QuestionQUESTION: Dear Cleggsan,
I was wondering what equipment, software and hardware, I would need to test headphones. I want to test sound performance from an objective perspective on things like frequency response, tracking, distortion, isolation and leakage. I would also like to test how well headphones can actively cancel noise. I am responsible for bringing this capability in house for the company I work for but do not know where to start. Thank you for all your help.
Regards,
Wayne
ANSWER: Good question. Unfortunately, there are no universally accepted testing methods or standards that cover all kinds of headphones.
One international standard that has merit, but is not universally accepted (because of the many subjective factors in deciding what is "objective") is the IEC document, 60268-7, which is quite comprehensive but takes much study and lots of certified test equipment to set up all the testing equipments. Here is where you can find out more about this measurement standard:
http://webstore.iec.ch/preview/info_iec60268-7%7Bed3.0%7Den.pdf
Here is a ppt you should take a look at:
www.gbcasa.org/.../Griesinger-Binaural-Hearing-EarCanals-Headphones.ppt
The above relates to hearing aids and to in-the-ear headphones but has some general principles outlined as well.
Let me suggest you do a google search on "headphone testing" and scan through some of the offerings and comments from the many in the business of doing what you want to do.
This HeadRoom site has some good tutorial grade comments and techniques. Please study it.
http://www.headphone.com/learning-center/headroom-test-lab.php
Many labs that are doing the kinds of tests you are thinking of devise their own methods and do comparison testing of various models or prototypes - which gives good comparable results against others products or ideas. This is often the only way as there are too many variables in determining the results you are hoping for.
The problem is:
ONE: Measuring acoustic transducers of any kind presents special problems because of boundary and dispersion characteristics and the variation thereof vs frequency and power level.
TWO: Air leakage in and around the ear is nigh impossible to control exactly.
THREE: Deciding on which dummy head or ear design model to use makes a big difference in the results. A few standard dummy heads have been proposed over the years in order to get repeatable results but that does not provide the correct result for every head size and shape.
FOUR: The test results are often greatly influenced by where and how the pickup microphone is placed when looking at the acoustic response at the inner ear.
FIVE: There are so many kinds, types and design constructions for headphones that it presents a very difficult measuring strategy to get comparable results from one basic type against another.
Because of the complexity in taking measurements and getting comparable results the testing scheme or strategy nearly always includes listen testing either with a panel of with "golden ear" candidates who have trained experiences on how to listen to them for best performance.
So far as leakage and noise cancellation features it is the same story all over again. There are two basic methods for eliminating noise: 1) muffs over the ear that block out the external sound and 2) cancellation of the external signal by electronic means (a small microphone is used to pickup the external sounds; it is then inverted and fed into the earphone amp to electronically cancel the noise stuff). Combinations of the above are most often used in the higher priced systems.
The testing can be objective methods with microphones placed around the head while signals are injected into the test heaphone and dummy head, etc. But, so many design and testing labs use as the principle testing or to augment the results a listen test with a human subject.
The problem of repeatability is big. A manufacturer may give certain quantitative data in the specifications of their headphone but no two labs can get the same results with the best testing methods employed. In deed, even testing the same exact model one day and the next often do not yield the same result.
So, back to the opening comment; often one merely develops their own testing sequences and learns over time to refine the methods and collect comparative data that provides an ongoing collection of data points that helps to put things in perspective and qualitative if not quantitative portfolio of success.
Sorry for sort of beating around the bush but it is a difficult area.
If you care to share with me the specific class of products you are going to be working with it would help me guide you more directly. Transducers used in wireless telephones and the like are quite a different proposition compared with music or high fidelity products. In fact, cup type "over the ear" phones are evaluated quite differently from, say, in-the-ear plugs.
Let me know how I can help further. And, best wishes.
PS: The Nuemann Dummy Head KU-81i is quite popular for many measurement references; it is likely that just about every design laboratory around the world has one or one essentially similar. You can see it here:
www.neumann.com/download.php?download=cata0059.PDF
---------- FOLLOW-UP ----------
QUESTION: Dear Cleggsan,
Thank you for all that information, I've studied it and realize this is much more difficult than I had initially anticipated. To further specify my needs, let me inform you of what I am trying to test. For industrial safety, we want to test well ear buds can clip out noises above a safe noise level of 85db and have the ability to amplify low level noises such as speech while at the same time eliminate loud white noise. I don't know if that means we amplify certain frequencies and eliminate others with a max point of 85db before it starts clipping out.
For quiet environments, we would like to test the performance of the ear buds to see how much noise is being leaked out of the ear buds into the surrounding environment. We essentially want the ear buds to not leak any noise.
Also, the third thing we are looking for is a fixture for measuring where noise comes from. This could be solved with a rotating stage to see the frequency response at different angles. If there is another method please let me know.
So I think I need the following equipment, please correct it if I am missing anything:
1. artificial head with microphones
2. sound proof chamber
3. audio analyzer
4. headphone amp
5. rotating stage
Again, thank you so much for your help.
Regards,
Wayne
AnswerPLEASE GET IN TOUCH WITH ME:
Could you contact with me regarding these headphone questions? My personal email is cleggsan at ieee dot org. 3/18/11 Thanks.
====================================================
You have summarized it well.
The problem with earbuds of any kind is getting the cup to fit tightly inside the ear in order to keep the sound produced by the transducer in the ear canal and not leaking out AND to keep outside sounds from getting into the ear canal. The fit is not only influenced by the shape of the sealing cup but the material used and the shape/size of the human's ear.
I do not know of any standard for this type of test nor do any of my colleagues have specific experience in this are. However, there are a couple of experts that are sufficiently experienced in working with headphones and earbuds of various kinds that they could help to advise you on how to go about doing an objective test sequence. My recommendation is that you use the services of a professional if for nothing else than to validate your testing idea and help you select the right kinds of equipment to use. If you intend to certify against some government or civic requirement, for example, you will need someone who can reference against industrial documentation, national or international safety and compliance issues.
If you would like a couple of suggestions for consulting engineers - who have much more experience in this area than I - please let me know your email address and I will communicate with you privately. cleggsan at ieee dot org
One engineer is located in the Chicago area and the other in Minneapolis.
PS: In addition to the items you will need, suggest you look at automated test set that can record each event and store all the results in memory for later comparisons. This company
http://ap.com/
makes such equipment and I can recommend them to you.
Best wishes.
==============================
FURTHER COMMENT: I would like to talk to you about this project. Could you contact me please? Email me at: cleggsan at ieee dot org
3/21/2022
==============================
