• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Headphones, Headphone Amplifiers, Tools and Methods

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
#1
Hello All,

Just for fun and learning this past Wednesday a shiny AP / GRAS 45CA and 12AA power supply appeared on my doorstep.

The effort is to sort a basic set measurements for headphones, headphone amplifiers equalizers, transfer functions and perhaps some software psychoacoustic convolutions.

First will be the easy stuff; frequency response, sensitivity / gain ......, and your know the basic 6 or 8 measurements.

Join in if you like.

Thanks DT
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #2
Hello All,

The Audio Precision test amplifier, APx 1701 transducer test interface has a SPEAKON connector for stereo output. Inside the 1701 is a pair, one for each channel, of current sense resistors to calculate the current through and impedance of each connected driver under test. Using the APx500 Impedance Theile Small software tool; Impedance, TS / P, Phase, Group Delay and similar parameters are calculated. Using a reference amplifier seems to be a valuable tool. Now I am sorting some bench made cables and such for headphone parameter measurement. See the attached photo of the newly fabricated SPEAKON cable.

speakon.jpg


Thoughts?

Thanks DT
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #3
Hello All,

Now that I have a GRAS 45CA on my bench along with an AP analyzer I have been doing some catch up with what type of testing is being done with headphones. We are aware of the research that Harman has been doing and published AES papers. There is also an IEC international standard for headphone testing. I suppose that I will need to pony up and purchase a copy for my bench.

IEC 60268-7, Sound system equipment, Part 7: Headphones and earphones

Audio Precision is an industry leader in audio test equipment. GRAS is also an industry in sound and vibration measurement instruments. AP and GRAS are now under one roof and ownership.

AP has published an application note “Application Note: Headphone Electroacoustic Measurements”. You can download the Application Note at AP.com, you will need to sign up and log in to download.

So what needs to be measured? First off to be put on the list is Frequency Response, “Frequency response is the single most important aspect of the performance of any audio device. If it is wrong, nothing else matters.” (Floyd Toole, 2009)

I am purposely leaving out the “Harman curve” for future discussion. The Harman research speaks of speakers measured and equalized in the free field as being the standard that our ears are used to hearing. Assuming that to be the case, headphones that sound most like a set of those speakers in a “standard” listening room will be preferred by the large majority of listeners. Of course your mileage may vary.

AP says that the IEC standard address most measurements as being “Objective” as in being measured with calibrated instruments and coordinated with measured human hearing, @overkilly will be pleased that the IEC standard allows some subjective equalization of frequency response curves. Human subjects do a qualitative determination of equal loudness of tones between speakers and headphones. This all ties back to laboratory quantified measurements.

This is a personal impression. The first Headphone thing to be included in the measurement report is Frequency Response. The next thing included should be some sort of equalization to bring the FR back to near the Frequency Response of free field equalized speakers in a “standard” listening room.

More of my musing; personal equalization curves are the last thing to measure and do not belong in a headphone measurement report.

Thanks DT

Comments about what should be measured and reported?
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #6
Hello All,

I have determined that that a good portion of the “Distortion” shown above on the plot is AP instrument Fan noise that the 45CA microphones are picking up.

The first thing to do is ditch the APx1701 test amplifiers and use a Geshelli headphone amplifier without a fan.

The next thing is to get some longer BNC cables and place the 45CA test fixture “Hammer” in the closet in the room across the hall. That removes the “hammer” by two hallway doors plus an additional closet door. That may be some technical attenuation of the fan noise. I will try it.

The next thing to try is to ditch the APx555 analyzer and fan. In November Audio Precision ran a special; 20% off everything; the 45CA, the 12AA power supply and the Flex version of the APx500 software. The discount largely made the software free. My computer has large heat sinks and slow turning quiet heat controlled fans. The APx555 fan is near 30db noisier than the Coffee Lake computer. The AP Flex software will work with the ASIO sound card. For measuring headphone distortion the magic of the APx555 is not required. We will see the difference.

Thanks DT
 

Dreyfus

Active Member
Joined
Feb 7, 2019
Messages
225
Likes
249
Location
Germany
#7
How much TDP does the fan have to handle in your scenario?
Maybe you could support or even replace it by using (as large as possible) passive heatsinks.

Regards
Dreyfus
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #8
@Dreyfus,

I like the idea of passive heatsinks.

When it comes to it I do not want to attempt such a thing with the expensive instruments.

Thanks DT
 

Dreyfus

Active Member
Joined
Feb 7, 2019
Messages
225
Likes
249
Location
Germany
#9
Then it is probably best to build an isolated, well damped chamber for the fixture. You will need that for low THD measurements anyway. :)

Innerfidelity.png rtings.png headfi.png
 
Last edited:
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #10
Hello All,

Yes, isolated, and damped are key words, they go along with vibration and sound engineering.

I am going to try a few simple things before I use a 1300 pound isolation platform as Head-Fi does.

I wonder why AP does not use isolated cases, engineered, air foil, and thermostat controlled fans? MY pc is quiet, the video card has a big heat sink and is without a vibrating, screaming fan.

Thanks DT
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #11
Hello All,

I just purchased a Topping L30 Headphone Amplifier to power bench testing of headphones.

Now one less amplifier fan to make noise.

Thanks DT
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #12
Hello All,

When the 45CA arrived it came with copy of AP APx500 V 6.0 software and a Flex dongle. The AP software will operate with a lesser analyzer than the APx555.

The software will operate with an ASIO based interface. I have a couple of sound cards and a RTX6001 analyzer to try. The key feature of the ASIO devises here to test drive is the absence of instrument cooling fans. I believe that the resolution of the ASIO devises is still 10dB better than the sound floor and distortion the 45CA test microphones. The absence of the instrument fans will be a plus as well.

I will give it a go.



Thanks DT
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #15
Hello All,

This is an acoustic sweep of the AKG K712PRO tested on a GRAS45CA-9 test hammer.

To the best of my information this is the headphone used by Harman to test preference with varying frequency response curves. The headphone has orange trim and cord. I call it the Chameleon headphone”; it can sound like anybody else. The sound is smooth, as in not harsh. The headphone sits comfortably on my head and ears.

I plan to test these K712PRO headphones with different amplifiers; different output impedance, tubes, Equalizers, FET’s and you know whatever else we think of.

This test plot was done with the APx555 and APx1701 transducer test interface. To reduce ambient noise I turned off all the things like the furnace, refrigerator and humming light transformers. The test hammer was positioned across the hall in another room with two doors closed in between the instrument fans and the test hammer.

Thanks DT
K712ProLevel and Distortion.png
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #17
Is there a special objective reason for that compared to other good (low resonance and distortion) headphones?
Hello,

Yes the reason is that Harman/AKG picked it to use for their frequency response preference testing. If they picked it for their testing, objectively it must be as good or better for the job than most headphones.

Sure there are other headphones that will do the job as well. My thought is that people will accept the AKG K712PRO as the standard "Chameleon Headphone”.

Thanks DT
 

thewas

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
1,889
Likes
4,082
#18
Yes the reason is that Harman/AKG picked it to use for their frequency response preference testing. If they picked it for their testing, objectively it must be as good or better for the job than most headphones.
For me that is rather a brand imposed choice (which headphone of our group brand fulfils the desired task better) than an objective choice over other good brands headphones for that, meaning it may be one of the best translating HP from the Harman brand, but possibly other great headphones might do an even better "chameleon job".
 
OP
D

DualTriode

Active Member
Joined
Oct 24, 2019
Messages
250
Likes
144
Thread Starter #19
You got it, Harman is going to pick AKG over some other brand.

In terms of data some other brand has some catching up to do to prove itself.

Thanks DT
 

Dreyfus

Active Member
Joined
Feb 7, 2019
Messages
225
Likes
249
Location
Germany
#20
What do you think qualifies a headphone to be a good chameleon?
I would translate that into a consistent inter-subjective seal and PRTF.

By the way...

olive.png


I hope you don't mind the meme. :D
 
Top Bottom