Power response comparison with Linkwitz-Riley crossovers


Introduction

In this section I will be comparing the power response of a 2nd order Linkwitz-Riley crossover with that of a 4th order Linkwitz-Riley crossover. Both crossovers have a crossover frequency of 2kHz. Just in case you aren't familiar with the term power response I'll describe it briefly. Power response is a term that describes a speaker's frequency response when all listening, or measuring, angles are taken into account. To have a flat power response the speaker would have to radiate sound equally in all directions as is the case with a true omnidirectional speaker. This comparison will mainly focus on frequencies from the midrange region and up. This analysis will assume a two-way speaker design with a 1" dome tweeter and a 6.5" mid-bass driver.

Analysis

The plot above shows the frequency and power response of a 2nd order Linkwitz-Riley crossover. If you place your mouse over the plot you'll see the response for a 4th order Linkwitz-Riley crossover. You'll notice that both designs exhibit a rolloff in the upper frequencies which is due to the poor off-axis response common to most 1" dome tweeters. Some of the dip in the power response above the crossover region is due to the directionality of the mid-bass driver. When you compare the two power responses you'll notice they have the same amount of dip at the crossover frequency but above and below that point the 4th order crossover has a stronger power response. This is due to the different vertical lobing patterns for these two crossover designs and also due to the fact that the 4th order Linkwitz-Riley crossover has a narrower crossover region.

The plots only say so much, what really matters is how they sound relative to each other. To find out how they compare sonically I designed two different crossovers (an LR2 and an LR4) for my Galatea speaker design. I ensured that they had the same optimal listening height on-axis with the tweeter and that the frequency responses at this axis were nearly identical within 0.5 dB across the entire frequency range. I used a relay box to instantaneously switch between the two crossover designs. What I discovered was that the 4th order design sounded stronger in the region just above the 2kHz crossover point near 4kHz. This correlates directly with the relative peak in the power response for that design in that region. I probably also heard some of the increased output in the region just below the crossover point but it wasn't as distinguishable. The differences in the sounds of the two crossovers was subtle but easily distinguishable. What surprised me about this analysis is that I didn't notice any difference in the soundstage or imaging characteristics between the two designs. I was previously under the impression that lower order crossovers provided a more natural soundstage and more pinpoint imaging but this didn't seem to be the case. So in conclusion I would say that the only differences that I noticed between the 2nd order and 4th order Linkwitz-Riley crossovers was in the power response where the 4th order design was slightly stronger above and below the crossover frequency.