There are tons of things to watch out when making acoustical speaker measurements, but one of the most overlooked ones is a holder influence. I can’t even count how many times I saw people using standard mic holders for measurement microphones. Then they post pictures and their measurement results and don’t realize that they are corrupted. When someone points out their mistake,  they don’t understand how can that be of any significance. Can we put an end to that?

Let’s explore an example with Dayton EMM-6 measurement microphone and it’s holder, that is included in a kit.

I should first point out that something like this might be perfectly acceptable for low-frequency room mode measurements. There we are measuring all the reflections from all of the room surfaces. So these few additional ones from our mic stand and holder are of a little significance to the final result. But what happens when we are making full-bandwidth gated measurement of a tweeter?

Let’s set up an experiment as shown in a picture above. We will make 3 separate measurements. Only variable between them will be microphone membrane distance from the holder.

As you can see, the most devastating influence can be observed at the highest octave. But it’s not limited there, as all of the 1-20kHz frequency range is effected a little. We can make a couple of conclusions from this data:

• Minimum distance of 30cm is needed to get rid of the most nasty looking measuring artifacts.
• The artifacts seems to disappear at the distance of 50cm and above.

This is well in accordance with intuitive expectations. Look at the set up picture and imagine how a sound wave leaves a tweeter. Then, after 1 meter or some 2.8ms of time, hits the microphone membrane. It travels further 50cm to the holder. Hits it and resulting reflection travels back same 50cm to the membrane. It takes again 2.8ms to go 50cm and back. So if we are gating our measurement at lower then 2.8ms – all effects of these reflections are mitigated.

The solution

For the most measurements posted here I was using a simple PVC pipe. I have fitted XLR connector to one end with heat-shrunk tube and called it a day. Heck, I’m using it even today when making critical new driver measurements on a baffle.

It’s 50 cm long and together with my WM61A mic, sets membrane some 70cm from a holder. There are a couple of drawbacks though. It’s difficult to handle and it doesn’t look very professional. So on a rare occasions I’m visiting someone and measuring their stuff – this “ghetto” setup doesn’t give me much credibility points.

So I decided to design something more respectable and this is a result above. It sets  EMM-6 mic membrane some 37cm from a stand. Bottom part has a thread and it screws right into standard boom-stand thread. Then it attaches to holder with a 5mm screw and there is a relief for accompanying nut. 

There shouldn’t be any problems importing them in a modern slicer and 3D printing the parts.

Final result

After some 8 hours of printing time and 70 grams of PLA filament, I got a new microphone holder. It’s 21cm high because that’s a maximum I can fit in my Anet A6. 

You can also see a small fixation screw near XLR connector. End of the holder is internally cone shaped so XLR socket already sits pretty tight. Fixation screw just seals the deal. Hole has to be threaded manually as horizontal printing of 3mm threaded holes still belongs to sci-fi (or stereo-lithography) area. At least for regular material jetting printers. Hope it serves you well!