WIN ISD Pro and multiway design - Page 8 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

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 2nd March 2013, 03:15 PM   #71
diyAudio Member
 
Kindhornman's Avatar
 
Join Date: Aug 2012
Location: Los Angeles, California
One often overlooked point of a crossover design is the actual impedance measurement of the speakers at the crossover point chosen. Do not just use a number like 8ohms at crossover. You need to understand that isn't the real value that you need to use. The actual impedance of the speaker in the box you are using at the crossover point will give you the impedance value that you need to use for the crossover point. This goes for both the upper and lower crossover points of a band pass filter. So if the mid range driver has a 6.4 ohm value at the lower frequency of the bandpass and say 8.7 ohms at the higher frequency point those are the values that you will need to use in your filter design to actually produce a crossover with the correct electrical filter. If you are using a Zobel or conjugate network on the speaker to flatten the impedance rise on a device then you need to develop this first and again remeasure the speakers impedance curve to plug in the electrical impedance at the crossover points to truly have the frequency points you have chosen and also the electrical -3db or -6db points that will sum correctly! If you do not do this your values will not be correct, the crossover point will not sum correctly. Remember also that you should check your network connected to the speaker after you build the network as the normal quality of a capacitor and many times an inductor can be as wide as 10 to 20% of the real values you have chosen. You will then need to adjust your values to make the crossover function correctly. This is the real reason that people often talk about why when substituting capacitors in a crossover from one brand to another of the same type that they say they hear a difference in their speakers, the actual values of the substituted components are not exactly the same over their value range and this is actually shifting the crossover point and changing the slope. This is the difference between an accurate crossover and one that is chosen based strictly on theoretical values only!
  Reply With Quote
Old 3rd March 2013, 12:07 AM   #72
Stevenn is offline Stevenn  Australia
diyAudio Member
 
Join Date: Jan 2013
Location: Sydney
Thanks guys your replies.
The woofer and mid driver impedance charts have large peaks which i plan to attenuate with Zobel circuit.
Still not sure on the PCD qsns, my logic (?!) says that as target response shape is a function of the filter used, it doesn't make sense to try and match a First order Butterworth target response with a Third Order parallel filter, for example.

And what is electrical frequency?

Cheers
  Reply With Quote
Old 3rd March 2013, 12:14 AM   #73
diyAudio Member
 
Kindhornman's Avatar
 
Join Date: Aug 2012
Location: Los Angeles, California
Stevenn,
I think what he was trying to convey was that the electrical response can be different than the actual acoustical response curve. They are not one and the same. Though you can put a 2nd or 4th order filter on the speaker you have to add this to the acoustical rolloff of the speaker, they are additive and not necessarily coincident to each other. If the speaker is rolling off naturally this will add to the electrical filters curve. I don't think that the tendency for many to only look at the acoustical slope is important as much as that the two crossover points are coincident at the -3db or -6db point dependent on the slope that you chose. That is why I tried to point out that a simulation only using electrical response of an ideal network does not often do what it is supposed to. The values of the electrical components have to be exacting and the impedance at crossover of the two devices have to be known true values.
  Reply With Quote
Old 3rd March 2013, 12:33 AM   #74
diyAudio Member
 
Moondog55's Avatar
 
Join Date: Jun 2007
Location: Norlane; Geelong: Victoria: Australia
Yes but the target acoustic shape is often much greater than first order, often the target is a 4th order LR shape or similar.
But if you have a driver with a good second order shape roll-off, lets say a tweeter then a first order electrical ( a single cap in series ) will give you a third order response acoustically and all you need to do is use a frequency where little bass gets through.
Use a second order filter and power handling increases so you can play louder
So much depends on how much power ( how loud do you personally play music?) the driver needs to handle.
For example my bedroom speakers [ 2.5 Way] will only be asked to play to about 90dB at most so I am using a single cap on the tweeter and running the Vifa P-13 full range in a 7 liter sealed box; and the speaker will be handling only about 30 watts peak.
With a proper second order XO at 300 Hertz the little Vifa alone could handle triple that with ease.
Electrical frequency is my shorthand for the calculated filter response using components into a fixed resistance, and you can calculate or get from tables.

Just for your information at around 300 hertz most woofers have an impedance equal or close to their DCR and often [sometimes] a drivers nominal impedance is given as the impedance at the XO frequency.
Zobels are used to flatten the impedance at a higher frequency so a great help when using book crossovers based on nominal impedance.
__________________
QUOTE" The more I know, the more I know, I know (insert maniacal laugh >here<) NOTHING"
  Reply With Quote
Old 3rd March 2013, 12:39 AM   #75
diyAudio Member
 
Kindhornman's Avatar
 
Join Date: Aug 2012
Location: Los Angeles, California
Moondog,
I understand what you are saying. I am moving past the use of a simple Zobel network of a simple RC network to a true conjugate network using an LRC network that is counter to the speakers impedance curve. Not something that is done much here in diy audio.
  Reply With Quote
Old 3rd March 2013, 01:03 AM   #76
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Hi Steve, I've attached an example here from my own crossover.

The target was 4th order bessel at 2.8Khz. The red traces represent that target.

The crossover to actually achieve that target consisted of a couple of notch filters and a second order electrical filter on the woofers, and a third order electrical on the tweeter. This crossover is quite a complex one but I was struggling with a few things that made it hard to get a good match to my target, I got there in the end though

Other things to look at (other than seemingly flat frequency response) are how well the phase of the two speakers tracks through the crossover region (dotted lines on the graph)

What Kindhornman said about impedance is something to watch for as well. I spent ages coming up with a super flat frequency response in the simulator but had not been simulating the impedance, I decided I should check it only to find that the impedance was dropping to about 1 ohm in places. Luckily I realized before I built it!! So do check the impedance!!

Tony.
Attached Images
File Type: png crossover_slopes.png (34.6 KB, 45 views)
File Type: png crossover1.png (11.0 KB, 44 views)
File Type: png impedance1.png (15.0 KB, 43 views)
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 3rd March 2013, 01:33 AM   #77
Stevenn is offline Stevenn  Australia
diyAudio Member
 
Join Date: Jan 2013
Location: Sydney
Thanks yr replies and examples. I think we are all on the same wavelength (pun intended!).
My qsns refer directly to the use of Jeff Bagby's spread sheet and what selections to make.

So I select the target response curve I want but this is a name only and might as well be curve 1,2,3 or 4 and then try and match it with filter order and component value?

Clear as mud?

Comments about impedance noted, Bagby can model this as well and this is next step once I resolve above queries.

Thanks
  Reply With Quote
Old 3rd March 2013, 03:47 AM   #78
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Yes select your target response curve and then using the various electrical filters try and get the speakers curve to match it. Don't be afraid to set the filter to be different than what you would expect it to be, the end goal is to have the acoustic output of the drivers matching your goal.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 3rd March 2013, 03:56 AM   #79
diyAudio Member
 
Moondog55's Avatar
 
Join Date: Jun 2007
Location: Norlane; Geelong: Victoria: Australia
Even using factory data you can do a box design in JBs box program and save the impedance files and then use those in the XO design program.
May I suggest starting with LR 2nd as the design goal
__________________
QUOTE" The more I know, the more I know, I know (insert maniacal laugh >here<) NOTHING"
  Reply With Quote
Old 3rd March 2013, 08:25 PM   #80
Stevenn is offline Stevenn  Australia
diyAudio Member
 
Join Date: Jan 2013
Location: Sydney
Thanks
i created impedance files in the first place from datasheets using SPL trace as Bagby PCD will not work without them.
Moondog - 'LR 2nd as the design goal' - what does this mean?

Also can anyone answer qsn as to what 'selected electrical frequency' means and does in Bagby PCD?

Thanks
  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
Unconventional Speaker Design - thread carried over from multiway forum morgoe Full Range 59 30th October 2012 12:17 PM
Ideas and approaches to dialing in an active multiway design ?? flatfinger Multi-Way 4 17th October 2011 01:28 PM
Win XP Pro keeps alerting usb device Bengali Everything Else 5 16th April 2010 10:20 PM
Active multiway speaker design? leadbelly Multi-Way 12 24th May 2007 12:52 AM


New To Site? Need Help?

All times are GMT. The time now is 01:22 PM.


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