
Joined
·
2,298 Posts
Discussion Starter #65 (Edited)
I have gone into lobing and combing in many other threads to some detail. You'll have to search around. I didn't compile them, and I have no intention of doing another thread like this one. Sorry.

I didn't say you couldn't run a tweeter with a 6db at a lowish crossover point, I just said, it's not a very good idea.

The 6 db crossover is excellent for producing the best power response at the crossover point of any other filter order. However, with that comes the fact that your driver had better be able to handle that power.

I haven't had the luxury of playing with the TBI tweeter, but I'm pretty sure, they don't want you running 100 watts to that tweeter with only a 6 db filter at 3.5khz. Also, a 6.8 uF capacitor yields a crossover point 5850 hz "electrical", NOT 3.5khz. Acoustically, I don't know. However, I don't know where they came up with 3.5 khz. Even "acoustically" that would be quite a stretch. Most likely, they're compensating for the rising response of the tweeter from Fs, and saying that the 6.8 uF cap "flattens" out the response from 3.5 - 6khz. Still acoustically, the slope would be much higher below 3.5 khz.

They played with some some numbers, which is fine. In the end though, they know better than to give people actual tweeter with inline caps for a 3.5khz, 6 db slope. That's just asking for trouble.
 

·
Registered
Joined
·
27 Posts
I don't generally do Zobels in my networks. Zobels allow you to increase the top end acoustic response of a driver. However, in the case of a 7" mid, using it above 2khz, starts to degrade the off-axis response of the mid, and therefore the overall polar response will suffer for the system as a whole.

So, I like to use the natural roll off of a driver to achieve my acoustical response at a point where I'll get the best polar response, and off-axis response possible.
i had my experiment on zobel network. i agree that zobel is lowering the impedance of speaker.



and this one with zobel



the green and black is just different value of C.

and this is the response of use zobel (black) and no use (green)


use of zobel beside lowering the impedance to Re its also decrease the top end acoustic response of a driver.
 

·
Premium Member
Joined
·
4,235 Posts
Careful, the Zobel doesn't change the impedance of the speaker, it changes the impedance of the amplifier. If the Zobel is calculated precisely, the response of the speaker shuldn't change, but the impedance the amplifier "sees" will change. The Zobel makes it easier to calculate the required crossover components and that's its main purpose. Simply designing the filter to use the rising impedanace as one part of the passive network will result in a simpler circuit that costs less.

The important response is the measured acoustic response from the combination of drivers connected to the crossover and not the electrical network itself.
 

·
Registered
Joined
·
2,298 Posts
Discussion Starter #68
Careful, the Zobel doesn't change the impedance of the speaker, it changes the impedance of the amplifier. If the Zobel is calculated precisely, the response of the speaker shuldn't change, but the impedance the amplifier "sees" will change. The Zobel makes it easier to calculate the required crossover components and that's its main purpose. Simply designing the filter to use the rising impedanace as one part of the passive network will result in a simpler circuit that costs less.

The important response is the measured acoustic response from the combination of drivers connected to the crossover and not the electrical network itself.
Exactly!

If you are resistant to doing complicated calculations yourself, and decide to use a modeling program for your filters, then the Zobel is almost an antiquated network.

When I say "modeling program", I'm not talking about a simple crossover designer where you input a frequency and an impedance then it shoots out some values for capacitors and inductors. I'm talking your Speaker Workshops, SoundEasy, LEAP, etc.

All those programs use the impedance curve, coupled with driver response, to predict the response shape after applying various capacitor and inductor values into your circuit. At that point, it becomes pointless to add a Zobel, as the program will be taking into account the variations in impedance anyway.

Another benefit to going without a Zobel is that it's an easier load on the amplifier. The higher impedance of your mid (in a 2-way generally), will increase the impedance of the system overall at the crossover point, making your amplifier sweat a little less.

However, if you really want extended response from a driver, adding a Zobel is a way to do it. I haven't played with an extended range driver and a Zobel yet, but I would venture to guess that that type of driver could benefit. I could also be totally mistaken.
 

·
Registered
Joined
·
27 Posts
Careful, the Zobel doesn't change the impedance of the speaker, it changes the impedance of the amplifier. If the Zobel is calculated precisely, the response of the speaker shuldn't change, but the impedance the amplifier "sees" will change. The Zobel makes it easier to calculate the required crossover components and that's its main purpose. Simply designing the filter to use the rising impedanace as one part of the passive network will result in a simpler circuit that costs less.

The important response is the measured acoustic response from the combination of drivers connected to the crossover and not the electrical network itself.
Exactly!

If you are resistant to doing complicated calculations yourself, and decide to use a modeling program for your filters, then the Zobel is almost an antiquated network.

When I say "modeling program", I'm not talking about a simple crossover designer where you input a frequency and an impedance then it shoots out some values for capacitors and inductors. I'm talking your Speaker Workshops, SoundEasy, LEAP, etc.

All those programs use the impedance curve, coupled with driver response, to predict the response shape after applying various capacitor and inductor values into your circuit. At that point, it becomes pointless to add a Zobel, as the program will be taking into account the variations in impedance anyway.

Another benefit to going without a Zobel is that it's an easier load on the amplifier. The higher impedance of your mid (in a 2-way generally), will increase the impedance of the system overall at the crossover point, making your amplifier sweat a little less.

However, if you really want extended response from a driver, adding a Zobel is a way to do it. I haven't played with an extended range driver and a Zobel yet, but I would venture to guess that that type of driver could benefit. I could also be totally mistaken.
thanks Andi and mini

but if don't have any simulation software maybe zobel can give an instant solution, just count the le and re and u are done.

the same driver and the target response 12 db at 2700 hz


just 12db butterworth passive xo without zobel



with zobel



zobel, with and without passive xo the higher frequency is always decrease. so i don't think zobel can extend the response.
 

·
Registered
Joined
·
86 Posts
Thanks Mini (and others), I'll have to go over the other thread you posted for crossovers (as well as search for lobing, etc.). I just want to say that these threads help to really dispel some of the myths of crossovers and have helped me to understand that simple passive crossovers are anything but "simple"! Thanks again,
Chris
 

·
Registered
Joined
·
5 Posts
Hello everyone. This is a real help for me, what a great forum! I'm just starting out in car audio but I have a good background in sound, mostly in music related stuff (live p.a. and recording studios).
I hope this is an appropriate post for this thread...

I'm thieving the crossovers from the factory amp in my '89 bmw 525i to use with a 4x40w head unit. (this is just the beginning...)

So with the original setup the speaker negatives are shared (ie FL and BL run together - you can see on the backside of the board)... does this affect things? I'm not sure if I should run separate negatives to the head or tie them together first - does it matter?

I'd be happy to get any other advice too - nice to be here!
 

Attachments

·
Registered
Joined
·
2,298 Posts
Discussion Starter #75
Hello everyone. This is a real help for me, what a great forum! I'm just starting out in car audio but I have a good background in sound, mostly in music related stuff (live p.a. and recording studios).
I hope this is an appropriate post for this thread...

I'm thieving the crossovers from the factory amp in my '89 bmw 525i to use with a 4x40w head unit. (this is just the beginning...)

So with the original setup the speaker negatives are shared (ie FL and BL run together - you can see on the backside of the board)... does this affect things? I'm not sure if I should run separate negatives to the head or tie them together first - does it matter?

I'd be happy to get any other advice too - nice to be here!
I'm not quite following you. Can you be more specific? What is FL, BL, and the other labeling you're using?
 

·
Registered
Joined
·
5 Posts
I'm not quite following you. Can you be more specific? What is FL, BL, and the other labeling you're using?
Sorry... Front Left and Back Left, (I am copying what's on the schematics I have so as not to confuse myself) - I will use RL (Rear Left).
Actually you can't see it on the circuitboard but they are spliced together in the factory wiring.

The factory head unit (which I never had, I bought the car with no h/u) had 4 line outs but only L and R speaker negatives. It was connected to an amp in the trunk that contained crossovers for the (10) speakers, 3way pair in the front and 2way pair in the rear. Each set of speakers (FL, FR, RL, RR) shares a common negative (total 4), so I'm wondering if it affects things (ie ohms & f/b fading) if I use the factory L and R speaker negatives and tie fronts and backs together or if I should connect seperate negatives for each of the 4 sets of speakers straight to my head unit. Sorry if it doesn't make sense, typical noob eh?

edit: these are the old amp connections that I am using.

edit: found out through trial and error that you have to run speaker grounds separately.
(...success! -> factory amp bypass (my way - lots of pics) - BimmerNut.com Forums )
 

Attachments

·
Registered
Joined
·
2,298 Posts
Discussion Starter #79
For those stumbling onto this thread, understand that I no longer support DIYMA, and will not affiliate myself with the administration of this website. So, I have pulled all my pictures from the relevant technical threads that I've generated over the years here.

For those that do wish to discuss the contents of this thread, or any other technical thread that I've generated, or participated in, I do still contribute, and discuss at another venue.
 
61 - 80 of 96 Posts
Top