The drivers for my future enclosures will be ordered this week, and it is recommended that I should order the parts for the crossovers also.
My last questions are.
1. How to choose the crossover frequency? The woofer has a nasty bump in the SPL response at approx 4.5KHz, while the resonance of the tweeter is at 1.8KHz. I have 2 options: a low XO freq (I'm thinking 2.5), in order to get far enough from that bump, but then I get close to the tweeter resonance. Or... high XO freq (3.5), to get far from the tweeter resonance, but then I'm getting closer to the woofer bump. Which version is preferrable? I'm thinking that the first one, if I use impedance EQ and/or a high damping factor amp. Am I right? Btw, the amp will be a Quad 707.
2. I saw this phrase in different occasions "incorporating the impedance correction in the XO itself is possible". What exactly does that mean?
3. Can I compensate for the lack of impedance correction with a good (high damping factor) amp?
My last questions are.
1. How to choose the crossover frequency? The woofer has a nasty bump in the SPL response at approx 4.5KHz, while the resonance of the tweeter is at 1.8KHz. I have 2 options: a low XO freq (I'm thinking 2.5), in order to get far enough from that bump, but then I get close to the tweeter resonance. Or... high XO freq (3.5), to get far from the tweeter resonance, but then I'm getting closer to the woofer bump. Which version is preferrable? I'm thinking that the first one, if I use impedance EQ and/or a high damping factor amp. Am I right? Btw, the amp will be a Quad 707.
2. I saw this phrase in different occasions "incorporating the impedance correction in the XO itself is possible". What exactly does that mean?
3. Can I compensate for the lack of impedance correction with a good (high damping factor) amp?
Choose your crossover frequency based upon the response of the woofer, staying below any bumps and also staying below the frequency where the 30 degree off-axis response dips more than 6dB below the on-axis. Then minimize the reponse bump and the bandwidth overlap by using at least 3rd order slopes, which also better protects the tweeters. With a 3rd order slope you can go to 1.5 x Fs of the tweeter; 1.3 with 4th order and 1.1 with 5th order. State of the art is 3rd order LP/5th order HP for maximum performance/minimum cost/lowest Fc.
In theory you should not calculate the crossover component values in advance as they are best arrived at after measuring the in box response of the drivers, but if you lack the equipment and experience required that's not possible. Impedance correction is over prescribed and shouldn't be done without prior testing to make sure that something's broken before you try to fix it.
In theory you should not calculate the crossover component values in advance as they are best arrived at after measuring the in box response of the drivers, but if you lack the equipment and experience required that's not possible. Impedance correction is over prescribed and shouldn't be done without prior testing to make sure that something's broken before you try to fix it.
BillFitzmaurice said:
Not to thread hijack, but in your opinion what is the best order for a cross over? If you can stay far enough away from the Fs, is lower order better? I see many cross overs that are 3rd order now, it seems to be preferred. In the question asked, disregarding the woofer resonance at 4,500 Hz, would a 2nd order x-over at (say) 3,600 Hz (2 x Fs) be preferred over the 3rd order at 2,700 (1.5 x Fs) Hz? Or something different?
For the impedance correction, what would you consider 'broken'? For many of the metal/ceramic woofers, there is a nasty resonance in the 3,000-4,000 Hz range. Would you advise lowering the x-over (about 1,500-2,000 Hz?) and avoiding the resonance, or allowing the cross over around 2,500-3,000, and correcting the resonance?
Thanks for your help!
I use what the job requires. Less is better to some extent, but I think that 1st order allows too much bandwidth overlap.
2nd order requires careful attention to phase issues, but is OK if the acoustic rolloff of the drivers is adequate to arrive at a total 3rd order or better.
When power issues arise in mid or HF units 3rd or 5th is best for protection, and allows the use of tweeters down closer to Fs. That can be useful to end up with a two way rather than three way system, or to get around off-axis response issues with larger woofers by crossing lower to the tweeter.
I don't see any need of more than 3rd order LP, and if there's a place I use 1st order it's on LP if the driver acoustic rolloff is adequate; that's what I use on folded horn subs, for instance, as much for impedance issues as anything else. As for peaking problems if they can't be avoided by the choice of corner frequency and slope then bandpass L/C filtering is in order.
2nd order requires careful attention to phase issues, but is OK if the acoustic rolloff of the drivers is adequate to arrive at a total 3rd order or better.
When power issues arise in mid or HF units 3rd or 5th is best for protection, and allows the use of tweeters down closer to Fs. That can be useful to end up with a two way rather than three way system, or to get around off-axis response issues with larger woofers by crossing lower to the tweeter.
I don't see any need of more than 3rd order LP, and if there's a place I use 1st order it's on LP if the driver acoustic rolloff is adequate; that's what I use on folded horn subs, for instance, as much for impedance issues as anything else. As for peaking problems if they can't be avoided by the choice of corner frequency and slope then bandpass L/C filtering is in order.
So, 3rd order seems to be the norm... Hm, I was just about to order the components for 2nd order filters with zobel for the woofer, but I saw this... Oh my God, how many times am I going to change my mind? 
What's the big deal with 3rd order anyway?
And about measuring the drivers before designing the xover... It's not that I don't have the "experience" to do that, after all it doesn't require a degree in Electrical Engineeering, but I don't afford a good mic for SPL measurements. A $40 Behringer mic is not that expensive, but now I'm using all my financial resources for the drivers. I would really like good tweeters, that are around $50 each... Some of my friends would thing that I went nuts if they knew that I am going to pay that for those... donut-sized thingies...
But since many people have assured me that the published specs for the drivers I'm buying are close to the real ones, I kinda trust those specs.
What's the big deal with 3rd order anyway?
And about measuring the drivers before designing the xover... It's not that I don't have the "experience" to do that, after all it doesn't require a degree in Electrical Engineeering, but I don't afford a good mic for SPL measurements. A $40 Behringer mic is not that expensive, but now I'm using all my financial resources for the drivers. I would really like good tweeters, that are around $50 each... Some of my friends would thing that I went nuts if they knew that I am going to pay that for those... donut-sized thingies...
But since many people have assured me that the published specs for the drivers I'm buying are close to the real ones, I kinda trust those specs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.