Options
Crossover/filter type, what do I choose and why?
You may have noticed that we often suggest a “BW� filter type for this, or an LR filter for that. What does that stand for? It’s an abbreviation of the filter type. In the Driverack we have “Butterworth or BW�, “Bessel or BS� and “Linkwitz Riley or LR� as you may have guessed, these were named after the developers.
I guess we should also explain why the low cutoff of the system is called the ‘hi-pass’. Seems kind of counter intuitive doesn’t it? Actually, the answer is quite simple. It’s called a hi-pass filter because it “passes� high frequencies well that lie above the cutoff (crossover) point. Conversely then a ‘low-pass’ filter passes low frequencies below the crossover frequency well and attenuates them above the cutoff.
With a BW filter the response also falls off at a steeper rate (over the LR type filter… even over a 24 db/octave versus an BW18 filter…) therefore carrying the response flatter, longer, making it a good candidate for a high-pass filter for a sub where frequencies that are too low could damage the speakers. In general terms though, as the number increases, the complexity, and the steepness of the crossover point increase as well. When selecting a Butterworth 24 filter the 3db down point, is the ½ power point at the tuning frequency. In the LR filter it’s the 6db down point at the tuning frequency. When we crossover to an “inter-cabinetâ€? component …( like a low mid (15â€?, or dual 15â€?to a 12â€? or 10â€? or multiple cones, or a horn with the potential to go down to 1000-1500hz) staying with a similar crossover type on both sides of the crossover smooths the response of the crossover response over using dissimilar types.
Each crossover type in a digital loudspeaker management device has its own characteristics, trade offs, and drawbacks. Generally, the less steep… the less complex, the least interaction of the crossover, but alas the more interaction of the drivers(comb filtering and interaction creating havoc in the crossover region). We also have a good handle on the delay that the LR 24 crossover presents between lo-pass and hi-pass points. This is the internal delay that all crossover types create, and the time it takes for the signal to pass through the filter. This was measured using SMAART, an FFT (fast fourier transform) measurement system that unlike the RTA system, take a number of factors like time and coherence into consideration, not just frequency based volume as the RTA does.
Your given a large number of crossover types and slopes to work with in the Driverack. The reason is, not all types of speaker systems work best with the BW 18 and LR 24 we specify mostly here. If you look at the overall component mix here we typically have a large woofer (or two.. sometimes with a pseudo two way design that has one of the woofers limited in its response) and a horn and driver for the high frequency component. If we were crossing a small woofer with a phenolic ring dome tweeter in a studio monitor we might be more likely to choose a high order Bessel filter. The thing is, with the Digital crossover like the Driverack we can very easily change crossover type and slope, the point the crossover takes place and the gain of the different components and see what happens, or more accurately HEAR… the difference. So give it a try if you get some time. The differences are not “stunning� but generally subtle, and don’t feel bad if you can’t tell the difference. It often takes a while to train the ear to discern theses differences, and sometimes the system just isn’t capable of recreating the detail necessary to exhibit these differences. Remember also that when tuning a speaker system it’s incredibly important to have all the drivers working together, which means that when a positive going pulse is sent to the speakers than ALL components are moving in the same direction at the same time. With certain crossover types this can mean that phase, delay or even polarity need to be addressed. That however is beyond the scope of this discussion.
As I have been heard to say…� The more you learn, the less you know�…
Gadget
I guess we should also explain why the low cutoff of the system is called the ‘hi-pass’. Seems kind of counter intuitive doesn’t it? Actually, the answer is quite simple. It’s called a hi-pass filter because it “passes� high frequencies well that lie above the cutoff (crossover) point. Conversely then a ‘low-pass’ filter passes low frequencies below the crossover frequency well and attenuates them above the cutoff.
With a BW filter the response also falls off at a steeper rate (over the LR type filter… even over a 24 db/octave versus an BW18 filter…) therefore carrying the response flatter, longer, making it a good candidate for a high-pass filter for a sub where frequencies that are too low could damage the speakers. In general terms though, as the number increases, the complexity, and the steepness of the crossover point increase as well. When selecting a Butterworth 24 filter the 3db down point, is the ½ power point at the tuning frequency. In the LR filter it’s the 6db down point at the tuning frequency. When we crossover to an “inter-cabinetâ€? component …( like a low mid (15â€?, or dual 15â€?to a 12â€? or 10â€? or multiple cones, or a horn with the potential to go down to 1000-1500hz) staying with a similar crossover type on both sides of the crossover smooths the response of the crossover response over using dissimilar types.
Each crossover type in a digital loudspeaker management device has its own characteristics, trade offs, and drawbacks. Generally, the less steep… the less complex, the least interaction of the crossover, but alas the more interaction of the drivers(comb filtering and interaction creating havoc in the crossover region). We also have a good handle on the delay that the LR 24 crossover presents between lo-pass and hi-pass points. This is the internal delay that all crossover types create, and the time it takes for the signal to pass through the filter. This was measured using SMAART, an FFT (fast fourier transform) measurement system that unlike the RTA system, take a number of factors like time and coherence into consideration, not just frequency based volume as the RTA does.
Your given a large number of crossover types and slopes to work with in the Driverack. The reason is, not all types of speaker systems work best with the BW 18 and LR 24 we specify mostly here. If you look at the overall component mix here we typically have a large woofer (or two.. sometimes with a pseudo two way design that has one of the woofers limited in its response) and a horn and driver for the high frequency component. If we were crossing a small woofer with a phenolic ring dome tweeter in a studio monitor we might be more likely to choose a high order Bessel filter. The thing is, with the Digital crossover like the Driverack we can very easily change crossover type and slope, the point the crossover takes place and the gain of the different components and see what happens, or more accurately HEAR… the difference. So give it a try if you get some time. The differences are not “stunning� but generally subtle, and don’t feel bad if you can’t tell the difference. It often takes a while to train the ear to discern theses differences, and sometimes the system just isn’t capable of recreating the detail necessary to exhibit these differences. Remember also that when tuning a speaker system it’s incredibly important to have all the drivers working together, which means that when a positive going pulse is sent to the speakers than ALL components are moving in the same direction at the same time. With certain crossover types this can mean that phase, delay or even polarity need to be addressed. That however is beyond the scope of this discussion.
As I have been heard to say…� The more you learn, the less you know�…
Gadget
0