Measurement Approach Questions

[aps]

I’ve got a two-way active (DIY) speaker with three subwoofers and have built filters for each driver in Audiolense based on in-room measurements at the main listening position. This approach works well, but I have a few nagging questions about the overall approach.

• Question 1: Is there a benefit of using PEQ to linearise near-field subwoofer performance before running the in-room measurements at the main listening position? (One of subwoofers needs a bit of EQ to go down to 20Hz.)
• Question 2: What is the rationale, more broadly, for doing all measurements in-room at the main listening position rather than, for example, first using near-field measurements to linearize the drivers before doing measurements at the main listening position?

 

[Ofer]

As for Q1. I have a sub that needs a bit of EQ to go down to 20hz. It was achieved by creating a target that bumps the area ,20-25hz in my case, by 2-3db and it works. The port is at 18hz and it was suggested (Marty Cube sub) to do a LPF at 18hz which was also done with the target.
BTW are you still using MiniDSP as your digital pre-amp?

 

[aps]

As for Q1. I have a sub that needs a bit of EQ to go down to 20hz. It was achieved by creating a target that bumps the area ,20-25hz in my case, by 2-3db and it works. The port is at 18hz and it was suggested (Marty Cube sub) to do a LPF at 18hz which was also done with the target.
BTW are you still using MiniDSP as your digital pre-amp?

Yes, I've been using DDRC22D but am moving to mini-DSP SHD Studio for the in-built Roon.

 

[juicehifi]

1 usually no, but perhaps if you have a very special sub that needs excessive boost in the bottom, and can handle it. An open baffel with a bunch of 15 inch plus drivers. But even this may be manageable in Audiolense.

2 The rationale has several facets.

A) The sweet spot is where it matters. The aim is to correct the blend of direct and reflected sound that is perceived as direct sound by the listener, who obviously sits in the sweet spot.

B) What you do in a near-field correction will be almost completely undone by the “final” sweet spot correction. An appropriate sweet spot correction will literally cover the near field correction in every meaning of the word. So what the near field correction basically does is modify the measurement sweep. Less basically the near field part may be dysfunctional, which brings me to… C) A nearfield per driver correction including crossovers can and will often be technically inferior to doing it in the sweet spot, especially at frequencies in the hundreds & in the Schroeder region, where reflections sometimes makes a huge difference. Bass and midrange will be somewhat out of phase in the sweet spot when they are in-phase close up. And in some extreme situations they will be in opposite phase at the crossover. So the sonic impact of reflections in the sweet spot should ideally be accounted for when the crossovers are chosen and when the drivers are aligned.

This is btw an old discussion where I stuck my neck out from the getgo almost 20 years ago. I have never been served a good reason for doing the nearfield thing. And I have probably off a number of “experts” by asking for one. The belief was strong but a rational reason was always missing.

I can think of two reasons, where only one is rational. The rational is if you have inadequate tools for correcting in the sweet spot… inadequate messurement, interpretation of measurement and correction … all of them are more complex in the sweet than nearfield. The other reason is “… but this is how speakers have always been made!”