Reduced level when measuring drivers in parallel? - Home Theater Forum and Systems - HomeTheaterShack.com
Home Theater Shack Forums
Marantz OPPO Digital - Universal Network 3D Blu-ray Players - Buy Direct from the Manufacturer Chane Music & Cinema Madisound Parts Express: Excellent Source for DIY Speaker and Subwoofer Projects! Musicians Friend: Your one stop shop for microphones, mic amps and Behringer products. Value Electronics: A trusted retailer and custom installer that specializes in whole-house audio, home theater, Plasma, LCD, LED, HDTVs, Blu-ray and more. Genelec: Active Speakers and Speaker Systems Audio General, Inc.: Great deals on projectors, screens and electronics. Underwood HiFi - Excellent discounts on some awesome name brands. Power Sound Audio designs, develops, and assembles high quality loudspeakers and subwoofers for home theater environments. Visual Apex: The most competitive pricing for home theater projectors... and built on customer satisfaction! SVSound Emotiva is your Home Theater Component Source for Audiophile Quality Home Theater Equipment at Factory Direct Prices Newegg: The perfect solution for all your HTPC and media server build needs! Creative Sound Solutions: Loudspeaker kits and components for subwoofers, midwoofers, woofers and full range speakers! XTZ Sound

Like Us on facebookREW
Go Back   Home Theater Forum and Systems - HomeTheaterShack.com > Equalization | Calibration > REW Forum
Forgot Password?
Donations Reviews Vendors

 Reply     Post New Thread
Views: 2125 - Replies: 36  
Thread Tools
Old 12-02-11, 05:43 PM   #1
Shackster

Since: Apr 2011
Posts: 26
UNITED STATES
  foxfire3 is offline    
I started this thread.
Reduced level when measuring drivers in parallel?


I'm using REW v.5.0 to measure a woofer, tweeter and both in parallel.
The woofer is a Tang Band 6" unit, the tweeter is Fountek 3.0 and all are mounted in a box.
The amplifier is a $100 Sherwood 2 channel receiver from Radio Shack.

When I measure the woofer and tweeter in parallel(using a 30uF cap. in series with the ribbon), the level is lower than measuring the woofer and tweeter by themselves.
I'm not changing any volume controls or moving the mic.

Is this normal, or am I doing something wrong?

I've included a photo:
Red is the summed response...gold is the woofer by itself and purple is the tweeter by itself.

Attachments
 

Forum Rules Reply With Quote
Sponsored Links
Advertisement
 
Old 12-02-11, 06:50 PM   #2
HTS Senior Moderator

robbo266317's Avatar
bigbadbill
Since: Sep 2008
Posts: 4,360
Newcastle Australia
  robbo266317 is offline       
Re: Reduced level when measuring drivers in parallel?


You should use a proper crossover to get proper measurements, even a simple LC will suffice.

Here is a simple crossover calculator:- http://www.ajdesigner.com/crossover/crossoverfirst.php
Plug in the impedance of the drivers and the desired crossover frequency.

In your case it's probably going to be around 2kHz
Assuming they are 8Ω drivers this gives C1 = 9.9 microfarads (round up to ten) and L1 = 0.64 millihenries

Cheers,
Bill.


Silence is golden but duct tape is silver.

DIY completed:
http://www.hometheatershack.com/foru...-shelf.html

Forum Rules Reply With Quote
Old 12-02-11, 07:51 PM   #3
Shackster

Since: Apr 2011
Posts: 26
UNITED STATES
  foxfire3 is offline    
I started this thread.
Re: Reduced level when measuring drivers in parallel?


Bill, I'm not trying to come up with a crossover right now.

I'm measuring the drivers' responses and phase so I can import them into crossover design software later.

My question is really why the paralleled response is lower in level than the individual responses?


Forum Rules Reply With Quote
Old 12-02-11, 10:12 PM   #4
HTS Senior Moderator

robbo266317's Avatar
bigbadbill
Since: Sep 2008
Posts: 4,360
Newcastle Australia
  robbo266317 is offline       
Re: Reduced level when measuring drivers in parallel?


Quote:
foxfire3 wrote: View Post
Bill, I'm not trying to come up with a crossover right now.

I'm measuring the drivers' responses and phase so I can import them into crossover design software later.

My question is really why the paralleled response is lower in level than the individual responses?
Ok, to get frequency and phase response you only need the individual drivers responses which should be referenced to the system you are using to measure it. It's only after you get these figures that it enables you to design the crossover and then you can get the full system frequency vs phase response.

In effect your parallel response is lower because the load you are applying to the amp with a simple 30 microfarad cap is not good. Assuming the drivers are 8Ω then you are effectively crossing over around 700 to 800 Hz.

For accurate crossover design software you really need to obtain the Theile/Small parameters and model your drivers and then choose a suitable crossover point.

Without the Thiele/Small parameters you either need to measure the individual drivers response and design the system from there, or you need to measure the combined response with a "best guess" crossover in place.

Unfortunately there are no shortcuts that will give meaningful results as you have demonstrated.

Can you tell me the driver models and manufacturer so I have a better idea as to how they will interact?

Cheers,
Bill.

PS A 10 microfarad cap and 0.64 millihenry inductor should only set you back a couple of dollars to do some preliminary measurements.


Silence is golden but duct tape is silver.

DIY completed:
http://www.hometheatershack.com/foru...-shelf.html

Forum Rules Reply With Quote
Old 12-02-11, 10:48 PM   #5
Shackster

Since: Apr 2011
Posts: 26
UNITED STATES
  foxfire3 is offline    
I started this thread.
Re: Reduced level when measuring drivers in parallel?


Bill, thank you for your replies.

The cost of the inductor and capacitor really isn't an issue.
I probably have one of each lying around in one of my parts boxes.

The woofer is a Tang Band W6-789E and the tweeter is the Fountek Neo3.

I measured the same drivers using HOLM in the photo below.
The black line is the woofer, blue is the tweeter and red is the summed response.
As you can tell their levels are pretty close, unlike what I'm getting with REW above.

I'm not here trying to make HOLM look great and REW look bad...NOT at all.
I really like REW, but I'm trying to figure out what is causing the difference in summed response level in REW.

Attachments
 

Forum Rules Reply With Quote
Old 12-03-11, 09:16 PM   #6
Shackster

AvantGuy's Avatar
Bob
Since: Oct 2011
Posts: 6
New York upstate
  AvantGuy is offline    
Re: Reduced level when measuring drivers in parallel?


But there are still a couple dips in the middle, I think. Could this possibly be a result of phase cancellation? Unless you're certain the drivers are wired in phase, you might take a minute or two to retest with one of the drivers reversed from its current polarity.

--Bob


Forum Rules Reply With Quote
Old 12-03-11, 09:34 PM   #7
Shackster

Since: Apr 2011
Posts: 26
UNITED STATES
  foxfire3 is offline    
I started this thread.
Re: Reduced level when measuring drivers in parallel?


Quote:
AvantGuy wrote: View Post
But there are still a couple dips in the middle, I think. Could this possibly be a result of phase cancellation? Unless you're certain the drivers are wired in phase, you might take a minute or two to retest with one of the drivers reversed from its current polarity.

--Bob
Hi Bob,

You are correct...there is quite a bit of phase cancellation where the tweeter's response overlaps the woofer's.
If I reverse the polarity on the tweeter, most of the cancellation goes away.

Still I'm wondering why REW's summed response is almost 26dB lower than the individual responses?

I also tested these same drivers using the free version of ARTA.
Again, the summed response was very close in level to the individual responses.


Forum Rules Reply With Quote
Old 12-09-11, 09:41 PM   #8
Shackster

Since: Apr 2011
Posts: 26
UNITED STATES
  foxfire3 is offline    
I started this thread.
Re: Reduced level when measuring drivers in parallel?


Wow...129 views and no one knows why the lower level in the paralleled response?

Perhaps I'm the only one who measures drivers in parallel to establish their acoustic offsets.

I was hoping JohnM might know what's causing this.


Forum Rules Reply With Quote
Old 12-09-11, 10:47 PM   #9
Senior Shackster
Silver Supporter

Since: Mar 2009
Posts: 779
Michigan, USA
My System
  jtalden is offline    
Re: Reduced level when measuring drivers in parallel?


I have no good guess as to what went wrong here and there is little info provided.
The HOLMImpulse measurements appear correct. The REW measurements are not. To find out where the problem is, we probably need to start with a proper calibration (it's doubtful that your noise floor was at 100 dB for instance.

Start by showing us the soundcard cal and mic cal on you charts and get a proper adjustment of the REW SPL Meter. Calibration is probably not the major issue here, but if that isn't correct is harder to get to the real problem.

Some questions:
Did you repeat the series of measurements and replicate these same results?
Are you using multiple sweeps? if so, use just one and see if that has an impact. It should not.
Is your soundcard and REW set to the same sample rate?
The mic was in the same position and none of the mic or speaker levels were changed?


Forum Rules Reply With Quote
Old 12-10-11, 06:36 AM   #10
REW Author

JohnM's Avatar
John
Since: Apr 2006
Posts: 5,096
UK
  JohnM is offline       
Re: Reduced level when measuring drivers in parallel?


Quote:
foxfire3 wrote: View Post
Is this normal, or am I doing something wrong?
Strange result indeed. Could you attach the mdat file with the three measurements?


Forum Rules Reply With Quote
 Reply     Post New Thread     Back to Top of Page

Check out the Woot product for today!   Musicians Friend: Your one stop shop for microphones, mic amps and Behringer products.   Cmple   Monoprice Electronics   Amazon   Velodyne

Parts Express: Excellent Source for DIY Speaker and Subwoofer Projects!   Onkyo   GIK Acoustics: Home audio acoustics at its best... especially when you have help from the owners right here at the Shack!  Check out their very affordable acoustic panels!   Nyrius   Aperion Audio   Axiom Audio


« Home Theater Shack > Equalization | Calibration > REW Forum »

« Previous Thread   Next Thread »

Bookmarks

Tags
drivers, level, measuring, parallel, reduced
Thread Tools

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 is not allowed!
Trackbacks are Off
Pingbacks are Off
Refbacks are Off







Bookmark and Share


Parts Express: The #1 Internet source for all your DIY and electronics needs!

This site is best viewed with a screen resolution width of 1280 or higher!




Powered by vBulletin
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©2006 - 2014, Home Theater Shack, LLC.



OUR SPONSORS


OUR AFFILIATES
Abe's of Maine

Adorama



Best Buy

Boston Acoustics A Series Speakers

Cables.com

Circuit City

Denon

DIRECTV

e-Cost

emusic.com

Fluance

GameFly

GamersGate

Harman

Headphone Authority

HDtracks

iTunes

J&R

Logitech

MarvelStore

MCM Electronics

Musician's Friend

Monoprice

Newegg

OneCall

Overstock

Panasonic

Parts Express

Paul's TV

Roku

SMS Audio

Solid Signal

Sonos

Sony

SquareTrade Warranties

Super Media Store!

Vann's

Velodyne

vudu

Warner Bros

Winegard

Woot

World Wide Stereo

ZVOX

Zzounds




Search Engine Friendly URLs by vBSEO