[delebru]

Hi all! I've been reading the forums for a while and following amazing builds and mods :) A year ago I've built a couple of gr subs running my lounge system with a pair of Focal Aria 936 I grabbed on a great discount. I am very happy with them, it's the best sounding system I've owned, but I'm curious to hear the difference a better crossover can produce.

The speaker's response on my lounge as measured by dirac live is fairly smooth. Gets spiky above 10k but dirac's filters should flatten that out.

I've removed the crossover and put the design down on paper but can't figure out the design for the mids other than the notch filter. I'm also not sure why the lows have the polarity reversed but it seems to work... Driver's and speaker have the DCR marked, I don't have anything to measure speaker's impedance.

Unfortunately I can't ship the speaker to Danny internationally for a crossover redesign, and although I want to try measuring the drivers individually with and without crossover, I still need to learn how to do that with REW. I'm also not sure if I can get enough data for a full redesign, so was wondering if it would be a good idea to try rebuilding the same crossover with good components?

[Shives]

Hey! Awesome ideas!

So, normally Danny’s takes the full networks out, reads the responses and starts to build a custom network right before his eyes for the speakers. Then takes a measurement for every component replaced each time, see the response of it helped, or made it worse. He’s really good at this! Obviously! Lol. 

I don’t know if you have the ability to do this. I’m sure they make a virtual crossover design, but doing it with the raw parts at hand is really what’s needed. This is the only way you can really factor the drivers response in the box! With measuring.

So, let’s say you can’t do that… you could always get rid of the board, go point to point wire, and replace the cheese parts with better versions.

Or, keep the board, and replace the crap parts (see many) The resistors (first, get rid of those crap sand blocks!) then go to caps. Inductors will be hard to figure out the value of you don’t have the proper tools to read inductor value! Plus you would want to do the measurements out of circuit. You may find that a simple rebuild of better parts will gain you some better sound your looking for.

Have to go re-read your graph.

Now, I’m not speaking for anyone at all, I’m just trying to give small direction that could be helpful.

[Shives]

How are you measuring this? In your room? Outside?

To see the speakers measurements you could take this outside away from anything and try to measure.

Well, I guess the first question.. how are you measuring? A lot of the time drivers polarity can be swapped for phase relation, and timing. As well, can smooth out the response.

[77SunsetStrip]

It is highly likely better quality components will result in a different sound experience.  There will likely not be a significant change in the FR curve.  Only you can decide if any difference is to your liking.  Air core inductors and good poly caps will require much more space.  For example, a board for woofer and a second for mid and high components.

Cyril Bateman did extensive analysis of capacitor types back in the 1970s. He showed different types of capacitors do perform differently in circuits.  One clear conclusion reached was electrolytic capacitors should not be used in audio crossovers.  So why do they appear in so many speakers regardless of retail price.  They are cheap and compact in size.  Manufacturers also know the vast majority of buyers will never disassemble and modify their speakers.  If a buyer wants better sound quality, the manufacturer always offer a more expensive model claimed to be better - still with cheap parts.

A well regarded speaker designer from the 1980s-1990s, John Bau, openly discussed the price point pressures of speaker design.  His designs were at the time, still are, highly sought after.  In an interview, Mr. Bau suggested replacement of electrolytic caps with poly caps because of deterioration.

My experience covers upgrading three different speakers - inexpensive, mid-level, and multi-thousand dollar.  All had electrolytic caps, iron core inductors and sand cast resistors.  The most expensive speaker crossover was no better in quality than the inexpensive model.  All three speakers had a good sound for their price point.  The most expensive claimed the flattest FR curve, with an accurate sound quality.  Unfortunately sterile in nature making every type of music sound lifeless.

All three were upgraded with good quality parts, not expensive or boutique.  All three produced more natural sounding music with wider dynamic range.  A grand piano sounded grand, not like a tinkly electric.  Cymbals were crisp and clear.  Bass full, no longer muddy.  These were differences noted by comparing modified to un modified speakers each time.

If you want to tackle modifications, it will be a learning experience and hopefully enjoyable results. 

 

[DannyBadorine]

I build my own speakers and design my own crossovers now, but my journey began by replacing crossover parts with higher quality caps, resistors and inductors.  I replaced with the exact same values but higher quality parts and I was blown away by the difference.  It's worth it if you want a fun project and want to learn more about how everything works.  But as someone mentioned above, better parts will take up more space.  Figuring out all of the logistics is part of the fun.

[delebru]

Thanks for the responses, I really appreciate it!

The measurement from my previous post was done in my lounge and contains room interactions. The capacitor's and resistor's values on the crossover circuit is what's specified on them, driver's DCR and coil values were measured with a multimeter (DigiTech QM-1552) which is supposed to be +/-3% accurate.

I know the components in the original crossover are far from ideal, I was thinking on throwing it away (well, putting it aside :P) and building the new one from scratch with only poly caps, air core inductors, good resistors and point to point wiring. I'm not worried about the final size, the floor standing speaker should have plenty of space. What I'm worried about is spending a decent amount of money to rebuild what may be a poor crossover design, though from your comments I'm starting to feel it may be worth a try.

I have a UMIK-1 but very little experience with REW. I'm hoping to do some gated response measurements and get useable data to help with the decision. If you have any ideas for useful tests please let me know.

[Danny Richie]

Wow, those are really budget level crossover parts. There is nothing there worth reusing.

I'd replace all of the parts value for value using some air core inductors, better resistors, and poly caps. You only need a small 3.9uF value for the tweeter circuit so I would use a Sonicap there. I would also by-pass that big 120uF cap value atr the front of the mid circuit with a small by-pass cap.

It is also normal for the polarity of the lower woofers to be flipped in order to keep them in phase.

I stock all of those parts and can put them together and ship them to anywhere.

I would also replace all of the wiring and the binding post with some tube connectors. That will really help as well.

[MattSelinger]

I recently rebuilt the crossovers for a DIY mass loaded transmission line speaker kit I built 3 years ago (not one of Danny's kits).  The crossover components that came with the DIY kit were mid(ish) level components (nicer than what you have in your Focals for sure though).  I replaced the mid level capacitors and resistors with higher quality and also added a bypass capacitor for the tweeter circuit.  The air core inductors that were part of the kit were reasonable ones - so I kept them.   This was all stuff I learned by building a couple of sets of Danny's speakers.

The results were pretty astounding going from mid-level components to high quality components.  I was thinking about selling the speakers - not thinking that anymore.  Now they have much greater clarity, resolution and spatial sound.  I listen to them often now and smile knowing I helped make them better.

Funny thing - once you hear the clarity of really good quality systems/speakers, your ears kind of long for it and notice when it is missing.  For me this is the blessing (or curse) of having built and enjoyed speakers like the X-LS Encores and NX-Studios.

Oh - forgot to mention I did add some norez to the cabinet walls to address some resonance.  You might consider that as part of your upgrade.