[playmusic]

I would like to keep listening levels with my headphones at safe levels.
This means that the rms of the sound should be in general below 85 dBSPL.
(There are various formulae for upper bounds on safe durations at higher dBSPL exposure per week,
but for simplicity I only consider the 85 dBSPL limit.)

In particular Stax listeners claim that the "clean" sound seduces people to turn up the volume beyond safe levels.
So, high volumes seem to be reasonable concern.

Let us consider the case that no fully equipped headphone measurement device is around to measure dbSPL at ear level.

For simplicity I consider a digital setup with integrated DAC and amp with a single volume control:
digital sound file -> amp -> headphone.
The following steps are required:

1. Determine the rms of the digital soundfile. (Easy: several free tools available).
2. Determine the voltage that is fed into the headphone for a full scale sine signal at maximum volume of the amp.
3. Find out the sensitivity of the headphone. (dBSPL @ 1 Vrms @ 1 kHz. Should be easy: check product sheet.)

If 2. can be achieved and the amp has a -dB display (or some printed -dB scale next to the volume control knob) for the volume control then it is just a matter of multiplying the different parts (or adding the dB).

Here is my first question. If 2. can be achieved but the amp has no -dB display: What are you doing to keep dBSPL at safe levels?
Painting some small dots in -5 dB intervals next to the volume control knob are one option.
It seems that the more expensive an amp the more restrained its design.
In particular amps for Stax headphones have the least amount of information around the volume control knob.
[Side note: I do not consider here controlling the volume via the media player when listening to music. This would be directly reducing the dynamic range of the audio. Controlling the volume via amp/DAC offers at least the chance that DR is not reduced if the amp has a reasonable quality. On the other hand, controlling via mediaplayer might be useful during analysis of the amp volume control.]

My second question is: How can 2. be achieved for Stax?
Here is what I have done for 2. for a dynamic headphone and an amp with -dB display:
I measured with a multimeter the unloaded voltage of the amp at max setting of the volume control for a full scale sine signal.
Then I looked up the output impedance of the amp. (Actually, I also calculated the output impedance with a multimeter and a shunt.)
I looked up the resistance of the headphone.
Then it was straightforward to determine the voltage over the headphone at max volume.

Now: How can I determine 2. for, say, an amplifier for a Stax headphone? (Ok, here amp and DAC have to separate except for the SRM-D10.)
Things get complicated because of the capacitive reactance of the headphone.
Output impedance of an amp is not specified by Stax and I would rather not connect a shunt at those high voltages to make measurements on my own.
Furthermore, even if the output impedance of the amp was known, the impedance of Stax headphones is specified as a real value in Ohm. I am not sure about the complex (real + imaginary) calculations required in analogy to the dynamic headphone case since I have only a very rudimentary understanding of electrophysics.
The maximum output voltage as specified by Stax might be helpful in this calculation. But then again I wonder if the specified amp gain of, say, 60 dB applies to the unloaded output voltage of the amp. Or if the unloaded voltage of the DAC output is multiplied by the amp by 1000 into the output voltage with the headphone as the load connected.

I posted this here in the science forum because my guess is that in the Stax thread in the high-end audio subforum people would stop reading after my proposal to paint small dots next to the amp volume control knob.

 

[castleofargh]

First, my serious advice is to rely on your ears to determine what is loud. When in doubt, you're probably louder than you should. If you're concerned about your hearing, set up some way to take pauses from sound at regular intervals. Maybe even wear some earplugs if the environment itself remains pretty noisy, just so your ears can sort of rest. They never completely do, but at the very least it can be nice to go below the level triggering the acoustic reflex for a while(about 60 or 70dB SPL usually)so the muscle can rest for a while and do its job later on when it's needed.

About measurements, you clearly have some ideas. If fooling around with amps that output several hundred volts makes you have second thoughts, congrats on having a survival instinct.
Why don't you just go with any other sound source available to you(non electrostat stuff) where you'll have maybe a volt or 3, and measure that instead without having to fear for yourself or for your gear? After that you can just use your ears to try and match the level of the Stax to that source. With a single tone at 1kHz or 2kHz you can get very close in level by ear(within a dB isn't far fetched at those frequencies). It's not good enough for blind testing and all that where 0.2dB can cause a change in our impressions, but for your needs, I'd say that it's very good.

Then you could confirm your result with some crappy meter app on your cellphone(it's always good to have alternative approaches to find out if anything is dramatically wrong, like my math so often is ^_^). I would expect some variations even if you go through the effort of trying to seal the space between the earpads and the cellphone, but even that shouldn't fall too far away from your other method(after all, 6dB is already twice the voltage). I'm guessing Apple has an advantage here as the app doesn't have to know tens of cellphones, so it's more likely that the calibration would be more accurate on an Iphone(IDK, I'm just guessing out loud).
Even if you decide not to trust the phone's app, you could still keep it around to measure the variations in SPL between what you once determined as your reasonabble max, and whatever you're getting anytime you want to check. that could work instead of your dot next to the volume knob maybe. you'd put the cellphone under your ear from time to time and check to see if you've drifted into the danger zone(whatever value you decided was too much, which is probably the hardest stuff to determine in practice because we're dealing with music, not constant noise).

I personally have everything I need to calibrate my mics(within reason, I only have amateur devices), measure voltage(not Stax!), even a bunch of cables made so I can insert the multimeter to measure at the load(headphone), and a bunch of resistors for when I can't, so I don't have to rely on unloaded measurements. I only bother with all that when I do IEM measurements nowadays, because stuff like distortions may greatly change if the output level is different. So it's important to be calibrated. But for my own listening I just stick to how I feel and try to lower the volume from time to time if I spend more than an hour with a headphone. But I guess a little dot on the volume knob isn't a bad idea for general purpose(until you get a different DAC and have to pray that the full scale voltage is close to that of the previous DAC. Luckily it usually is).
Personally I had to rely on many half reliable sources of information because the worst part of my tools was the multimeter itself that I got for less than 15euro. So I first had to use anything I could find. The full scale output of some DAP I had that a guy online had measured with good tools. The sensitivity of some IEM I owned that someone had also checked and confirmed to be what the manufacturer gave(which isn't always the case and reassured me that mine was also likely to fall in that same area of sensi and impedance). Some IEM or headphone that came with per unit measurements(that's so very cool when it happens). Some cheap measurement rig with a suspicious but existing self calibration procedure, etc. I really pulled together anything I could, while not really trusting anything. And it's the agreement between different approaches that increased my confidence that I was on the right track. For example I'm now pretty confident that I have the right voltage from my multimeter into most loads if I use a 400Hz tone. And cellphone apps were honestly a good surprise in my case when I tried a few back in the days. I got too much variations from just the way I would place the phone, making it useless as a precise calibration tool. But to get a general idea, it was simple and fast. so that might just be what you're after.

 

[bigshot]

I agree. 85dB is a little over the edge of comfort. Why listen to things at a volume that is uncomfortable? It's unlikely that most people would even want to listen that loud.

 

[playmusic]

Thank you very much for your comprehensive reply!

to go below the level triggering the acoustic reflex for a while

This concept was an unknown unknown to me. I will read up on it.

Why don't you just go with any other sound source available to you(non electrostat stuff) where you'll have maybe a volt or 3, and measure that instead without having to fear for yourself or for your gear?

Yes, this is what I did.

After that you can just use your ears to try and match the level of the Stax to that source.

Oh ...
Why did this obvious approach not occur to me?

Btw, I have no electrostat, yet. I am still improving my listening skills with the harman "how to listen" software.
Once I have reached a reasonable level, I will audition some Stax headphones again.

because the worst part of my tools was the multimeter itself that I got for less than 15euro

Same with me. One issue to watch out is the specified frequency for AC voltage. Quite a lot of multimeters are only specified around 50/60 Hz for mains voltage.
So, a 50 Hz sine test tone might give the best results.

even a bunch of cables made so I can insert the multimeter to measure at the load(headphone)

I have not got such cables, yet.
I guess this is made by removing the isolation(s) in the middle of a 3.5 mm headphone cable?
What I did was to use some unstable crocodile cable connection setup and make the measurements at the bare metal.

Are you aware of any passive electronics parts that can be purchased for making this type of measurements?
For example a small box where you can plug in two cables (one output to headphone, one input from amp) and have access to the cables in the box so that one can make measurements of the system under load (headphone)?
Or a small box with 3 cable sockets (2 input from amp, 1 output to headphone) and a switch to make a DBT of two amps easier?

 

[playmusic]

I agree. 85dB is a little over the edge of comfort. Why listen to things at a volume that is uncomfortable? It's unlikely that most people would even want to listen that loud.

Fair question.

Musicians in a symphony orchestra are exposed to 85-90 dB SPL rms during a concert.
So, the audience sitting towards the front is exposed to 80-85 dB SPL rms.
Which is the range I listen at for classical music.

I agree that this is rather on the loud side since most people in the audience sit further back.
Still, this is the level I like best for the dynamic range of classical music.
I estimate that I listen with headphones for less than 10 hours a week. So, this should be ok, I hope.

But if the sound level goes up, I soon hit a barrier.
I cannot stand 95 dB SPL rms for 15 seconds.
Still, 90% of people seem to be ok with those volumes at parties.

 

[castleofargh]

I have not got such cables, yet.
I guess this is made by removing the isolation(s) in the middle of a 3.5 mm headphone cable?
What I did was to use some unstable crocodile cable connection setup and make the measurements at the bare metal.

Are you aware of any passive electronics parts that can be purchased for making this type of measurements?
For example a small box where you can plug in two cables (one output to headphone, one input from amp) and have access to the cables in the box so that one can make measurements of the system under load (headphone)?
Or a small box with 3 cable sockets (2 input from amp, 1 output to headphone) and a switch to make a DBT of two amps easier?

I wouldn't advise to open up your headphone cables just so you can stick the multimeter to it. In my case, I initially used short male to male adapters, with crocodile cables to close the circuit between 2 male jacks, so I could put the multimeter there. I would on occasion add a little bread board to hold some resistor.
Later on I finally got around to create some abominations using a soldering iron, zero soldering skill, and some cheap cable parts bought online.
I'm sure there must be some manufacturers doing all forms of adapters like we see in labs and even in schools with banana plug everything, but I never looked into those.

 

[playmusic]

I wouldn't advise to open up your headphone cables just so you can stick the multimeter to it.

I realize that I was unprecise in my description.
What I meant was a female to male cable in which I remove part of the isolation in the middle.
The male plug goes into the output of the amp, the female plug into the male plug of the headphone cable.
So, this cable could be used for measurements of all headphones (of this plug type 3.5 mm or 6.3 mm).

Later on I finally got around to create some abominations using a soldering iron, zero soldering skill

I am afraid that I might have to go down this road if nobody directs me to a shop where I can by these things.

I'm sure there must be some manufacturers doing all forms of adapters

Any forum members around who know where to buy these small devices mentioned in the last paragraph of post #4?