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

B&W Matrix 802 S3 crossover refresh
B&W Matrix 802 S3 crossover refresh
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 27th January 2015, 02:59 PM   #1
glina is offline glina  Poland
diyAudio Member
 
Join Date: Oct 2003
Location: Warsaw
Default B&W Matrix 802 S3 crossover refresh

Hello all,

I'm starting a project to refresh the XO of my 802s3. In fact, to build new ones from scratch.

The original crossovers are built with basic components (electrolytic capacitors, iron core coils) which I'm aiming to upgrade.

I'm aware that the original system was calculated with component imperfections accounted for and I want to compensate for this. I also intend to match the DCR of all coils and ESR of all caps.

Here is the schematic of original B&W crossovers.
B&W Matrix 802 S3 crossover refresh

source: B&W crossover rebuild

C1 and C2 are MKP
C3, C4 and C5 are Electrolytic
L1 is Air core
L2, L3 is P-Core
L4, L5 is Ferrite core

I'm especially careful about C3 and C4 which I want to replace with MKP.

Assuming a D.F. of 0.03% for these capacitors, I planned to add approx 0.15Ohm in series with C3 and 0.33Ohm in series with C4 to compensate for the difference, however the DF is frequency dependent and my approximations are therefore not perfect. The frequency cutoff points may be affected. Is there any other way to better simulate the behavior of Electrolytics in this circuit?

I will use electrolytic for C5. Or should I not?

Can I use an air core coil with higher ESR for L2 and compensate for the difference with R1 values? The original is a P-Core with approx 0.2 DCR.

Same question for L3. Can I use an air core coil with higher ESR to compensate for lower ESR of the C4?

The component types I plan to use are following:
C1, C2 - Mundorf Supreme
C3 - Mundorf EVO 22uF + Mundorf Supreme 3.9uF (or Jantzen Superior 3.9uF)
C4 - Mundorf Evo 10uF
R1, R2 - Mills or Jantzen Superes (any recommendations?)
L1 - Air Core 0.4mm wire Jantzen
L2, L3 - Mundorf copper foil 14mm I-core (Feron core) (or air core coils if I can compensate for their increased DCR)
L4, L5 - Mundorf 1.4mm (Feron core)

I'd appreciate any tips.
  Reply With Quote
Old 28th January 2015, 01:00 PM   #2
nannoo is offline nannoo
diyAudio Member
 
Join Date: Jan 2013
Certainly a good idea, but before we go further can we clarify a few things;

Do you have a decent quality LCR meter? (Or meters)

Can you show pictures of the existing crossovers?

I'll spend a few more minutes with this later today, are you uk based?
  Reply With Quote
Old 28th January 2015, 01:20 PM   #3
glina is offline glina  Poland
diyAudio Member
 
Join Date: Oct 2003
Location: Warsaw
There are some photos of the original crossover, as well as his own ones `on the website of Graham O'Neill
B&W crossover photos
Lots of good information there, but I don't think he followed through with value matching.

I'm in Switzerland, but will be ordering my parts from Hificollective (UK) as they have all I need and want.

I can organize an LCR meter, no problem.
  Reply With Quote
Old 28th January 2015, 09:28 PM   #4
nannoo is offline nannoo
diyAudio Member
 
Join Date: Jan 2013
Also, what are your end goals? Are you happy with imaging etc?

I find B&W crossovers have a very particular voicing, which I can only describe as fluffy... they're not as analytical as they can be, but are a very easy listen.

There's a very important question of what do you want to achieve, and would you consider a re-design?
  Reply With Quote
Old 29th January 2015, 06:22 AM   #5
glina is offline glina  Poland
diyAudio Member
 
Join Date: Oct 2003
Location: Warsaw
My goal is to clean up the sound a little bit. I'm happy with the imaging and overall balance, but I feel they're a bit dull, grey overall and hazy in the very top end.

I have another set of boxes based on D260 and a Davis woofers, and while they can't quite keep up with midrange detail, they are smoother and more involving.

I wouldn't mind a bit more speed at the bass, at expense of weight. Can I tweak R2/C5 values somehow to achieve this?

As for a re-design - I'd rather not, I don't have the resources or parts to fine tune the end effect and I want to keep it close to the original.
  Reply With Quote
Old 29th January 2015, 08:31 AM   #6
nannoo is offline nannoo
diyAudio Member
 
Join Date: Jan 2013
Givem all that you've said, I'd be looking at replacing the capacitors only and keeping the inductors - changing the inductors is likely to be costly and compromise the sound. Small variations in the DCR of the inductors will effect the transfer functions of the network and could mess up BSC etc.

Going that route, personally I'd be looking at mundorf supremes in the tweeter circuit, standard m-cap for the woofer bypass, and evo for the mid cap bypass.

The 26uF series cap in the midrange I'd personally replace with a 22uF & 4.7uF Evo oil, though if you realy wanted you could change the 4.7 to a supereme - I wouldn't see any advantage to doing that though.

One advantage of ordering through mundorf/hi-fi collective is they match all the components for you to very tight tolerances.

If you wanted to tweak, personally I'd look at reducing the series resistance on the mid either to 0 or with bypass resistors (you could check that immediately with a little wire around the resistot, that way you can remove it without any physical alterations), I'd almost definately increas the 4.7 cap to 5.6, and probably swap the values of L2 and L3... those observations are made from previous experience with B&W xo's.
  Reply With Quote
Old 29th January 2015, 09:35 AM   #7
glina is offline glina  Poland
diyAudio Member
 
Join Date: Oct 2003
Location: Warsaw
Thanks a lot for the tips. I planned to build new XO's anyway, as I intend to keep the originals for comparisons sake. In any case I want to match DCR down to <0.03 Ohm or better.

Could you please explain your reasoning for changing C1 from 4.7uF to 5.6uF?
Also, why would C3 = 22uF+4.7uF be a better match than 22uF+3.9uF?

Wouldn't swapping L2 and L3 values affect the frequency cutoff points?

Also, would you suggest keeping C4 electrolytic or is my suggestion of MKP + series resistor reasonable?
  Reply With Quote
Old 3rd February 2015, 03:10 PM   #8
glina is offline glina  Poland
diyAudio Member
 
Join Date: Oct 2003
Location: Warsaw
A couple of questions to more experienced DIYers:

1. What effect could I expect by using MKP C4 without a series resistance to compensate for the ESR of the electrolytic which is there by design? I used a crossover simulator software with idealized 8ohm load and I can only see a very minor shift (~0.2dB) in SPL and an equally small group delay shift.

Original (10uF + estimated 0.7R ESR):
An externally hosted image should be here but it no longer works. Please upload images instead of linking to them to prevent this.

An externally hosted image should be here but it no longer works. Please upload images instead of linking to them to prevent this.

Modified (10uF + estimated 0.1R ESR):
An externally hosted image should be here but it no longer works. Please upload images instead of linking to them to prevent this.

An externally hosted image should be here but it no longer works. Please upload images instead of linking to them to prevent this.


Could the effect be much more pronounced if simulated with real instead of idealized load?


2. I tried simulating the effect of R2/C2 notch filter but failed. How can I estimate the frequency affected by this filter? Can I tune the bass response by modifying R2 (lower R2 = stronger attenuation at filter frequency?)
  Reply With Quote
Old 6th February 2015, 10:20 PM   #9
nannoo is offline nannoo
diyAudio Member
 
Join Date: Jan 2013
glina,

sorry, feel like I'm really letting you down as haven't had time to come back to this (middle of dissertation) will give it the time it deserves shortly, let's not mess about - if you're set on re-building from scratch you may as well improve at the same time.

Do you have any ways to measure transfer function of current xo and speaker response??
  Reply With Quote
Old 7th February 2015, 01:19 AM   #10
AllenB is online now AllenB  Australia
diyAudio Member
 
Join Date: Oct 2008
Quote:
Originally Posted by glina View Post
Could the effect be much more pronounced if simulated with real instead of idealized load?
I have a plot of the difference here. A tweeter (green) has the crossover applied (blue) to be close to the target slope (black). This has been done using a full impedance plot.

Where the impedance magnitude (level) of the plot is correct but phase information is missing (zeroed in this case) the response will come out on the simulator as shown in pink. When pure resistance has been used, you'll get the yellow plot for 8Ω, cyan for 12.

It will not be possible to have confidence in a simulation to the level you seek unless you have impedance plus phase information, and of course acoustic (frequency response) plots for each driver. For the combined responses you'll also need acoustic phase (e.g. four plots per driver).

To re-engineer the crossover for improved performance I would take directivity plots seeking the power response of the drivers to augment the single plots.. and for finding the right angle to listen at.
Attached Images
File Type: gif xo.gif (17.6 KB, 210 views)
  Reply With Quote

Reply


B&amp;W Matrix 802 S3 crossover refreshHide 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
Pensil 12 and B&W Matrix 802 goneill Planet 10 hifi 65 26th April 2011 04:31 AM
B&W 801 Matrix S3 zavan Multi-Way 2 7th October 2010 12:46 PM
Sound Anchors for B&W 801 Matrix S3 Julius Swap Meet 0 14th January 2008 05:04 AM
B&amp;W 801 Matrix S3 crossover schematics? arildji Multi-Way 12 13th September 2006 02:27 PM
Internal mods on B&amp;W 802 Matrix? Jai Multi-Way 5 18th May 2004 06:16 PM


New To Site? Need Help?

All times are GMT. The time now is 02:16 AM.


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