Complaint thread about speaker measurements

[amirm]

* it might be interesting to post the ambient temperature and relative humidity at the time of the test. Cone materials (paper, treated or not, come to mind), cabinet materials (wood, MDF, plastic, etc.), box stuffing (polyfill, rockwool, fiberglass, etc.) and even motors (voice coil heat generation), can behave a bit differently in varying climates. Not mission critical, just interesting.

I was reporting the temp and such but people got concerned that it made a difference so I stopped. So far I have found no correlation between environmental factors and measurements. Speakers have wide operating range and they better work as well throughout that range.

Nothing in measurements should be examined at a level deep enough to expose such factors.

 

[amirm]

* perhaps one measurement can be included based on a mfrs. recommendation of speaker placement and accompanying listener position. IOW, if a mfr. specifies the optimal listening position is 15 degrees off-axis, it would be a handy data point to know what you'd measured according to the mfrs. recommendation

This data is available in the graph (vertical and horizontal directivity) and in full exports of the same. It is at 10 degree increments though as the standard stipulates.

 

[RickS]

Looks like speaker boxes we were building in our teens

very much like my first 3-way build about 30 years ago! Not likely they are controlling much cabinet resonance without any crossbracing. Also, the woofer is secured with only 4 bolts?

otoh, the crossover does appear to be relatively substantial. Do they make any claims about “tuning” individual speakers?

 

[dense]

I think the improvements in the measurement conditions can be suggested, but Amir should decide the rest.
There can always be more measurements, but it will become too much burden for a single person unless ASR becomes a Consumer Report of audios.
I have been working in the electronics industry for a while, and measuring something is a time consuming and sometimes very stressful process. Although it's good to see some of the detailed info., tolerances in the products and test conditions can not be fully trusted with just one sample.
IMHO, SPL, THD, impedance, CSD measurements in full/semi anechoic chamber, and subjective listening is enough data for purchasing a speaker. Anything more is for engineering and it should be a paid service.

 

[RayDunzl]

So far I have found no correlation between environmental factors and measurements.

My 'stats might dim a little in the high frequencies with excessively elevated humidity.

PS: I hope you don't miss an opportunity to spin a panel.

 

[Frank Dernie]

I am not sure I would go as far as ‘substantial’

measuring a few Harbeths the treble seems to diminish very quickly with distance.
keith

It depends on what you know about resonance and damping.
To what would you attribute a more than usual reduction of treble with distance?

 

[Frank Dernie]

Not likely they are controlling much cabinet resonance without any crossbracing.

Crossbracing doesn't control resonance.
Common bit of static thinking BS seen around the internet.
Non-facts are commonly touted on the internet.

 

[Frank Dernie]

Looks like speaker boxes we were building in our teens

Actually unlikely but t may look it to the un-knowledgeable.

 

[Purité Audio]

It depends on what you know about resonance and damping.
To what would you attribute a more than usual reduction of treble with distance?

If you say on bitumen pad is ‘substantial’ who am I to argue.
re treble just measuring customers speakers, and then comparing them to the speaker I am demonstrating, I ask customers if possible keep their current system up.
I remember comparing Kiis and Harbeth, both treble outputs similar at one metre very different at two, both measurements and speakers on axis.
Keith

 

[Frank Dernie]

If you say on bitumen pad is ‘substantial’ who am I to argue.
re treble just measuring customers speakers, and then comparing them to the speaker I am demonstrating, I ask customers if possible keep their current system up.
I remember comparing Kiis and Harbeth, both treble outputs similar at one metre very different at two, both measurements and speakers on axis.
Keith

The thickness and size of the bitumen panel is related to the thickness of the panel and the Q of the resonance you are aiming for. A good engineer will have a simple solution. Complex solutions normally mean marketing lead BS rather than engineering.
The fall of of sound with distance is related to physics not speaker manufacturer. I can see that the radiation pattern may make a difference but not that big unless horn loaded.
Mystified why the laws of physics may apply differently to Harbeth than Kii????

 

[pma]

Several screws to hold rear cover - real SOTA. Maybe some crackling added makes the part of the remarkable Harbeth sound. Fairy tales about careful handling of cabinet's resonances? Or another marketing department story?

 

[Purité Audio]

I agree re the tweeter, the Harbeth’s were already more rolled off ,in terms of treble than the Kiis, probably my measuring error.
I am not particularly criticising the Harbeths they are ok Perhaps a little on the expensive side for what’s in them, but that applies to many audio products.
Keith

 

[Frank Dernie]

Several screws to hold rear cover - real SOTA. Maybe some crackling added makes the part of the remarkable Harbeth sound. Fairy tales about careful handling of cabinet's resonances? Or another marketing department story?

Bolted junctions add damping. The number and tightness of the screws effects the damping.
Bolted together cabinets have more damping than glued ones but are more expensive to make and assemble.
Read the BBC papers. You may not agree, and whether than means anything depends on how much you know about resonance and dynamics, but don't assume it hasn't been researched, because it has, for years.
Cabinet radiation is part of how a speaker sounds and is not separated from the radiation from the drivers in measurements, as far as I understand from a specialist, meaning the sort of speaker measurements we see do not correctly quantify the cabinet (ie spurious) contribution to the sound field being measured.

 

[sergeauckland]

Several screws to hold rear cover - real SOTA. Maybe some crackling added makes the part of the remarkable Harbeth sound. Fairy tales about careful handling of cabinet's resonances? Or another marketing department story?

The point of the BBC/Harbeth cabinet is that front and rear panels should NOT be rigidly fixed, so as to create a lossy structure, like a cracked bell. Hence only a few screws and no glue.

S.

 

[Frank Dernie]

I agree re the tweeter, the Harbeth’s were already more rolled off ,in terms of treble than the Kiis, probably my measuring error.
I am not particularly criticising the Harbeths they are ok Perhaps a little on the expensive side for what’s in them, but that applies to many audio products.
Keith

I think the manufacturing cost of both the cabinets and main drivers in Harbeths which are custom made in fairly small numbers are probably quite high.
The tweeters aren't.

 

[pma]

I like Wilson Audio cabinets, not Harbeth. And I also do not like Sugden amplifiers. We may have different preferences, I agree.

 

[Purité Audio]

Bolted junctions add damping. The number and tightness of the screws effects the damping.
Bolted together cabinets have more damping than glued ones but are more expensive to make and assemble.
Read the BBC papers. You may not agree, and whether than means anything depends on how much you know about resonance and dynamics, but don't assume it hasn't been researched, because it has, for years.
Cabinet radiation is part of how a speaker sounds and is not separated from the radiation from the drivers in measurements, as far as I understand from a specialist, meaning the sort of speaker measurements we see do not correctly quantify the cabinet (ie spurious) contribution to the sound field being measured.

Surely it would be straightforward to seperate the driver and enclosures contribution?
Keith

 

[Frank Dernie]

Surely it would be straightforward to seperate the driver and enclosures contribution?
Keith

How?
It can be calculated from surface vibration measurement, the engineer I know who designs speakers for OEM clients told me he does it this way.
How would a measurement microphone (or your ear) know which bit of the air pressure wibbling it was picking up was coming from the drive unit and which proportion was the spurious bit coming from the cabinet?
Maybe somebody has come up with a way of doing it but If so I haven't seen it published yet.

 

[Frank Dernie]

I like Wilson Audio cabinets, not Harbeth. And I also do not like Sugden amplifiers. We may have different preferences, I agree.

Wilson achieve by brute force and cash what Harbeth achieve by clever engineering.
I quite like some of the Wilson speakers too.
I don't write much about electronics since it is not my expertise. Resonance and vibration is.
A lot of complete bollox is written about it.
Everything resonates, making it stiffer increases the frequency, making it heavier lowers it.

 

[RickS]

Crossbracing doesn't control resonance.
Common bit of static thinking BS seen around the internet.
Non-facts are commonly touted on the internet.

Unless you prefer changing over controlling, not sure what you are going after?

This non-fact about cross-bracing is published in the Loudspeaker Design Cookbook and is widely applied in better cabinet designs.