Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

Measurement technology
Measurement technology
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
Reply
 
Thread Tools Search this Thread
Old 13th September 2014, 11:49 PM   #131
bwaslo is offline bwaslo  United States
Old guy with soldering iron
diyAudio Member
 
bwaslo's Avatar
 
Join Date: May 2006
Location: Portland, Oregon!
Measurement technology
Quote:
Originally Posted by _Wim_ View Post
Ok, I thought with “locked” you actually meant "Keep in/out stream active", and this is what I have done.
Just a caveat -- don't assume all sound cards use the same reference clock for record and playback. There are some that use an entirely different source for playback as for record, so you won't get them to lock because of the (even quite small) differences in sample rate. For instance, audio interfaces based on the Texas Instruments PCM290x series chips use the USB clock for recording and a separate crystal for playback. Some popular devices that use these parts are the iMic and Behrenger USB interfaces). I've seen some motherboards also to have this problem (not sure what chips they had inside). The way to tell is to "lock" an impulse response from a speaker and leave things set up while continuing to watch successive impulses. Mismatched rates will cause the IR to drift forward or back, usually visible over 10 seconds or so (but some on every single cycle).
__________________
[W9MJE] Horn spreadsheet SynergyCalc/; SmallSyns SmallSyns;
Crossover design Xsim; Depot diffusor super-easy diffusors
  Reply With Quote
Old 14th September 2014, 12:34 AM   #132
gedlee is offline gedlee  United States
diyAudio Member
 
gedlee's Avatar
 
Join Date: Dec 2004
Location: Novi, Michigan
Quote:
Originally Posted by bolserst View Post
Agreed. Just realize that most people(me included) don’t understand the nuances of how your process works, where as you have been working at refining it for years geared toward your application and a particular set of measurement angles. It is only natural to be a little skeptical and want to understand/test the limits and tolerances to see if something unexpected results.
Understood and accepted.

I will try and teach the skeptics what I know as long as the discussion is civil. I will not deal with impolite. It is very true that this technique is new and no one else does it, but that does not make it wrong. People are always skeptical of the new, that's just nature, but the true professional is always looking for the new that benefits his work. Not doing so causes stagnation - which could be the hash tag for "loudspeaker technology" - #stagnant_loudspeaker_technology.

Horns had not changed in decades till I showed that they could be better.
__________________
Earl Geddes Gedlee Website
  Reply With Quote
Old 14th September 2014, 12:39 AM   #133
gedlee is offline gedlee  United States
diyAudio Member
 
gedlee's Avatar
 
Join Date: Dec 2004
Location: Novi, Michigan
Quote:
Originally Posted by bwaslo View Post
Just a caveat -- don't assume all sound cards use the same reference clock for record and playback. There are some that use an entirely different source for playback as for record, so you won't get them to lock because of the (even quite small) differences in sample rate. For instance, audio interfaces based on the Texas Instruments PCM290x series chips use the USB clock for recording and a separate crystal for playback. Some popular devices that use these parts are the iMic and Behrenger USB interfaces). I've seen some motherboards also to have this problem (not sure what chips they had inside). The way to tell is to "lock" an impulse response from a speaker and leave things set up while continuing to watch successive impulses. Mismatched rates will cause the IR to drift forward or back, usually visible over 10 seconds or so (but some on every single cycle).
Bill, are we talking ms. here or micro-s? A sample variance or two is not going to be a big issue. In the above case where it was an issue it was several ms. because Holm locked on to the wrong impulse. That is an issue.

At any rate, I was unaware of what you are saying, but have not seen it in any of my work to date.
__________________
Earl Geddes Gedlee Website
  Reply With Quote
Old 14th September 2014, 03:59 AM   #134
natehansen66 is offline natehansen66  United States
diyAudio Member
 
Join Date: Mar 2011
Location: Rochester, MN
Outdoor measurements complete. I'm glad I did that, there's quite a bit below 1khz that I was missing in my shorter gate inside data. That said, I was having trouble locking the time in Holm. I think it's because I use a separate DAC and mic preamp, and I have audio software that grabs the audio that Holm sends to the default Windows driver for processing. I manually had it detect the largest peak for each measurement, and offset by hand the last few to get them inline.

You said you don't want them gated, so when I export the full data set from Holm is the gate included? I exported each impulse as a .wav file so I could look at all the data in REW, and my gate was included there.

I measured the Pioneer bookshelf, and honestly I think it's broke

Not sure if I want to submit this:

Click the image to open in full size.

That 9khz notch is evident until about 20° off axis.
  Reply With Quote
Old 14th September 2014, 04:04 AM   #135
Pallas is offline Pallas  Pakistan
diyAudio Member
 
Join Date: Aug 2004
Quote:
Originally Posted by gedlee View Post
Users would be wise to carefully read the procedure. the problems caused by not locking the time base can be significant.
Does setting up a loopback correction give the same results as a time-lock in Holm?

I'm not familiar with Holm, as it's a Windows program. I use FuzzMeasure on my Mac for audio measurements. It doesn't do time lock that I could see, but is designed to use loopback correction.
__________________
Serious Audio Blog
  Reply With Quote
Old 14th September 2014, 12:57 PM   #136
Zvu is offline Zvu  Serbia
diyAudio Member
 
Zvu's Avatar
 
Join Date: Jan 2009
Location: Belgrade
Quote:
Originally Posted by natehansen66 View Post
Outdoor measurements complete. I'm glad I did that, there's quite a bit below 1khz that I was missing in my shorter gate inside data. That said, I was having trouble locking the time in Holm. I think it's because I use a separate DAC and mic preamp, and I have audio software that grabs the audio that Holm sends to the default Windows driver for processing. I manually had it detect the largest peak for each measurement, and offset by hand the last few to get them inline.

You said you don't want them gated, so when I export the full data set from Holm is the gate included? I exported each impulse as a .wav file so I could look at all the data in REW, and my gate was included there.

I measured the Pioneer bookshelf, and honestly I think it's broke

Not sure if I want to submit this:

Click the image to open in full size.

That 9khz notch is evident until about 20° off axis.
Now that's what i call measurement. I would like to see more people giving their measurements in 1dB or 2dB grid but that's really rare. We see 5dB or even 10dB gridline. What am i going to see from that other than really big issues that pops up. I always give my measurements in 2dB gridlines so if anything is wrong it becomes obvious.

Other than that notch the FR of that loudspeaker looks good. Could it be malfunction ? I doubt that it is intended to play like that. Thanks for this measurement
__________________
Tesla; George Carlin
  Reply With Quote
Old 14th September 2014, 01:09 PM   #137
natehansen66 is offline natehansen66  United States
diyAudio Member
 
Join Date: Mar 2011
Location: Rochester, MN
Quote:
Originally Posted by Zvu View Post
Now that's what i call measurement. I would like to see more people giving their measurements in 1dB or 2dB grid but that's really rare. We see 5dB or even 10dB gridline. What am i going to see from that other than really big issues that pops up. I always give my measurements in 2dB gridlines so if anything is wrong it becomes obvious.

Other than that notch the FR of that loudspeaker looks good. Could it be malfunction ? I doubt that it is intended to play like that. Thanks for this measurement
Actually that's how Holm defaults the grid lines. I typically use ARTA or REW and I use 5dB lines there. It's not the grid lines that's important it's the scale. A 40dB scale is the same look at the data whether you have 20dB grid or a 1dB grid.

I should have tested the other speaker to see if it behaves in the same way but my neighbor had just got home and was firing up his grill. His grill is about 10' from my deck so I didn't want to bug him.
  Reply With Quote
Old 14th September 2014, 04:56 PM   #138
bwaslo is offline bwaslo  United States
Old guy with soldering iron
diyAudio Member
 
bwaslo's Avatar
 
Join Date: May 2006
Location: Portland, Oregon!
Measurement technology
Quote:
Originally Posted by gedlee View Post
Bill, are we talking ms. here or micro-s? A sample variance or two is not going to be a big issue. In the above case where it was an issue it was several ms. because Holm locked on to the wrong impulse. That is an issue.

At any rate, I was unaware of what you are saying, but have not seen it in any of my work to date.
Earl,

No, this is a long term very slightly different sample rate, with the recording not synchronous with the playback and the integrated error (apparent delay) increasing over time. If you want to see an example, pick up one of the inexpensive Behrenger USB audio interfaces and try with that. I learned that lesson the hard way with calls from customers who were unable to get synchronous averaging to work with Praxis when configuring a portable laptop based measuring system using those kinds of USB devices.
__________________
[W9MJE] Horn spreadsheet SynergyCalc/; SmallSyns SmallSyns;
Crossover design Xsim; Depot diffusor super-easy diffusors
  Reply With Quote
Old 14th September 2014, 06:06 PM   #139
gedlee is offline gedlee  United States
diyAudio Member
 
gedlee's Avatar
 
Join Date: Dec 2004
Location: Novi, Michigan
Quote:
Originally Posted by natehansen66 View Post
Outdoor measurements complete. I'm glad I did that, there's quite a bit below 1khz that I was missing in my shorter gate inside data. That said, I was having trouble locking the time in Holm. I think it's because I use a separate DAC and mic preamp, and I have audio software that grabs the audio that Holm sends to the default Windows driver for processing. I manually had it detect the largest peak for each measurement, and offset by hand the last few to get them inline.
That is not actually correct, Holm will do that as well. As the box is rotated the impulse do and should arrive later and later in time.

I never realized that this was such a big issue. I have used numerous different computers over the years and none has had these problems.

Quote:
Originally Posted by Pallas View Post
Does setting up a loopback correction give the same results as a time-lock in Holm?

I'm not familiar with Holm, as it's a Windows program. I use FuzzMeasure on my Mac for audio measurements. It doesn't do time lock that I could see, but is designed to use loopback correction.
I do not know if a loopback would work. Seems to me that will only correct for frequency response and lag errors in your system. It will not lock the time base of each measurement.

Quote:
Originally Posted by bwaslo View Post
Earl,

No, this is a long term very slightly different sample rate, with the recording not synchronous with the playback and the integrated error (apparent delay) increasing over time. If you want to see an example, pick up one of the inexpensive Behrenger USB audio interfaces and try with that. I learned that lesson the hard way with calls from customers who were unable to get synchronous averaging to work with Praxis when configuring a portable laptop based measuring system using those kinds of USB devices.
Hi Bill - yes I can see where this would null out any advantage of synchronous averaging. I wish that Holm would do synchronous averaging. But it does not seem to be an issue that would affect what I do. At least not to any great extent.

Can Praxis export an array of impulse responses as a single file? Could it do this in a similar format to Holm?

I sound like a big fan of Holm, but that is just because it works. When I wanted to try the Farina technique Holm was free and it did this technique very well. Not many other software did. Since I just wanted to try the technique and I always did any post processing myself, I did not want to pay just to try it. Holm fit the bill and in the beginning was well supported. The single array export was done at my request. But now Holm is not supported at all and that worries me. I would rather use a software that is supported and I know will not just go away some day. But the software does have to export unmodified impulse responses as an array in a single file or it is just too hard to deal with.

An example export is available on my website.

I can't see ARTA or REW ever doing this.
__________________
Earl Geddes Gedlee Website
  Reply With Quote
Old 14th September 2014, 06:53 PM   #140
natehansen66 is offline natehansen66  United States
diyAudio Member
 
Join Date: Mar 2011
Location: Rochester, MN
Quote:
Originally Posted by gedlee View Post
But the software does have to export unmodified impulse responses as an array in a single file or it is just too hard to deal with.

An example export is available on my website.

I can't see ARTA or REW ever doing this.
That's my main problem with ARTA. It's a pain saving each individual impulse. REW can but unfortunately it's only for its own file type.

When I get a chance in the next couple days (maybe today) I'll send you my data.
  Reply With Quote

Reply


Measurement technologyHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
New technology tvrgeek Planars & Exotics 1 22nd March 2013 07:46 AM
New technology from the UK Ed Holland Planars & Exotics 3 6th April 2009 05:56 PM
Sound Technology 1700B Distortion Measurement Systems for sale nwaero Swap Meet 0 15th December 2006 08:09 AM
new technology Chris8sirhC Planars & Exotics 10 4th May 2005 01:47 PM
Another new technology... Kongen Multi-Way 0 2nd November 2003 06:12 PM


New To Site? Need Help?

All times are GMT. The time now is 07:43 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki