Crossover help needed for Dynaudio 17W75XL(ohm) and D260

[gordonhunsaker]

View attachment Frequency Response-plot.pdf

I have Dynaudio 17W75XL {pn#84862}8 ohm mid/woofers from Dynaudio BM6A studio monitors [put into in 12 litre vented enclosure] and D260's tweeters out of Vision Acoustique model 702 speakers. I have worked hard on a very nice-damped/braced cabinet for these drive units. I now need optimum crossover design for this combination with high/low pass sections, respective zobels for woofer and tweeter (if recommended), maybe resonant or notch filters to improve sound and tweeter attenuation if necessary plus any phase corrections, etc. that might also help sound quality. Currently I have below configuration crossover that sounds good already and will rebuild crossovers a final time to live with the speakers forever upon getting proven ideal x-over spec and finally put this fun project to bed. Had based initial crossover design with attached theoretical plot with online "Speaker Wizard.uk" resource. Using Solen standard air core coils and polypropylene caps now but will consider Solen litz coils and Solen film/foil caps next time for final crossover build. Also using Solen achrohmic mil-spec resistors. Love the Dynaudio sound so far...!!!

Thanks in advance for any help!,
Gordon Hunsaker
Calgary, Canada

 

[Lojzek]

Hello Gordon, very nice cabinets, also my kind of style finish. Unfortunately the XO filter isn't accounting for baffle step loss of frequencies and the calculus is based on idealized impedance (and frequency response) being a straight line (8 ohm). The best I can offer is tracing Dynaudio original measurements, applying simulator brewed baffle step response and make a crossover that will be closer to real deal but not entirely as real measurements which are proven to be harder to execute than one would like too. I have a set of 4 ohm car audio MW160GT and D260 but my tweeters have dented domes and come from different years of production, looking different at the back chamber. I'll use 17W75 LQ datasheet for I don't know if XL version existed in DIY form since XL actually stands for higher Bl parameter meaning it was intended for a vented application and LQ is that also. I need the baffle measures, width, height and center to center distance of the drive units. Thank you.

 

[gordonhunsaker]

Thanks for the quick reply! I cannot take credit for the initial cabinet construction but damped all internal walls with bitumen padding (from Solen) and mechanically braced the inside walls (side to side and front to back). The original speakers I picked up were Chinese made (by Cayin) but had fake Dynaudio drivers so I sourced the real ones to put into the speaker enclosure and re-built the crossovers accordingly with the knowledge base I have. As an additional note you will see the first L-pad on tweet [-3dB] but I subsequently put in the in series resistor to take the tweeter amplitude down another (-4dB) for a (-7dB) total to satisfy what I thought was the tweeter being too loud. Do you think the new 17W75XL's I have will gain amplitude slightly after breaking in 100+ hours of use to catch up more with the tweeter? I am already thinking of laying off (-1dB) on the tweeter padding to net (-6dB) but wanted to reach out to the world to obtain suggestions on the entire crossover topology. After altering the crossover design and even substituting components such as the litz inductors and film/foil caps will ultimately lead me to a final assessment of tweeter level to suit my ears. Wanted to re-think the entire x-over and hope to approach the sound quality of perhaps Dynaudio "Focus" or "Contour" series with all the attention put into cabinets and crossover albeit I don't have the "LQ" version 17W75's nor the D280's. Am working out of country at the moment and would return home in a few weeks to provide exact cabinet dimensions plus exact driver spacing for you...


Thanks again!

Regards,
Gordon

 

[gordonhunsaker]

FYI: Dynaudio data sheets I used for reference in considering crossover for 17W75XL-8 & D260:

 

[gordonhunsaker]

Lojzek,

Pdf's of drivers missed on last post:

Thx,
G.

 

[system7]

I am sure Lojzek will come up with something here.

Your current crossover looks quite close to the Foccus4 design, if you untangle the doubled-up excesses of that one. But your polarity looks wrong to me. You usually get a rather echoish or hollow sound on steel guitar like Dire Straits if it is wrong. Box seems a bit at small at 12L too.

Anyway, what I glean so far. I'm just watching the action here.

 

[gordonhunsaker]

Thanks Steve,

Could try the original Dynaudio Sirius kit crossover but alter component values for 8 ohm version instead of their 4 ohm 17W75 (i.e. 1.8mH?) plus respective Zobel values. Anyone try Solen Hepta-Litz coils for x-over application on mid/woofer? On the Sirius x-over, what would the value of Rs be as its not indicated? Also excuse my lack of knowledge but what would the two final sections in their design be paralleled to D260 tweeter for: 0.68uF/22 ohm[?] + 0.27mH/15uF/10 ohm[resonant cct?] - for response shaping??

 

[Lojzek]

Gordon, how do you find the tweeter level with this filter of yours? Would you like it changed in any way? I'm close to making the sim. Btw. hepta litz sounds like heavy snake oil!

 

[gordonhunsaker]

Hi Lozjek,

I find the tweeter level with that much attenuation within a dB or two of me liking. I cannot stand for over achieving tweeters! Still would like to find the optimum crossover for my Dynaudio combination however. According to Solen themselves midrange should come out more coherent with the Hepta Litz coil in series to mid/woofer but maybe it will be snake oil in the end?! What are your thoughts on a better cap such as film/foil compared to regular polypropylene - had you tried? Anyone else...?

Thanks,
Gordon

 

[ShaneF]

I am using the crossover for that tweeter from the Sirius Dynaudio design. It is very good, in fact, the sibilance is completely gone. If you you know the D260, you know that it has a sibilance that will destroy your life.

 

[ShaneF]

Gordon: I am using the Solen silvers. They are excellent, a "little" more expensive... but not stupid expensive.

Oh yeah... the hepta litz is bogus. So are those tape coils.

 

[Lojzek]

Hi Gordon, I don't recommend any brand or type of caps. Poly caps are excellent by any means and personally I'm fine with NPE's as well but I use MKT Philips and Roederstein for I have bought these at discount prices for a lifetime stash.

I am supplying the plots I promised to. It was a little bit odd to work on it. I used the official Dynaudio woofer measurements in their respective cabinet and I didn't apply any changes to it. It is what it is. My filter proposal is a simple one which I thought would be a proof that tweeter is better protected from lower frequencies level and the XO point is in the ballpark. However you are free to modify it to your taste. Padding resistor in tweeter branch can be any value in either direction so if you are confident, try it.
Tweeter very high frequencies are 0 deg axis so as soon as you start moving horizontally, the level goes down. I would not worry about it. The important range is at about 5 kHz. That should be set right. Black curve is summed response of drivers.

 

[Lojzek]

I am using the crossover for that tweeter from the Sirius Dynaudio design.

In so many words, thanks Lojzek for wasting your time doing filters that don't work!

 

[ShaneF]

In so many words, thanks Lojzek for wasting your time doing filters that don't work!

Actually, I "am" using the crossover you helped me with for the mid range and woofer. It is working beautifully! We didn't discuss the tweeter crossover because I knew I was going to go with the sirius design already.

Thanks, it's awesome!!

Here it is before I glued the parts in and finalized the placement!

 

[Lojzek]

Hmm, interesting look of the blue speakers. Reminds me of one of the tetris game building blocks.

 

[gordonhunsaker]

Lojzek,

Amazing what you have presented! Your net O/P response looks great!! After looking into what you have used (Xsim) it has opened a new world for me as I downloaded the software today!! The only thing I need to know is how to trace the response curves and impedance curves from factory spec sheets to make respective .frd / zma files. Also how do you import these files into Xsim to get going to start modeling a X-over? Would you so kind as to send me the .frd /zma files you had traced so I could play and tweak the x-over design myself until I learn how to trace my own?

What are your thoughts about Shane's sibilance killer X-over by Dynaudio D260 (Sirius design)?

Thanks again in advance,

Gordon

 

[gordonhunsaker]

Thanks for you input Steve!

Cheers,
Gordon

 

[Lojzek]

Hey Gordon, there is this great tracing software at fprawn labs that makes so much easier the job since you don't have to click 1000 times to get frd and zma for one driver like I used to do with SPL Tools.

In XSim you place the driver icon on a sheet and right click to tune properties. There you'll see a dialogue to browse for the files. Additionally I have used +0,5" of delay to the woofer part which states that woofer is half of an inch further away from the baffle.

The files are attached and what I think of Sirius tweeter filter is in the simmed photo. It's okay for a 1st order filter but I am very reluctant to drive a sensitive tweeter voice coil like that so I always use steeper ones. Blue curve is with the filter of mine (yours) additionally padded down to meet the grey which represents Sirius.

 

[gordonhunsaker]

Lojzek,

You rock! Thank so much for these and all your help/education along the way!! I will play around tomorrow with all of this...and let you know what I settle on...

Best Regards,
Gordon

 

[gordonhunsaker]

Trying the fprawn labs software but having difficulty locking onto driver curve..
How did you mange to capture them properly off of the data sheets?

Thanks,
Gordon