Future testing, data analysis and presentation - Page 2 - diyAudio
Go Back   Home > Forums > Commercial Sector > Manufacturers > Markaudio

Markaudio Designers and builders of audiophile grade drivers

Reply
 
Thread Tools Search this Thread
Old 26th February 2014, 05:26 PM   #11
diyAudio Member
 
Bob Brines's Avatar
 
Join Date: Jan 2003
Location: Hot Spring Village AR
Some random thoughts on the subject:

I personally have come to the conclusion the home measurements -- decidedly non-anechoic -- are at best a gross approximation of a driver's actual performance. Gating does remove reflections, but it also smooths the top end (more later), and totally destroys the bottom end. Nearfield driver/port measurements of the bottom end look real nice, but are optimistic as to real bass output and I can’t ever seem to make them agree with farfield measurements in my shop or listening room.

I have the Smith and Larson Woofer Tester Pro, which for $1000US does a pretty good job. Quite flexible, and has a lot more options than needed in a home environment. Lately, I have been doing quick-and-dirty FR measurements with REW and a UMM-6 mic. For an $80 investment, this is the way DIY’ers should go. (I am not sure how well REW does of IMP measurements. Never tried it.)

I have published raw FR plots and smoothed plots. There is no way that anyone will convince the geeks that all of those little wiggles in a raw plot are meaningless. The brain will repossess the data and smooth it to something like 1/6 octave. Therefore, the way I display the data on my recent products is to present the raw 16k FFT is gray overlaid with the 1/6 octave smoothed in black.

Carrying this above though one step farther, the reality is that full-range drivers are never listened to on axis. Usually around 15* off axis. (BWT, I personally find that toed out 15* gives the best compromise of imaging, sound stage and width of the sweet spot. That means the speakers are square with the room and gives the best visual presentation. But I digress….) What I am thinking is that I will present the 0* in red and the 15* in a thicker black.

And given the above, the graph should display 0*, 15*, 30*, 45* and maybe 60*. The 100dB vertical scale is industry standard and has the singular advantage of allowing the FR and IMP traces on the same graph. A 50-100dB scale would be ideal for the DIY’ers. A distortion graph would be nice and satisfy the curiosity of DIY’ers, although I doubt that one it ten really understands what they are looking at (“Look at that huge 3rd order spike 60dB down!”) Same goes for a waterfall plot. IMO a waterfall done in other than a true anechoic chamber is worthless. If the plot is not gated, you are looking at the room, not the driver, and if you gate, any time outside of the gate (~4ms in a standard 8’ high room) is meaningless.

Speaking of anechoic chambers, you mount the DUT in a box, don’t you? If so, what you measure is representative of what a builder will get. However, I am thinking that the 10kHz+ notch off axis may be edge defraction. The plot I displayed on the “other” thread certainly is subject to edge defraction and is very similar to your A7.3 trace. May I suggest that you use an IEC baffle with a box behind it. If indeed defraction is a problem, this will solve it. I think Seas does it this way
Enough for now. More thoughts as they come up

Bob
__________________
Brines Acoustics Web Site
  Reply With Quote
Old 26th February 2014, 05:31 PM   #12
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
I see some of the same resonance peaks appearing in the bare mounted vs the mounted configuration in the TC9FD. The bare mount just confirms that the inherent resonance frequency is there because screw mount pressure along the bezel does not necessarily translate to more damping or reinforcement along the fundamental vibrational modes in question.
  Reply With Quote
Old 26th February 2014, 05:38 PM   #13
diyAudio Member
 
markaudio's Avatar
 
Join Date: Nov 2008
Quote:
Originally Posted by xrk971 View Post
Do you mean photos of the bare driver ultra-nearfield test setup?
Pics of you whole set up (room, baffles, gear etc). It all helps to illustrate a home set-up. We need to get some sort of consensus on home testing so future beta testers can operate with some agreed uniformity (we hope).

Thanks
Mark.
  Reply With Quote
Old 26th February 2014, 05:46 PM   #14
frugal-phile(tm)
diyAudio Moderator
 
planet10's Avatar
 
Join Date: Oct 2001
Location: Victoria, BC, NA, Sol III
Blog Entries: 5
I would suggest a data set that cooresponds with at least a subset of the presentation suggested by Floyd Toole in "Sound Reproduction". He has a full set of 360 degree data for loudspeakers in their boxes that is probably not suitable, but later they describe what they consider an important subset of that.

As to distortion measures, i tend towards Geddes whose scentific research that shows that distortion as currently measured in not very useful so i'm not all that concerened with them. Most also happens low down where room & box dominate.

Other measures that extend the things that are measured wouldalsobenice, but since these test have yet, to my knowledge, been designed, and the R&D and tooks needed to execute them are likely beyond a small manufacturer.

dave
__________________
community sites t-linespeakers.org, frugal-horn.com, frugal-phile.com ........ commercial site planet10-HiFi
p10-hifi forum here at diyA

Last edited by markaudio; 26th February 2014 at 10:27 PM. Reason: conflict resolution
  Reply With Quote
Old 26th February 2014, 05:48 PM   #15
dewardh is offline dewardh  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Berkeley, CA
Quote:
Originally Posted by Scottmoose View Post
-One possibility: use a similar range to that employed over on Zaph Audio, since you don't find many complaining about that.

-0, 10, 15 & 30 degrees axis measures off would be very useful. I'd question if more than 45 degrees off is of any value for wideband drivers. I can see the value of out to, say, 60 for tweeters / midbass units since it has some value for XO design, but single driver types? Rather less.
I don't see "range" as an issue, but rather "display scaling" . . . that's where the "confusion" comes in. These published charts are commonly used as comparison tools . . . for that one needs to be able to (mentally) overlay them and have the scales coincide. It then doesn't matter if one chart is "short" (displays 60-110dB) and the other is "tall" (displays 0-110 dB and includes the distortion curves at the bottom). We also have to remember "worldwide market" . . . Zaph is clearly an American phenomenon (despite the Australian software), so for European and Asian markets charts that mimic the more common products in those markets might be more appropriate. I don't see people complaining about the charts Seas provides either . . .

The other potential "use" is, as you note, crossover design . . . and while it's true that far off-axis behavior is entirely baffle determined at any plausible crossover frequency it's still at least plausibly helpful to know what the driver's "internal crossover" is doing . . . so I'd vote to include 60 degrees as the metric of "off axis" performance with 0/15/30 describing behavior in the "listening window.
  Reply With Quote
Old 26th February 2014, 07:06 PM   #16
dewardh is offline dewardh  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Berkeley, CA
Quote:
Originally Posted by Bob Brines View Post
If you have problems comparing FR graphs in different scales, bring them up in separate windows and expand the smaller until they are both visually the same size. Do it all the time.
It's not so much a question of size, as the relative horizontal/vertical axis ratios. Getting those to overlap is a bit more difficult. It's complicated further by the MA charts extending out to 40kHz (totally useless IMO and no one else does it), which leads to "intuitive" misinterpretation even after you "intellectually" recognize the difference.
  Reply With Quote
Old 26th February 2014, 07:35 PM   #17
diyAudio Member
 
Scottmoose's Avatar
 
Join Date: Mar 2005
Location: UK
Quote:
Originally Posted by dewardh View Post
I don't see "range" as an issue, but rather "display scaling" . . .
Just so (but this is largely a matter linguistic semantics since everybody knows what was meant).

Quote:
Zaph is clearly an American phenomenon (despite the Australian software), so for European and Asian markets charts that mimic the more common products in those markets might be more appropriate. I don't see people complaining about the charts Seas provides either . . .
No argument re the quality of Seas's data. I regularly use their products, including in three commercial designs, and it's generally excellent. Their reputation was not gained without reason. However, they're primarily a 'conventional' drive unit manufacturer. Couple of widebanders, which are not big sellers, and I would suggest that copying their data presentation format is not likely to be a bright idea in presentation terms for a manufacturer who specialises in this type of driver.

Re the Zaph site, I disagree. I'm a European (UK), and regularly refer to John's data as a useful additional reference, so it's far from being only known to those in the US. Many of those I know in the UK & on the continent are also well aware of it. I'm not suggesting using SoundEasy (excellent though it is for what it is); simply suggesting that similar chart ranges / scale / whatever you wish to call it are an option that might be worth thinking about. YMMV as always of course.
__________________
Community sites www.frugal-horn.com http://frugal-phile.com/

Last edited by Scottmoose; 26th February 2014 at 07:41 PM.
  Reply With Quote
Old 26th February 2014, 07:55 PM   #18
chrisb is offline chrisb  Canada
diyAudio Member
 
Join Date: Aug 2002
Location: victoria BC
Quote:
Originally Posted by dewardh View Post
It's not so much a question of size, as the relative horizontal/vertical axis ratios. Getting those to overlap is a bit more difficult. It's complicated further by the MA charts extending out to 40kHz (totally useless IMO and no one else does it), which leads to "intuitive" misinterpretation even after you "intellectually" recognize the difference.
Well, a lot of cone drivers of the size range in question simply don't extend much output that far - some domes and ribbons, etc yes and I think you'll see that on their graphs (graphs for some bullet horn supertweeters extend to 50kHz)

I would concur that absent a peer reviewed, controlled double blind ,etc study as to the audibility and subjective difference playing available source material between a speaker system (not headphones) that extends relatively cleanly to past 40kHz and an identical system filtered without artifacts to only 20kHz, the importance of striving for the 11th octave is lost on me.
__________________
now on sabbatical
  Reply With Quote
Old 26th February 2014, 07:58 PM   #19
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Quote:
Originally Posted by markaudio View Post
Pics of you whole set up (room, baffles, gear etc). It all helps to illustrate a home set-up. We need to get some sort of consensus on home testing so future beta testers can operate with some agreed uniformity (we hope).

Thanks
Mark.
Lab space:
Click the image to open in full size.

Near field measurement of a sub:
Click the image to open in full size.
  Reply With Quote
Old 26th February 2014, 08:12 PM   #20
diyAudio Member
 
Scottmoose's Avatar
 
Join Date: Mar 2005
Location: UK
That's a lot of foam-core.
__________________
Community sites www.frugal-horn.com http://frugal-phile.com/
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
The future for driver data Electron Markaudio 15 3rd March 2010 10:34 PM
New testing data ucla88 Multi-Way 7 5th June 2008 01:58 AM
The future of Internet speed and data transfer Netlist Everything Else 3 3rd January 2007 01:42 AM
graphical data analysis software bocka Everything Else 3 19th July 2005 05:56 PM
Software Analysis Testing RobPhill33 Everything Else 9 24th February 2003 10:37 AM


New To Site? Need Help?

All times are GMT. The time now is 07:33 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2