What do you use to measure sound levels? - Page 2

Quote:

Originally Posted by Ham Sandwich 

This is what I rigged up for my SPL meter:

 

 

The clear disc is a protector disc that you get when you buy a spindle of CD or DVD discs.  The center hole of the CD disc is actually a little larger than the mic so I glued on a rubber washer that fits the mic more snugly (the diameter of the mic is approx 1/2 inch).  It works and the clear disc is nice because you can see in to see where the mic is positioned.

That is identical to my rig--same SPL meter and using a junked CD. I don't allow the mic to stick out like that though, since I think having it flush with the CD is better. If the mic gets too close to the driver it'll skew the SPL reading to be louder than what you hear with headphones on. Remember, your ear canal is actually some distance away from the side of your face. If we rigged a short distance of tunnel with rolled up material to simulate the ear canal, and then attach the meter's mic to the end of it, we might get even closer (without actually buying an expensive dummy head for headphone testing).
 

This is an awesome thread, as I'm looking at doing something like this.  Although, I am looking for something to measure IEM's.  Has anybody constructed a way to do this?

Thanks!

Quote:

Originally Posted by wiz561 

This is an awesome thread, as I'm looking at doing something like this.  Although, I am looking for something to measure IEM's.  Has anybody constructed a way to do this?

 

Thanks!

This was mentioned in one post but otherwise didn't get much attention—have you considered just measuring voltage sent to the IEMs (some kind of Y-splitter, measure at jack with multimeter or similar) and then calculating based on IEM efficiency?  It may be easier than rigging a microphone and SPL-testing rig.

im glad i saw that disc rig, thats exactly what im gonna do when my spl meter arrives (which should be soon i hope).

i also got a scosche one thats only meant to check 60-130 db but i thought it doesnt matter since im only aiming to have it below 80 (or rather 75 just in case)? couldnt find a listing of the frequency range though, i hope its alright... ill check with the scosche site later, if it isnt 20hz-20khz does that mean that after a certain frequency it just wont measure?

if i boost the bass with eq, and i mean like a 10 db boost, will the spl meter pick up on it?

edit: just got it! initial measurements show i was litening to music at about 90 db, which is too loud for safety... unfortunately i cant do the disc thing because the tip of my meter isnt round - its a rather bulky triangle, ill have to improvise a cardboard rig. i noticed though that if i measure the headphones WITHOUT any music playing it immediately jumps to around 74 db... i dont get it, theres no music playing... i think its a bad meter... and the scosche site doesnt have a frequency range spec either... 

Quote:

Originally Posted by mikeaj 

 

This was mentioned in one post but otherwise didn't get much attention—have you considered just measuring voltage sent to the IEMs (some kind of Y-splitter, measure at jack with multimeter or similar) and then calculating based on IEM efficiency?  It may be easier than rigging a microphone and SPL-testing rig.

Thanks for the idea, but I'm not exactly sure that this would give an accurate view of the IEM itself.  I suppose that by doing something like what you're suggesting, it would equalize the equipment.  To make sure I understand you, you're saying to do that, then look at the frequency curve for the IEM, and do a diff to set the equalizer?  Are there any references or howto's out there that explain how to do this better?

Last time I hooked a multimeter up to some equipment, I think I was about 9 or 10 years old.  I forget what it was, but I want to say something started smoking!  Ever since then, I've been a little leery about doing it again.  On the other hand, it was 20 years ago and I didn't know what an ohm was, or the difference between ac and dc!

Thanks!

Quote:

Originally Posted by wiz561 

 

Thanks for the idea, but I'm not exactly sure that this would give an accurate view of the IEM itself.  I suppose that by doing something like what you're suggesting, it would equalize the equipment.  To make sure I understand you, you're saying to do that, then look at the frequency curve for the IEM, and do a diff to set the equalizer?  Are there any references or howto's out there that explain how to do this better?

 

Last time I hooked a multimeter up to some equipment, I think I was about 9 or 10 years old.  I forget what it was, but I want to say something started smoking!  Ever since then, I've been a little leery about doing it again.  On the other hand, it was 20 years ago and I didn't know what an ohm was, or the difference between ac and dc!

Thanks!

I'm not sure if I follow what you're asking.  Equalizing?  There are some compromises and inaccuracies here, but there may be more in trying to actually do the acoustic measurement.  Anyway, let's first assume we have an IEM with a flat frequency response, known to have sensitivity of 120 dB SPL / 1V input.  This figure should be given by the manufacturer; a review site like InnerFidelity may have measured it as well.  If it's given in dB SPL / 1 mW input, you can do some easy math to convert if you know the impedance.

Play music, set the volume how you would listen.  Measure the voltage applied to the IEMs while this is going on (a splitter cable may be useful, just so you have something you can touch the measurement probes to; you want to measure between L or R channel and ground).  Let's say it was 0.02V (rms).  Then we know that the output SPL should be 120 - 20*log10(1 / 0.02) = 86 dB.  Done.

In practice, manufacturer's spec may be slightly off.  A cheap multimeter is not going to give an accurate true rms voltage reading.  I doubt you have an oscilloscope handy.  If you want, you could instead just play say a 50 Hz tone at roughly your listening volume, try again for a 500 Hz tone, etc., because they can actually read pure sine waves with no problem, unless they're out of calibration or can't handle that level or frequency.  If the IEM has a much stronger response in some frequencies than others, you may have to consider that when picking out test tones, if you want to use pure sine waves.

How accurate is an SPL meter without the cardboard or plastic baffle?

I'm measuring koss portapros and apple earpods by sticking the tip of the meter directly against the foam covers of the drivers for these 'phones. I really have no interest in building a specialized baffle for every pair and styles of headphones I own. 

Good bump, but also a good question. I found that "without baffle" was only a small number of db less than "with baffle" on 2 different headphones (in no case was the tip of the meter touching the driver). I can't remember the exact difference now, but let's round it up to 5db to be safe.

This makes it less accurate, but the whole process isn't that accurate anyway, so I reckon "without baffle" is fine as a quick check and just add 5db to the reading. And if you stay below say 80db peak, then you should be safe by any standards.

A much bigger difference in db reading will be the average v peak setting. Average will usually result in several db lower than peak (sometimes these are called slow and fast).

Quote:

Originally Posted by TheAttorney 

Good bump, but also a good question. I found that "without baffle" was only a small number of db less than "with baffle" on 2 different headphones (in no case was the tip of the meter touching the driver). I can't remember the exact difference now, but let's round it up to 5db to be safe.

 

This makes it less accurate, but the whole process isn't that accurate anyway, so I reckon "without baffle" is fine as a quick check and just add 5db to the reading. And if you stay below say 80db peak, then you should be safe by any standards.

 

A much bigger difference in db reading will be the average v peak setting. Average will usually result in several db lower than peak (sometimes these are called slow and fast).

Thanks for the reply. I did a quick and dirty reading with and without on my portapro's. I just rolled up a piece of paper, taped one end to the meter, and created a larger cup for the foam cover/driver. 

I'd say the readings were very close, within say about  2 dBA of each other. 

I don't have any other procedure for measuring IEM spl except to stick the meter directly against the foam covering the driver of the IEM. I haven't done a before/after (with/without foam), but I tend to doubt that there's any more than a 2 or 3 dBA difference there either.