Some SpeakerWorkshop crossover design q's - 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 10th July 2008, 08:33 PM   #1
diyAudio Member
 
Join Date: Jul 2004
Location: Beveren
Default Some SpeakerWorkshop crossover design q's

I have choosen these drivers for my next project:
- Beyma SMC 2012/N (tweeter)
- Beyma PRO 6MI (mid)
- Subwoofer will be added later, for more info on my goals I refer to this thread:

http://www.diyaudio.com/forums/showt...34#post1547534


To design the crossover, I used SpeakerWorkshop. Unfortunately I couldn't find a "SpeakerWorkshop for Dummies", but this great walktrough provided what I needed:

http://bellsouthpwp.net/l/j/ljfrank/...html#crossover


If you look at the attachment on the bottom, you'll see the result I have now.
The dip at 1000 Hz can be equalized out later I think - or choose bessel slope ??? there is a dip in the woofer responce so it is difficult to raise the responce thetr anyway IMO. Padding down the tweeter some more also shouldn't be a problem.

My questions are about the components SpeakerWorkshop proposed.
- Tweet: The acoustical slope is almost perfect to the 2nd order LR goal. SW seems to hint that I could do with only the cap, but as said/explained in the walktrough above, I'll keep the 2nd order electrical to protect the tweet.
- Woofer: The slope descends less than the 2nd Order LR goal, but since it will be listened to off-axis the high freq content will drop more and I think I shouldn't be too far from what I want. Am i correct about this?
SW says "422.2uH" so I think this is a 0,42 mH ?

Since I could do with only a cap on the tweeter, will the extra inductor provide enough protection?

Other ideas to improve?
Attached Images
File Type: jpg combined.jpg (76.1 KB, 257 views)
__________________
-- If in doubt: overdesign!
  Reply With Quote
Old 10th July 2008, 08:40 PM   #2
diyAudio Member
 
Join Date: Jul 2004
Location: Beveren
And perhaps the .swd file could be of some help:

http://users.telenet.be/Hellhouse/pu...hine2order.zip

(I put it here for the moment, as it is too big to attach in this reply.)
__________________
-- If in doubt: overdesign!
  Reply With Quote
Old 11th July 2008, 04:28 PM   #3
JLC7 is offline JLC7  United States
diyAudio Member
 
Join Date: Apr 2008
First thing I noticed is that you have a 1000 H inductor in your tweeter crossover. That's ridiculously big. Even if you wanted 2nd order electrical you can't buy an inductor anywhere near that big.

Usually speaker workshop gives you 1000 H components after optimization if the component isn't needed or if the values your components started with led to a solution that didn't converge. In the latter case, you can change the values back down to something reasonable and optimize again to see if it will give you a more reasonable result.

Quote:
The slope descends less than the 2nd Order LR goal, but since it will be listened to off-axis the high freq content will drop more and I think I shouldn't be too far from what I want. Am i correct about this? SW says "422.2uH" so I think this is a 0,42 mH ?
The thing is you'll also be listening to your tweeter off axis which changes things a lot too.

What were the goals you set for your HP and LP?
  Reply With Quote
Old 11th July 2008, 04:47 PM   #4
JLC7 is offline JLC7  United States
diyAudio Member
 
Join Date: Apr 2008
You also don't have any valid phase information. What you have is just the generic electrical component phase which doesn't take into account the phase of the drivers or their offsets. THe dip around 1k is from the electrical phases being 180 degrees apart.
  Reply With Quote
Old 12th July 2008, 03:47 AM   #5
diyAudio Member
 
Join Date: Aug 2006
Quote:
Originally posted by JLC7
First thing I noticed is that you have a 1000 H inductor in your tweeter crossover. That's ridiculously big. Even if you wanted 2nd order electrical you can't buy an inductor anywhere near that big.
One thousand henrys like in 1,000,000mH? Such a passive inductor made of copper wire would weigh several tons I imagine.
  Reply With Quote
Old 12th July 2008, 03:49 AM   #6
diyAudio Member
 
Join Date: Aug 2006
Default Re: Some SpeakerWorkshop crossover design q's

Quote:
Originally posted by Cordraconis
SW says "422.2uH" so I think this is a 0,42 mH ?
Yes
  Reply With Quote
Old 12th July 2008, 04:33 AM   #7
diyAudio Member
 
Join Date: Dec 2006
Location: Where the sky loves the sea
I'm no expert on SW, but I have figured out most of it to a first order anyway.

looking at your .swd, a couple of things:

1) no impedance information for the drivers? This is easy to measure with SW, a sound card, and minimal external components.

2) driver parameters for the tweeter are nonsense. The woofer looked more reasonable but still many of the values looked like a round value was just tossed in. This info can either come from the data sheet for the driver or via measurement with SW.

3) as others have pointed out, a 1000H inductor is clearly a nonsense result. If using the optimize feature gives either very large inductance values, or very small capacitance (pF) values, then you can delete the component. Make a change, re-calculate the response, then use ctl-Z / ctl-Y (undo / redo) to see the change.

I found that 4th order acoustic response could be closely approximated with a 2nd order network.

also see this website for a helpful tutorial:

http://www.claudionegro.com/

the learning curve with SW is tough but once you figure it out it is very powerful, frankly amazing for free SW.
  Reply With Quote
Old 12th July 2008, 09:07 AM   #8
diyAudio Member
 
Join Date: Jul 2004
Location: Beveren
Thanks for the replies guys! A few things:

@JLC7:
Yes, I couldn't find such a large coil in the online shop, and the largest they had was a ridiculously 125 (+-200$) so I already concluded something wasn't right. When I said I would follow the explanation in the walktrough, I meant indeed changing to a lower value and reoptimized. Sorry if it wasn't clear and if I didn't upload the newer version.
Also I will be listening to the tweets on-axis as I'll mount them lying *on* the dashboard and not flush mounted (where they would have to reflect via the windshield.)
The goals were 2nd order L.R. for both, but after I created this thread I played around further and found out - like mightydub - that I could do a 4th order L.R. and still have the 2nd order electrical filter to protect the tweeter, so I'll change it to 4th order L.R.

What do you mean with the phase information? The SPL plot? I got the SPL plot and the impendance plot for the datasheets which I did an SPLTrace on.
The offset will be 0 after the addition of digital delays (much) later, so for now this shouldn't be an issue in the design of the passive crossover, unless I can be convinced otherwise?


@mightydub:
Strange, I did the SPLTrace of the impendance plot but in SW I don't see anything. Didn't notice that, thanks for bringing it up!
When I open the .zma file in SPL View I can see it so the data is there, however, the impendance rises slowly towards 4 Ohm at 20000Hz, while it should be around 16 Ohms in the plot I have here printed out next to me. Maybe something went wrong in the tracing or this is from the tweets T/S parameters I forgot to input in SW ...
Thanks for the link, but the site and the SW tutorial there seem to be orientated to doing measurements which is a bit hard one week prior to ordering the drivers and not having the equipment to measure.
My goal is to use the traces from the datasheet to make an as good as possible simulation of the passive crossover and to order the right caps and coils together with the drivers.


@willitwork: thanks for the reassurance. btw, for the current copper prices an inductor like that probably comes with its own sherpa to install it.



I have to go now, but tomorrow I'll try to:
- Sanitize my inputs (i.e. re-do the impendance traces and input the T/S parameters and double check everything in SW!)
- change the desired slopes to 4th order L.R. and do the simulations/optimisations to get reasonable cap and ind values.
- upload everything, including the traces to the site for you all
- ask about what possible bumps I hit when trying to do the above.

If something is still not clear, feel free to ask me.

thanks again and hang on for more!

Joris
__________________
-- If in doubt: overdesign!
  Reply With Quote
Old 13th July 2008, 01:15 AM   #9
JLC7 is offline JLC7  United States
diyAudio Member
 
Join Date: Apr 2008
The SPLtrace data you got from the spec sheet is the response of the speaker in the manufacturer's specific test environment. This could be an infinite baffle, a box of set volume, etc. They usually tell you what it is. It's typically not exactly the actual response you get in real life. That will vary based on the box you build for the speaker and where it's placed.

So you run into a few problems. The frequency response data you have probably doesn't represent what the driver's response will be in the car. Any optimization you do with speaker workshop will be as much a shot in the dark as if you didn't have the manufacturer's spec data. It might be a good idea just using a stock crossover.

T/S parameters aren't necessary for SW to do crossover optimization. You typically use a few of the parameters to design a box, but that's it. SW just uses the frequency response, phase, and impedance to optimize the crossover.

Minimum phase data can be extracted from the frequency response data, but that data doesn't reflect what will happen in the car, so the phase data won't be valid either.

If you don't put the woofer in a suitable enclosure, it might not sound good. If you just mount the driver in the door it's anyone's guess what it might sound like.

Matching tweeter and woofer levels goes out the door because the tweeter will be on axis with your head and the woofer will be somewhere further away, mounted at an angle, and probably reflecting off of stuff.


Wow. I didn't realize car audio was so complicated. Maybe that's why some people don't bother. Even zaph mentions how he doesn't bother because the T/S parameters of the drivers changes with the temperature of the voice coil which in a car could range from extremely hot to extremely cold.

Sure you don't want to make a nice home hifi speaker?
  Reply With Quote
Old 13th July 2008, 09:59 AM   #10
diyAudio Member
 
Join Date: Jul 2004
Location: Beveren
OK, here is the problem: I've made the impendance traces again (*) but I just can't get them to display in Speaker Workshop. Changeing the amount of pionts, upper and lower freq limits etc ... it all doesn't help.
Maybe there is some program to convert it to a file format Speaker Workshop understands?

http://users.telenet.be/Hellhouse/public/TranceMachine/
Here are the data files and the .pdf where I got the traces from.

@JLC7: I'll answer tomorrow, but you can see some answers yourself in the PDF's.




*: I have to check the one from the PRO6MI, the scale seems off but the shape is ok.
__________________
-- If in doubt: overdesign!
  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
FFT size in SpeakerWorkshop jm_kzo Multi-Way 2 9th July 2008 07:48 AM
Speakerworkshop question kro5998 Multi-Way 2 25th June 2006 06:18 PM
SpeakerWorkshop MLS measurement?? kimschips Multi-Way 3 21st June 2006 05:04 AM
SpeakerWorkshop again Dave Jones Multi-Way 19 17th June 2004 05:34 PM
speakerworkshop, meaning of L1 and R1? capslock Multi-Way 9 5th June 2003 07:55 AM


New To Site? Need Help?

All times are GMT. The time now is 05:31 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