Speaker Measurements | audiobymartin

This section will be tricky.

I've long tinkered about putting my measurements online, as I actually do measure quite a lot.

So much can go wrong with measurements (room/floor reflection) and what SPL to measure at?

SPL is huge factor for distortion. A subwoofer doesn’t distort very much at 70 dB output when the microphone is placed 3” away. The measurement at 70 dB may look good, but is actually quite useless, as it doesn't reflect real life.

Should measurements be taken in a speaker box or on an infinite baffle?
What about a baffle step compensation to mess frequency response, if a narrow speaker box is to be used?
Should it be taken with everything mounted and beveled in?
What level of smoothing should be used? (to hide flaws...)
Nearfield or Farfield?
Gathed or Natural?

I often conduct measurements at 2” – 3” for my personal use, and have a lot of data to support that way.

Close measurements are useful for certain aspects, when apple to apple comparisons are made, but is otherwise useless for others.

In my system, I tend to measure the "whole system" and I want to incorporate the room artifacts with it. This is especially true in the sub 200 Hz territory.

If you want to have good sound at the listening position, you have to measure in the listening position and correct what needs to be corrected in order of priority.

If it means the room ceiling reflection is the biggest issue to be corrected first, so be it. Time measurement can easily identify this.

Sometimes measuring at 1 meter makes a lot of sense, and is a good compromise, specifically for multi-way speaker time alignments.

For horns, 2 meters is more preferable.

In the drop-down tabs of this section, you’ll see a lot of various measurements.

Not all of them are made the same way, and should be taken with a grain of salt.

I’ll do my best to attach pictures of the measurement set-up, and describe how my measurements were performed.

Keep this in mind if you’re going to compare my measurements to other people’s data.

Use my findings as you see fit. Feel free to post them, and take notes of the results on specific measurement conditions and possible use of equalization as some measurement contain DSP correction built-in.

I'll also do my best to only post measurements of good quality, top tier product lines. Does anyone really care about the results of a $12 tweeter Part Express sells?

I certainly don’t.