Is there a cheap, good, and fast way to measure headhpones for parametric equalization?

[Deleted member 58865]

It doesn't have to be perfect, but it needs to be good enough.

I'm not looking for a way to do it perfectly.

Notice the emphasis on "good". I think "good" can be made cheap and fast in most cases. I didn't say "excellent" or "perfect".

 

[staticV3]

What do you actually want to do?
Correct the frequency response based on measurements?
Or check the frequency response difference between stock and EQed?

 

[isostasy]

What do you actually want to do?
Correct the frequency response based on measurements?
Or check the frequency response difference between stock and EQed?

Far too much at once, without taking on board any recommendations or advice anyone gives, and without contributing anything back to the forum.

 

[Plumbus the Wise]

Yes, go onto reddit and look to see if Oratory1990 has measured them

 

[Deleted member 58865]

What do you actually want to do?
Correct the frequency response based on measurements?
Or check the frequency response difference between stock and EQed?

Simple. Parametric equalization.

 

[kemmler3D]

There isn't really a 'cheap' way because the good headphone measuring setups are pretty expensive. I have the minidsp EARS and my measurements on there don't really resemble the measurements from high-end setups very well.

However, once you have a decent measurement jig (like GRAS or something) then it's both fast and good from there on. You can use (say) REW and AutoEq, or even just REW, I think.

As they say, it's the iron triangle: Fast, Cheap, Good, pick two.

 

[Deleted member 58865]

There isn't really a 'cheap' way because the good headphone measuring setups are pretty expensive.

In theory, something can be made cheap, good, and fast if it really gets people's attention.
Think Raspberry Pi. It is cheap, good, and fast to set up because it has recevied a lot of attention.
I guess it's not the case now.

 

[ClosDeLaRoche]

AutoEQ in Peace or ToneBoosters MorphIt is a good place to start.

To tune my Ether Flow I used the InnerFidelity AutoEQ profile in Peace as a starting point, then smoothed the EQ profile a bit and listened to sweeps with a tone generator. I listen for peaks and dips with the tone generator and tweak the AutoEQ profile accordingly.

If that is too much work or too fiddly then MorphIt is pretty awesome. I think MorphIt is best used as a cheap add-on to the USB Audio Player Pro app for android. And of course, your particular headphone needs to be supported.

 

[kemmler3D]

if it really gets people's attention.

Yes, if you can scale production of headphone measurement jigs to a few million units per year I think you'd see both good and cheap options. As it is I think the market for professional-grade headphone measurement is closer to 100 per year than 1,000,000, so I wouldn't hold my breath.

The RPi started out fast, decent and cheap, and became better over time because it attracted a lot of interest.

The MiniDSP EARS is okay for the money, but the problem is I wouldn't really trust it for absolute magnitude very much, which you do need for a good EQ setting. The EARS unit is good for telling how one headphone differs from another, I have found it good for that. But I would be much less confident in the measurements reflecting the actual SPL vs. neutral.

My advice is maybe buy one of those, and then use your ears to dial in the EQ settings using various listening techniques, i.e. sweeping bandpassed pink noise, tones, just listening to PN, etc.

Or shell out the $50K or whatever they want for a GRAS or B&K setup.

 

[Deleted member 58865]

AutoEQ in Peace or ToneBoosters MorphIt is a good place to start.

Is it possible to get good enough match with neutral sound signature with ears? If there is a way to get objective measurements with ears, then maybe.

 

[Deleted member 58865]

The EARS unit is good for telling how one headphone differs from another, I have found it good for that.

I think that's good enough if you want to use a different pair of pads.

 

[moosso]

Solutions depends on what you want:

Very expensive (50K~100K)

• GRAS 43AG, GRAS 45CA, GRAS 45BC
• B&K 5128

For serious people want to design their own headphone or doing review.
These measurement rigs are designed to match most people's HRTF. However, variance still exist.

Fair (around $300)

• MiniDSP EARS
• Taobao Clone IEC711 Rig
  • You can get artifical pinna from the same store, this is more close to GRAS 43AG than EARS

Less accurate than industry standard setups, but good enough for casual measurements.
From my experience <50Hz and >6KHz are mostly unreliable.

Cheap (around $100)

• MS-TFB-2
• Roland CS10-EM
• Sony ECM-TL3 ($20~$40 depends on your location)

These can be used to measure in ear response, which is good for people who want to adjust sound only for their pair.
The downside is it can't be compensated to GRAS or BK standard, which you can't EQ the measurement to a target based on GRAS or BK.

Free

• AutoEQ
• Tone Generator

AutoEQ can be a good start point though it doesn't count for unit variance and earpad difference.
Tone generator can be used to check the actual in ear FR (works better than any measurement rig), though it's diffcult to adjust because it won't tell you the exact gain and bandwidth you need.

 

[kemmler3D]

I think that's good enough if you want to use a different pair of pads.

Yeah, I think it will tell you with reasonable accuracy what the impact of swapping pads would be. You can probably arrive at a good guess as to whether the pads add +2 or +4dB at 80hz, for example. In the past I would tweak the global EQ on some cans and re-measure on the EARS, usually the difference measured is very close to what I changed on the EQ. So as far as that goes, they aren't bad.

The problem is that the overall measured response may not be very close to what a person actually hears, or what you'd measure on a high-end rig. If that doesn't matter to you then the MiniDSP setup is pretty cheap, good, and fast.

 

[MayaTlab]

• Taobao Clone IEC711 Rig
  • You can get artifical pinna from the same store, this is more close to GRAS 43AG than EARS

Sounds Good Store has started selling an updated pinna that is closer to GRAS' anthropomorphic pinna :

https://www.aliexpress.com/item/1005005260834672.html?spm=a2g0o.order_list.order_list_main.5.350e1802h3rBkc

It's already been received by Mad Economist. Comparison with a real GRAS pinna are pending but apparently the silicone isn't quite as soft as the original (but still softer than the KB0065 clone pinna).

There's also a complete mount :

https://www.aliexpress.com/item/1005005282002642.html?spm=a2g0o.store_pc_groupList.8148356.1.56e83bd0ZTf18b&pdp_npi=2%40dis%21EUR%21%E2%82%AC%20629%2C08%21%E2%82%AC%20629%2C08%21%21%21%21%21%4021038ede16782578955171726e10c7%2112000032477656645%21sh

For this one I have some concerns about how the pinna is mounted on the plate (the lip around it isn't covered by the plate and it protrudes).

For over-ears (heck even in-ears to a lesser degree) I'm tempted to think, based on csglinux's tour, and my own results with IEMs with a fairly low sample variation, that the clone couplers themselves should be adequate up to 8kHz or so, perhaps even higher, unless the one you get is a really poor copy, provided you can compensate them (if they're decent you shouldn't have too much compensation to so, 1dB at most in that range). By adequate I mean "good enough that you have other, more significant problems to worry about".

 

[Deleted member 58865]

I think I have surveyed all available options.

My conclusion is that in most cases, it should be good enough to buy a pair of neutral-sounding headphones for which there are replacement stock pads made of good materials or ones for which there are durable third party pads that don't reduce sound quality significantly.

Additionally, if the headband material is fake leather, either see whether it's possible to replace it with a durable material like tech-flex tubing or to cover it with a third-party headband cover.

If you are still not satisfied, then you should start looking into headphone measurements. If you just want to make third party pads sound like stock pads, then MiniDSP EARS should be good enough. If your preference is more specific, you need to spend a lot more money.

 

[L5730]

Cheap = economy of scale.

The famous RPi SBC is a bad example to compare a headphone measuring device to. An RPi can be used for almost infinite projects, radio streamers, home automation, media servers, an actual useable PC, firewall, weather station etc, etc. What on this green earth can a headphone measuring system be used for, with exception of measuring headphones of course?
Thus it will never have a large enough community who are interested in investing in it.

That's not so say that something can't be made useful for a reasonable cost that does an adequate job, in a quick enough time frame. All terms in italics maybe wildly different depending on who is being asked. Heck even the GRAS system has tradeoffs.

If you want cheap, quick 'n' dirty, then shove a measurement mic (Behringer ECM8000) through a flat bit of foam and measure one earphone sealed. Will it be accurate? Nope, but it might point out some issues. I suppose one could then go on to take moulds of their own ears and make latex pinnae, but I don't have a clue what the interaction would be like with those kinds of mics. Would in-ear microphones be a better solution if one could make some plastic 'ears'?

Pads might not necessarily 'reduce sound quality'. They may change the tonality from stock. Whether that is to detriment or not is in the eye, or rather the ear of the beholder.

 

[Deleted member 58865]

Pads might not necessarily 'reduce sound quality'. They may change the tonality from stock. Whether that is to detriment or not is in the eye, or rather the ear of the beholder.

You are right. But, different pads produce different outcomes for different headphones. That's why you should pay more attention to details when you buy third party pads.

You are not going to just buy arbitrary pads and be satisfied. You may have preferences regarding various frequencies.

It's a good idea to study how size, shape, and material of pads affect sound signature.

 

[Deleted member 58865]

shove a measurement mic (Behringer ECM8000) through a flat bit of foam

I don't know what that would look like with headphones. Can you show me a picture?

 

[L5730]

I don't know what that would look like with headphones. Can you show me a picture?

Something like this but much crappier.

Ideas for Building a New Flat Plate Measurement Rig

I'm looking to build a new headphone measurement rig for purposes of modding and EQing. My old/existing one is a Dayton EMM-6 mounted flush through a damped aluminum plate because that was the easiest to build at the time. Since it uses a full size measurement mic it's way too wide to actually...

www.audiosciencereview.com