Headphone Design

[jaxvon]

Ok, this is a call for help to all the pros and engineers lurking on this forum. This is not for sure, but for an engineering project we need to pick a product and find ways to improve it. It's a group project and one of the products that we have opted to have approved by the professor was a pair of inexpensive headphones. As most know, these could definitely be a lot better and still be cheap.

Here's the rub: We can't just propose improvements, we need a pretty good working knowledge of the product. Granted, we will be tearing apart two pairs of 'phones, but I still need more technical knowledge. Without telling me to go buy some $6000 testing equipment, does anyone know of some good ways to test headphones besides listening? I was thinking of a good SPL meter, but I'm not sure how that would work with headphones.

If we end up going with this project, we will need some ways to quantify the performance of the headphones. Listening tests will of course be used, but I need something else. I have no idea what kind of equipment is available here on campus, but I'm sure if I sniffed around I could find something I could use. The problem is, I haven't done that.

Also, what is a good resource (hopefully online) that I can use to get information about the construction and design of headphones and the drivers? Sorry about the vague questions. Post any concerns you might have, and I'll try to reply quickly. I'm on this forum a lot, so I won't miss your message.

Thanks in advance!

 

[BMXTRIX]

Without being able to provide you much help, make sure you use the search feature of the forums and look at the many headphones topics already posted and go through the links listed. They are what I used to buy headphones and had good info and headphone specific forums.

One of the big issues you have is these to statements:
1. "these could definitely be a lot better and still be cheap."
2. "I still need more technical knowledge"

Headphones, very likely can't be a lot better - or any better - and still be inexspensive. There are major manufacturers who focus on cheap headphones (Sony & others) as well as companies that specialize in headphone production who simply can't put out cheap headphones. I would question your hypothesis in line 1 before anything else. Ask yourself: Is there a way to make them of higher quality for less money? What production costs are involved? How many would need to be sold to make a profit? What is the raw price of materials? What about the cost of proper testing equipment to do accurate testing? How will they be used?

Anyway, it sounds like a fun project, but I don't personally find headphones to be overpriced for their quality in the least. There are dozens of different types of headphones that come in hundreds (or thousands) of different models. Just a quick search on 'headphones' on eBay gives me about 6,000 results and you have to go about 10 pages deep before a pair comes up that has bids greater than one penny.

Just some crap to consider from someone clueless about headphone engineering... and as I said, it still sounds like fun to make.

Want some real fun that is current and has potential? Why not a pair of really good on-wall speakers that are bigger than the crap designed to go with 42 or 50 inch plasmas? I've been looking all over for something to get up off of my floor to replace my Definitive Technology BP-30 speakers, but everything seems pretty wimpy by comparison and manufacturers claim that if they make them bigger they won't be wall mountable. Sounds like an engineering cop-out to me.

 

[sjdgpt]

Without telling me to go buy some $6000 testing equipment, does anyone know of some good ways to test headphones besides listening? I was thinking of a good SPL meter, but I'm not sure how that would work with headphones.

My suggestion is use your ears.


I read an headphone "debate" in an audio rag many years ago (circa 1986/87). I as I remember, there was a problem with the SPL meter, specifically the size of the receiptor on the meter was incorrect (too large) versus the size of the headphone "speaker", the distance the meter was from the "speaker" was difficult to properly calibrate, the angle of the meter was critical, and there was a problem with standing waves not being absorbed (or being absorbed).

This little "debate" was meaningful to me at the time, as I was working with a civil engineering firm, and we were having similar problems testing a scale model. So when I read the audio debate, it was like relieving my days work, and it just reaffirmed that testing was a problem for many industries.

 

[WmAx]

Without telling me to go buy some $6000 testing equipment, does anyone know of some good ways to test headphones besides listening? I was thinking of a good SPL meter, but I'm not sure how that would work with headphones.

The main priority is a REPEATABLE measurement system so that relative differences/changes can be accurately documented/recorded. So, a jig is needed that can precisely align the headphones to the microphone and obtain repeatable measurements. A simple SPL meter is not sufficient resolution. You need to be able lto analyse using a MLS signal; which is the easiest way to derive frequency/phase data and maintane a high signal noise ratio to reject environmental noise. A jig/baffle that the headphone rests up to with a hole containing a microphone, is needed. The baffle sruface that the headphone sits agains should be acoustically absorbative, otherwise the reflections will make interpretation of relative measurements more difficult.

I have no idea what kind of equipment is available here on campus, but I'm sure if I sniffed around I could find something I could use. The problem is, I haven't done that.

A software package such as Loudspeaker Lab or any number of others, coupled with a compatible soundcard and linear microphone/mic preamp is needed along with the jig.

Also, what is a good resource (hopefully online) that I can use to get information about the construction and design of headphones and the drivers?

Several documents exist in the AES database; but these are at a cost of course. As far as actual engineering guides for headphones go -- I don't know. I've never seen such material published online.

Sorry about the vague questions. Post any concerns you might have, and I'll try to reply quickly. I'm on this forum a lot, so I won't miss your message.

BMXTRIX brings up an excellent point: their is not really any room left for making a cheaper/better headphone. For example, the Sony MDR-7506 is one of the finest headphones for monitoring accuracy and has superb/durable construction quality. It's cost? About $100 USD.

Additionally, one of the best sounding headphones I've heard cost only $140(AKG K501--with modified ear pads to move drivers closer to ears) or so bucks on ebay. I've listened to units costing up to $4000, and the only thing they had over the K501 IMO was the exotic construction using fancy leathers and exotic woods. <shrug>

Of course, their are headphones that are overpriced for what you get. But the thing is -- headphones that are cheap vs. performance are already in existance.

Take the Sony MDR-7506 into consideration. How do you propose to keep the same construction quality and part quality and then cut the price in half and still make a profit that makes production worthwhile? You can find the Sony MDR-V6(same as 7506) for about $70 USD if you look around hard enough.

What more could you possibly expect for that price? I just don't know. The device has adequate mechanical capability to have no compromse bass response with no distortion, etc. The only thing that you are really ablel to change is the frequency response -- tune to preference so to speak. Since this is a headphone, it is much simpeler case than a speaker. The frame/enclsoure is subject to far less problems than a speaker cabinet and the headphone does not require analysis of off axis response and room effects as is required for a speaker analysis.

-Chris

 

[jaxvon]

We have a pretty low price ceiling for two models of a product that we have to improve (only $50). That's why I'm not going to be dissecting a pair of studio reference 'phones. Also, I don't need to do a HUGE project determining all the testing costs, production costs, etc. The gist of what I'm getting at is that for a cheap pair of headphones, I'm sure some improvements could be made in sound quality without turning them into something like my MDR-7506s.

My goal here isn't to make a headphone more affordable. Obviously, that has already been done to the extreme. However, I (with my group) am looking for ways to improve a product. Many times headphones that are convenient are lacking in comfort (earbuds, wrap-around). Many times, these headphones are subject to lacking bass response or extreme distortion.

Now, before you go off on me, I understand that the price is a big factor in quality. However, take a look at the wrap-around style 'phones. Those basically only come in one price range: $20-$25. These are convenient however, because they tend to be fairly durable, aren't as rediculous looking as normal headphones, and they don't mess up your hair if you have some gel on it. Earbuds do this too, but they really can't handle SPLs or bass response. I think people would be willing to pay a little more if they could get more comfort and quality out of a pair of relatively cheap headphones. And I do know about Etymotic Research, so don't even go there.

Thanks for the responses guys. I need to do some sniffing around in the EE building to see what kind of equipment I can find. This university has billions of dollars, they MUST have something I can use for this project. If not, I'm going to yell at people for wasting money or something. Any other ideas/resources anyone knows about? The U has an enormous library, so they might have papers on this kind of stuff. If you can give me some titles or at least what to look for in documents, it would be much appreciated.

Thanks!

 

[WmAx]

We have a pretty low price ceiling for two models of a product that we have to improve (only $50). That's why I'm not going to be dissecting a pair of studio reference 'phones. Also, I don't need to do a HUGE project determining all the testing costs, production costs, etc. The gist of what I'm getting at is that for a cheap pair of headphones, I'm sure some improvements could be made in sound quality without turning them into something like my MDR-7506s.

True. You could take a high grade of driver and install it into a very low priced headphone frame chassis ; it would increase perofmance substantially. But I can't help to think this is bit like taking a 300HP performance motor and popping it into a Festiva chassis. What's the market for this? What is the driver cost relative to the frame/chassis cost? Is this sellable or just a 'niche market item? It's an odd mix n' match.

However, I (with my group) am looking for ways to improve a product. Many times headphones that are convenient are lacking in comfort (earbuds, wrap-around). Many times, these headphones are subject to lacking bass response or extreme distortion.

For comfort, it might be useful to examine the Sony MDR-F1 design. It is probably one of the most comofrtable designs. It has no contact with the ear/pinnae and is totally open(literally), preventing ear sweating. It uses soft cloth pads. For a method to increase spatial properties refer to AKG K1000. The drivers are on a hinge and can be angled away from your ears -- in front of you. This allows the ear pinnae to apply it's custom transfer function. Of course this method requires you to leave behind your stated desire of having the device be non-objectionable, visually. This also increases cost significantly, as you need a driver with relatively robust motor/spl capabilities to produce the required SPLs at bass frequencies at the much greater distance from the ears.

The U has an enormous library, so they might have papers on this kind of stuff. If you can give me some titles or at least what to look for in documents, it would be much appreciated.

I can not direct to you a specific paper on headphones; as I have not read the type of papers on headphones that you seem to need. For peer reviewed articles on various headphone issues go to this link and type in headphones:

http://www.aes.org/journal/search.cfm

Hopefully you can find them in your library-- otherwise break out the CC.

-Chris

 

[jaxvon]

Thanks for the link. I'll need to check out the libraries here tomorrow. I think I'm going to join the AES. The student price is only $22, which is quite affordable.

Also, with regards to my project, we only need to propose improvements and how to implement them. We aren't actually going to be making better headphones. However, this still requires a good understanding of the science behind headphones, hence my requests. I'll keep looking.

Thanks again.