ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => Audio Measurement and Testing => Topic started by: boburtz on July 24, 2018, 03:41:04 PM

Title: Acceptable deviation between boxes?
Post by: boburtz on July 24, 2018, 03:41:04 PM
I measured transfer functions of 16 KF730 boxes individually, each with the exact same measurement conditions. I am curious as to what is an acceptable variance box to box for stuff like this. Is there a standard?  They average out to something close to the published EAW spec, but individually there are some variations that seem to be pretty large.  Does this look normal? If not, do you have any idea what could be causing this or a suggested solution?
Title: Re: Acceptable deviation between boxes?
Post by: Nathan Riddle on July 24, 2018, 03:49:33 PM
I measured transfer functions of 16 KF730 boxes individually, each with the exact same measurement conditions. I am curious as to what is an acceptable variance box to box for stuff like this. Is there a standard?  They average out to something close to the published EAW spec, but individually there are some variations that seem to be pretty large.  Does this look normal? If not, do you have any idea what could be causing this or a suggested solution?

I believe Ivan has spoke about individual components' tolerances are sometimes swings of 3dB. Could explain the 6dB discrepancy between two farthest deviated components?
Title: Re: Acceptable deviation between boxes?
Post by: Nathan Riddle on July 24, 2018, 03:51:49 PM
Oh and you've obviously been around via your post count, so I'm assuming the name is Bob Urtz and admins are fine? Threw me for a loop for a sec   ;D
Title: Re: Acceptable deviation between boxes?
Post by: David Sturzenbecher on July 24, 2018, 04:21:24 PM
I measured transfer functions of 16 KF730 boxes individually, each with the exact same measurement conditions. I am curious as to what is an acceptable variance box to box for stuff like this. Is there a standard?  They average out to something close to the published EAW spec, but individually there are some variations that seem to be pretty large.  Does this look normal? If not, do you have any idea what could be causing this or a suggested solution?

I would be interested to see the complex impedance curves of these boxes as well, so see if there is corresponding differences there too.


Sent from my iPhone using Tapatalk Pro
Title: Re: Acceptable deviation between boxes?
Post by: John Roberts {JR} on July 24, 2018, 05:01:28 PM
I measured transfer functions of 16 KF730 boxes individually, each with the exact same measurement conditions. I am curious as to what is an acceptable variance box to box for stuff like this. Is there a standard?  They average out to something close to the published EAW spec, but individually there are some variations that seem to be pretty large.  Does this look normal? If not, do you have any idea what could be causing this or a suggested solution?
It comes down to cost... When boxes are intended to be arrayed there is an ASSumption that they will match reasonably well to play together nicely.

I went through this last century with Peavey dealers being unable to convince them why the more expensive drivers required for boxes meant for being arrayed were justified over the otherwise similar boxes playing by themselves, where tolerance matching was not as important.

JR

PS: Of course I have no specific information about those exact boxes, just in general.

PPS: I am not the speaker guy here. I just know a little about a lot.
Title: Re: Acceptable deviation between boxes?
Post by: Don T. Williams on July 24, 2018, 06:36:09 PM
This is not specific to this model, but if these were purchased in different lots at different times there could be different brands and models of drivers.  How many versions of the KF850 were made?  Kudos to EAW for trying to keep improving the product.  Drivers do age, and not every box may go out on every show, so they may not have all seem the same use (or abuse).
Title: Re: Acceptable deviation between boxes?
Post by: Tim McCulloch on July 24, 2018, 11:33:39 PM
Oh and you've obviously been around via your post count, so I'm assuming the name is Bob Urtz and admins are fine? Threw me for a loop for a sec   ;D

Bob has been around for at least 2 iterations of the PSW forums....
Title: Re: Acceptable deviation between boxes?
Post by: Scott Holtzman on July 25, 2018, 01:45:58 AM
Man Nate you are going to shit yourself of Abdhul EQ shows up.

Sent from my VS996 using Tapatalk
Title: Re: Acceptable deviation between boxes?
Post by: John Roberts {JR} on July 25, 2018, 09:28:43 AM
Man Nate you are going to shit yourself of Abdhul EQ shows up.

Sent from my VS996 using Tapatalk
Careful if we repeat his name 3x the dessert wanderer may chime in.... 8) 

but that is not a bad thing to receive his wisdom from the sands of time.

JR
Title: Re: Acceptable deviation between boxes?
Post by: Tim McCulloch on July 25, 2018, 04:50:18 PM
Careful if we repeat his name 3x the dessert wanderer may chime in.... 8) 

but that is not a bad thing to receive his wisdom from the sands of time.

JR

It's about time for him to surface from the sandstorms, was Dec 2015 when he last checked in.
Title: Re: Acceptable deviation between boxes?
Post by: Matthias McCready on July 26, 2018, 04:25:04 AM
the dessert wanderer may chime in.... 8) 

JR

Does he come bearing icecreams and other sweet things?  ;D ;D ;D
Title: Re: Acceptable deviation between boxes?
Post by: boburtz on July 26, 2018, 05:55:09 PM
I would be interested to see the complex impedance curves of these boxes as well, so see if there is corresponding differences there too.
I'm using DATS for the impedance measurement, and it only allows 20 saved traces (10 biamp boxes) so I had to split it up into two groups. See anything alarming here?
Title: Re: Acceptable deviation between boxes?
Post by: Art Welter on July 27, 2018, 01:12:28 PM
See anything alarming here?
Bob,

+/- a coil winding (or two) can easily account for impedance minima/maxima variations. Though I'm not alarmed, the purple trace makes me curious, as the higher frequency peak around 2kHz (instead of around 1800 Hz) may indicate something is physically different in the cabinet, perhaps something in the HF horn.

Art
Title: Re: Acceptable deviation between boxes?
Post by: boburtz on July 27, 2018, 01:33:58 PM
Bob,

+/- a coil winding (or two) can easily account for impedance minima/maxima variations. Though I'm not alarmed, the purple trace makes me curious, as the higher frequency peak around 2kHz (instead of around 1800 Hz) may indicate something is physically different in the cabinet, perhaps something in the HF horn.

Art
Thanks Art. Any chance this could be caused by old diaphragms or an issue with the passive crossover? I am going to dismantle the box and see if there is anything physically obvious.
Title: Re: Acceptable deviation between boxes?
Post by: Art Welter on July 30, 2018, 06:17:45 PM
Thanks Art. Any chance this could be caused by old diaphragms or an issue with the passive crossover? I am going to dismantle the box and see if there is anything physically obvious.
I was not aware of the passive crossover in the path, either a diaphragm of a different impedance or slightly different passive crossover component values, or a combination of both could account for the differences.

Title: Re: Acceptable deviation between boxes?
Post by: boburtz on July 30, 2018, 07:28:29 PM
I was not aware of the passive crossover in the path, either a diaphragm of a different impedance or slightly different passive crossover component values, or a combination of both could account for the differences.
The diaphragms both measure identically when removed from the path and they look fine when removed from the magnet assembly. There is nothing about the horn that is physically different than the other boxes, no blockage or cracks or otherwise. I suspect something in the passive crossover network. Incidentally, the magnitude and phase of this particular box measures more closely to the EAW spec than some of the others. Do you suppose it falls within the realm of normal or is it something I should pursue further? As always, thank you for the time you take to answer my questions.
Title: Re: Acceptable deviation between boxes?
Post by: Matthew Knischewsky on July 30, 2018, 11:57:50 PM
I measured transfer functions of 16 KF730 boxes individually, each with the exact same measurement conditions. I am curious as to what is an acceptable variance box to box for stuff like this. Is there a standard?  They average out to something close to the published EAW spec, but individually there are some variations that seem to be pretty large.  Does this look normal? If not, do you have any idea what could be causing this or a suggested solution?

I did complete tear down and re-build of a 16 box 730 rig some 5-6 years ago. Your traces look (to me) to be very normal. The passive network in these speakers make these a bit of a bear to work on, not to mention each box has 2 drivers for each passband. Unless each driver is measured individually to get a baseline it becomes difficult to know for sure if it's any combination of the drivers, diaphragms, and the network thats contributing to a variance. This work is all time consuming, so how much time do you have? Are you the patient type?

The phase variance at the extreme top end is normal. That's minute variation in mic/cabinet placement. What's more important is what's happening at the MF/HF X-over point, 1k-3k or so. Phase shift here is going to have an impact on the magnitude response.

A note about the phase response- a lagging phase response might tell you something about a driver that you may not necessarily be able to hear - like a slight rub on a voice coil. On the 730s the original voice coil former was Nomex, a paper like material which tended swell up and rub over time. Later, the VC former was changed to Kapton, a high temperature plastic material which wouldn't expand over time/use.

Ultimately, you'll have to decide how much effort you're willing to expend for what returns. I was able to get most of "my" 730s within about 2dB of each other over most of their range, some a bit more like 3dB. But, my starting point I think was much worse- A thrifty person thought they were smarter than EAW's parts department. A long story for another day.
Title: Re: Acceptable deviation between boxes?
Post by: Art Welter on July 31, 2018, 03:11:19 PM
Do you suppose it falls within the realm of normal or is it something I should pursue further?
Well within the realm of normal.
Forgot to mention magnet strength in the list of things that can change with age, though the difference you see between units could have easily been there from day one.

As Matt said "This work is all time consuming, so how much time do you have?"

As long as no coils are dragging/buzzing, I'd call it "good to go".

Art