Short answer: a calibrated measurement mic, the free Room EQ Wizard (REW) software and a laptop will measure a speaker well enough to design a crossover, if you use gating and near-field tricks to get the room out of the way. A USB mic is the easiest start. An XLR mic with a two-channel audio interface is better for crossover work, because it gives you an accurate timing reference. Add an impedance tester and you've got a proper home lab.
What you need
| Gear | What it's for | Notes |
|---|---|---|
| REW (free) | Measurement software: frequency response, impedance, distortion, room analysis | roomeqwizard.com. Windows, Mac, Linux. |
| Calibrated measurement mic | Hears the speaker accurately | Must come with its own calibration file. |
| Audio interface (XLR mic route) | Gets the mic into the laptop, provides phantom power, gives a loopback timing reference | Any two-input USB interface with phantom power will do. |
| Amplifier | Drives the speaker | Whatever you've got. Keep the volume sensible. |
| Mic stand | Holds the mic still at the right height | A cheap boom stand is fine. |
| Impedance tester (optional but handy) | Impedance curves and T/S parameters | Needed for proper crossover design and checking drivers. |
From our measurement range: - USB mic: Dayton Audio UMM-6, plug it straight into a laptop. - XLR mic: Dayton Audio EMM-6, for use with an interface. - Phone or tablet mic: Dayton Audio iMM-6C (USB-C), handy for quick room checks with an app. - Impedance and T/S: Dayton Audio DATS V3.
USB mic or XLR mic plus interface?
USB mic: easiest. Plug in, load the calibration file, measure. Great for room measurements, sub EQ and checking a finished speaker.
The catch is timing. With a USB mic, the mic and the playback device run on separate clocks, so REW can't measure absolute timing directly. For crossover design you want to know the timing relationship between the drivers (their relative phase). REW has an acoustic timing reference feature that works around this, and it works, but it's another thing to get right.
XLR mic plus a two-channel interface: best for crossovers. One input takes the mic. The other input gets a loopback cable straight from the output. Now REW knows exactly when the signal left, so the timing (and the phase relationship between the drivers) is measured properly.
If you're only going to measure rooms and subs, go USB. If you're designing crossovers, go XLR.
Getting a clean measurement in a normal room
A room adds reflections from the floor, walls and ceiling. They make a measurement look like a mountain range. Three techniques deal with them.
1. Gating (for the mids and highs)
The direct sound from the speaker reaches the mic before any reflection. If you only look at the first few milliseconds of the impulse response (the "gate"), you see the speaker without the room.
The trade-off: a short gate can't see low frequencies. Roughly 1 ÷ gate time gives the frequency resolution: about 200 Hz for a 5 ms gate. That's a rough scale, not a promise that the response is accurate right down to 200 Hz, so don't read fine detail near it.
To get a longer gate, get the speaker and mic away from surfaces: up on a stand, in the middle of the biggest room you've got. Outdoors on a still day is even better.
2. Near-field (for the bass)
For the woofer below a few hundred hertz, put the mic almost touching the dust cap, a few millimetres away. At that distance the speaker is so much louder than the room that reflections don't matter. Measure the port the same way, right at its mouth. Before adding the port to the woofer, scale it for its area compared with the cone's (REW and VituixCAD can do this).
3. Merge them
Splice the near-field bass onto the gated far-field measurement. REW and VituixCAD can both merge measurements. VituixCAD's diffraction tool can simulate the baffle step that a near-field measurement misses.
A basic measurement session for crossover design
- Put the speaker on a stand, as far from walls as you can.
- Set the mic at the height you'll design for (usually the tweeter axis, or between the drivers), about 1 m away. Keep it there for every far-field measurement. Don't move the speaker or the mic between drivers. That's the common-axis method. VituixCAD's documented method measures each driver on its own axis instead. Either works: pick one, follow it from the first measurement to the finished crossover, and don't mix the two sets of files.
- Measure each driver on its own, at a safe level. Protect the tweeter with a capacitor in series while testing, so it doesn't get bass. The capacitor shows up in the measured response and phase, so note that it was in circuit.
- Measure the woofer and port near-field.
- Measure impedance for each driver, in the box.
- If you can, measure off-axis too (every 15° out to 60° or 90°). It shows how the speaker sounds across the room, not just straight ahead.
- Load the lot into your crossover software.
Passive crossovers explained covers what happens next.
Measuring a room and a sub
For room EQ and subwoofer setup:
- Measure at the listening position, then at a few spots around it. Averaging smooths out the results of moving your head 30 cm.
- Low-frequency room modes show up as big peaks and nulls. EQ can tame peaks well. Nulls can't be fixed with EQ (boosting into a null just wastes power), so move the sub or the seat instead.
- Two subs in different spots usually give a smoother response across more seats than one.
Common traps
- No calibration file loaded. Every mic needs its own file (matched to its serial number). Without it, the top end will be off.
- Too loud. Sweeps are hard on tweeters. Start quiet and creep up. Use a protection cap on the tweeter when it's measured on its own.
- Moving things between measurements. If the mic or speaker moves between the woofer and tweeter measurements, the timing relationship is wrong and the crossover sim won't match reality.
- Measuring the room, not the speaker. Forgetting to gate makes a smooth speaker look terrible.
- Wiring a driver backwards. Check polarity before trusting a phase measurement.
- Laptop power saving or Bluetooth audio. Use a wired output and switch off sound "enhancements".
When you don't need any of this
Building a proven kit or a published design? You don't need to measure anything. The designer already did. A mic is still handy afterwards for placing the speakers and setting up a sub, but it's optional. If you're designing your own, measuring isn't optional: it's the difference between guessing and knowing.
Quick questions
Can I measure with my phone's built-in mic? For a rough room check, sort of. Phone mics have processing and an unknown response. A calibrated mic is much more trustworthy.
What's the difference between an impedance tester and a measurement mic? The mic measures what you hear. The impedance tester measures the electrical load the speaker presents to the amp, which crossover design also needs. You want both for crossover work.
How far away should the mic be? For a bookshelf two-way, about 1 m is a common starting point, far enough that both drivers blend. Bigger speakers need more distance.
Is REW really free? Yes, the standard version is free to download and use. There's a paid Pro upgrade for some advanced features, but you don't need it for anything in this guide.