Measuring headphone FR?

natehansen66 · 15th February 2013, 12:23 AM

So I'm a DIY speaker guy, new to headphones. I thought I'd get a set for reference and listening later in the evening (I live in an apartment). After some research I picked up some Shure SRH-440s. I use the headphone out on my MobilePre that I use for measurements. Compared to my cheap in-ears the Shure's just didn't sound right. Little bass and overly bright. Headphone.com's measurement show fairly flat response, rolling off at 100hz.

Being a DIY type, I figured there had to be some way I could measure these things and correct the response. I searched the 'net, and came up with a site that had a tutorial for making an in-ear mic to measure headphones with a Panasonic mic capsule. I made my loudspeaker measurement mic with such a capsule, had it calibrated, and have a few capsules left. I wired one up, and took a measurement with it right next to my regular mic. They were close enough to use the cal file I had made for my "reference" mic.

I very carefully put the mic capsule into my ear, taped the lead to the side of my face to keep the cap from pulling out, put the 'phones on, and took some measurements. I'm not going to recommend anyone do this, for obvious reasons.

I adjusted the cap and the phones and took several measurements for consistency, and this is what I got:

No where near headphone.com's curve, but it follows what I hear. I figure my head isn't like their test dummy's head. My hair is thick and wavy too, so I suppose the cups can't get a good seal and that's why I'm seeing the bass drop like that. Is this a valid measurement? Obviously I can't place the capsule exactly where my ear drum is so there is some discrepancy there. I eq'd the response to flat and I think they sound quite good now.

Does this make sense or am I totally nuts?

cotdt · 15th February 2013, 01:15 AM

It's really hard to interpret headphone frequency responses. A dummy head would ruin a speaker's frequency response too.

You need a speaker (with flat response) playing through the same dummy head as reference.

cotdt · 15th February 2013, 01:17 AM

I measured several headphones too, and the frequency response varied wildly depending on my measurement setup. It's really hard.

mortron · 15th February 2013, 02:56 AM

maybe your measure is right, and the way they measure the FR of headphones is flawed?

jcx · 15th February 2013, 05:10 AM

Headphone Measurement Procedures - Frequency Response | InnerFidelity
Expert Tests InnerFidelity's Headphone Measurement Repeatability and Reproducibility Page 2 | InnerFidelity

cotdt · 15th February 2013, 05:16 AM

Sure, but it's not so much about reproducibility as much as what the heck the frequency response even means. What kind of response is ideal for headphones after the sound goes reflected through a dummy head?

31697B · 02:34 AM

I have done similar tests. Your test is OK. A dummy ear is so that the factory does not have to have a person sitting around all day being used as a dummy for testing. It DOES, as you say, correspond to what you hear. you can flatten the FR with an EQ... works well up to the treble and then it isnt so accurate... but still useful. Equalise it to be flat and see how it sounds... The references given above by jcx are very good basic info.

A useful book on the subject is called... gets into more depth ..... by Harvey Dillon.... Hearing Aids. Thieme publishers. $65 (used) at Amazon.
It will give a more complete appreciation for the hearing and testing issues.

-RNMarsh

natehansen66 · 16th February 2013, 02:46 AM

Quote:

Originally Posted by jcx

Headphone Measurement Procedures - Frequency Response | InnerFidelity
Expert Tests InnerFidelity's Headphone Measurement Repeatability and Reproducibility Page 2 | InnerFidelity

Thanks for the links. I looked through them before but didn't see/pay much attention to the HRTF compensation.

xnor · 16th February 2013, 09:03 AM

You only get that much bass as in the IF graph if you have perfect seal. When I tested the SRH440 I also felt that there was a strong roll-off below 300 Hz and that they were a bit bright for my liking.

natehansen66 · 16th February 2013, 03:55 PM

Quote:

Originally Posted by xnor

You only get that much bass as in the IF graph if you have perfect seal. When I tested the SRH440 I also felt that there was a strong roll-off below 300 Hz and that they were a bit bright for my liking.

Yeah, with my hair there's no way I'm getting a good seal. With eq below 1k, things are sounding pretty good.

15th February 2013, 12:23 AM	#1
natehansen66 diyAudio Member Join Date: Mar 2011 Location: Rochester, MN	Measuring headphone FR? So I'm a DIY speaker guy, new to headphones. I thought I'd get a set for reference and listening later in the evening (I live in an apartment). After some research I picked up some Shure SRH-440s. I use the headphone out on my MobilePre that I use for measurements. Compared to my cheap in-ears the Shure's just didn't sound right. Little bass and overly bright. Headphone.com's measurement show fairly flat response, rolling off at 100hz. Being a DIY type, I figured there had to be some way I could measure these things and correct the response. I searched the 'net, and came up with a site that had a tutorial for making an in-ear mic to measure headphones with a Panasonic mic capsule. I made my loudspeaker measurement mic with such a capsule, had it calibrated, and have a few capsules left. I wired one up, and took a measurement with it right next to my regular mic. They were close enough to use the cal file I had made for my "reference" mic. I very carefully put the mic capsule into my ear, taped the lead to the side of my face to keep the cap from pulling out, put the 'phones on, and took some measurements. I'm not going to recommend anyone do this, for obvious reasons. I adjusted the cap and the phones and took several measurements for consistency, and this is what I got: No where near headphone.com's curve, but it follows what I hear. I figure my head isn't like their test dummy's head. My hair is thick and wavy too, so I suppose the cups can't get a good seal and that's why I'm seeing the bass drop like that. Is this a valid measurement? Obviously I can't place the capsule exactly where my ear drum is so there is some discrepancy there. I eq'd the response to flat and I think they sound quite good now. Does this make sense or am I totally nuts?

16th February 2013, 02:11 AM	#7
31697B Account disabled at member's request Join Date: May 2012	I have done similar tests. Your test is OK. A dummy ear is so that the factory does not have to have a person sitting around all day being used as a dummy for testing. It DOES, as you say, correspond to what you hear. you can flatten the FR with an EQ... works well up to the treble and then it isnt so accurate... but still useful. Equalise it to be flat and see how it sounds... The references given above by jcx are very good basic info. A useful book on the subject is called... gets into more depth ..... by Harvey Dillon.... Hearing Aids. Thieme publishers. $65 (used) at Amazon. It will give a more complete appreciation for the hearing and testing issues. -RNMarsh Last edited by 31697B; 16th February 2013 at 02:34 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
measuring dc offset in OTL headphone amp	gh0st	Tubes / Valves	4	2nd March 2012 08:10 PM
Not quite FR, but a FR driver does most ~ A tower of power with balls (spheres)	TimboinOz	Full Range	3	28th October 2011 09:22 AM
Easy, simple to use software for measuring FR, impulse, etc?	makingmoney	Multi-Way	7	6th July 2010 09:45 AM
Measuring tweeter FR without a highpass filter	owdi	Multi-Way	3	14th March 2007 05:40 AM
IDEA: Measuring FR of a driver with room acoustics?	Rarkov	Everything Else	10	22nd April 2003 05:25 PM

New To Site?	Need Help?
Register to Participate Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

15th February 2013, 01:15 AM	#2
cotdt diyAudio Member Join Date: Oct 2005	It's really hard to interpret headphone frequency responses. A dummy head would ruin a speaker's frequency response too. You need a speaker (with flat response) playing through the same dummy head as reference.

15th February 2013, 01:17 AM	#3
cotdt diyAudio Member Join Date: Oct 2005	I measured several headphones too, and the frequency response varied wildly depending on my measurement setup. It's really hard.

15th February 2013, 02:56 AM	#4
mortron diyAudio Member Join Date: Aug 2011 Location: Town	maybe your measure is right, and the way they measure the FR of headphones is flawed?

15th February 2013, 05:10 AM	#5
jcx diyAudio Member Join Date: Feb 2003 Location: ..	Headphone Measurement Procedures - Frequency Response \| InnerFidelity Expert Tests InnerFidelity's Headphone Measurement Repeatability and Reproducibility Page 2 \| InnerFidelity

15th February 2013, 05:16 AM	#6
cotdt diyAudio Member Join Date: Oct 2005	Sure, but it's not so much about reproducibility as much as what the heck the frequency response even means. What kind of response is ideal for headphones after the sound goes reflected through a dummy head?

16th February 2013, 09:03 AM	#9
xnor diyAudio Member Join Date: Oct 2010	You only get that much bass as in the IF graph if you have perfect seal. When I tested the SRH440 I also felt that there was a strong roll-off below 300 Hz and that they were a bit bright for my liking.