Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 11, 2012 16:18:45 GMT
With the AClass amp build project nearing the launch pad I am now also thinking of my possibilities with an active crossover. I have been mooching around the interent looking at various designs, some don't go high enough e.g. max 2khz xo point. ( I need 2.2khz min. ) Others seem overly complicated and yet more wildly priced. The Jaycar is a three way so would need modifying and I'm not sure of its availability. The one I keep getting steered back to is the ESP Linkwitz-Riley. Is this still a good bet or can I do better without going mad on price. (Speakers are Naim SBLs) EDIT= there is also this Ebay jobby, too basic?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 0:43:12 GMT
Okay Chris looks like you've just posted my next project lots of reading up to do. Perhaps Frans or Alex have something up their sleeve? Alan
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 1:03:47 GMT
Okay Chris looks like you've just posted my next project lots of reading up to do. Perhaps Frans or Alex have something up their sleeve? Alan Hi Alan No, I don't have any plans along those lines. I am a firm believer in keeping the electronics to a minimum if possible, and start off with better than average speakers. Any active XOver design is going to degrade what you already have coming out of the Class A HA/preamp. Kind Regards Alex P.S. David2vk from Sydney, for example, further improved the sound of his big (and expensive) Infinity speakers by using quite expensive aftermarket XOver components that cost more than the full cost of most people's speakers.It's very hard without suitable test equipment to fine tune the balance between sections. Dave tried that on one occasion,, and he has bloody good hearing, but I felt that his speakers didn't sound as "together" as at the previous listening session.
|
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Jul 12, 2012 1:52:00 GMT
Oh, I'm also looking out for one too to go REAL triamp but now currently stuck at computer AV netwerk setup. The Linkwitz-Riley XO has a large following in Singapore for those into multiple drivers like me. To me multiple drivers array sounded more Wonderful, Live and Real overall than those pesky low wattage but efficient "SET" drivers and configurations except for lacking that bit of last order coherence to the musical picture. Well, maybe Alex likes "SET" type of sound. Btw, is there a passive version of that? That will be the best of the best. Btw, with upstream XO, the amps are obviously better able to control the drivers than with a downstream one. Well, you know all those tough phase and impedance loads with a downsteam XO. Wonderfull, Chirs. You dug up a topic that hit my "bull's" eyes too. Let's share here.
|
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Jul 12, 2012 2:02:08 GMT
Arrghh............, that Elliot's XO is with op amps. Any discrete version you have found?
|
|
rowuk
Been here a while!
Pain in the ass, ex-patriot yank living in the land of sauerkraut
Posts: 1,011
|
Post by rowuk on Jul 12, 2012 5:04:46 GMT
That would be 5x dual triodes (12AU7) if you wanted to do it right........
I am suspicious of Mr. Linkwitz. He likes dipoles........
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 6:36:56 GMT
BD says his gear is completely transparent and he can't hear the difference between the original and his recording. Can it become any better than that and do you trust his ears/judgement ? Good speakers (extremely expensive planars) and a power amp that can drive it is essential. Just like the whole recording chain (mic placement/usage/choice) and post 'processing'. :(Obviously 20 bits delta Sigma 1 bit conversion (AK5394/AK4395) and common bipolar opamps like the NE5532 and NJM4580 are good enough to be 'transparent' and thus add no coloration nor change anything which is good news to those who doubt wether to use 'sound degrading' opamps or choose to add much more harmonics. In the MIC channels OPA228 and TLE206x (J-FET based) opamps are obviously good enough. Fortunately also SMPS (3 of those inside the box) do not seem to be degrading the sound even on mic inputs. I would simply go for the ESP circuit. You can even buy the PCB's ! Myself I use my own design active filter but only for the lows. It can be adjusted with pots to create the Xover I like (steepness/freq) but the ESP circuit will do. Wether an active filter gives improvements over passive filtering is very dependent of the execution of both filter designs. Most don't realise this, but when using a phase-linear filter the tweeter and acoustical centre of the woofer/midrange should be alligned in the depth plane (viewed from the listening position) also to get the real benefits. Otherwise the linear phase is acoustically ruined by acoustical phase shifts. Personally I prefer NO crossover between 200Hz and 6kHz at all (actually no XO above 150Hz at all )
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 7:18:58 GMT
I accept that Barry is hearing no differences with his own playback chain.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 7:21:07 GMT
Attachments:
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 7:55:41 GMT
Ah well... the NE5532 and NJM4580 ARE used in BD's 20 bit resolution playback chain They are safely tucked away in his ULN-8 together with 3 SMPS's and other 'standard' grade components.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 8:31:06 GMT
Ah well... the NE5532 and NJM4580 ARE used in BD's 20 bit resolution playback chain They are safely tucked away in his ULN-8 together with 3 SMPS's and other 'standard' grade components. It's still much, much better, than going through numerous "cheap as chips" opamps designed around 30 years ago, in tandem in typical recording studio mixers ! I bet they aren't using typical consumer grade SMPS either.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 8:38:22 GMT
From the looks of it... they are:
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 8:48:08 GMT
Perhaps they are for professional stuff, but I have yet to see a SMPS with 4 x large electros in normal domestic consumer gear. Anyway, none of this has anything to do with the Topic. Chong would like to know if there is a suitable DISCRETE active crossover design that he could try.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 10:41:12 GMT
Since my DIY TL's are some Thirty Plus Years Old would it be more appropriate for me to rejuvenate their crossover components? Lost all the data I had when I built them so I'll have to open the beasts up and have a squint. As is I like the sound I have. I hear you say why change it then. Maybe fresh components will give it a small lift. Who knows? but I'd like your comments. Regards, Alan
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 11:08:23 GMT
Hi Alan It certainly wouldn't hurt to replace any electros.If the values in the tweeter section are low enough, you may be able to replace them with polyprops ? Regards Alex
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 12:32:33 GMT
Thanks Alex, I'll have a peek, the rain's off today I'm torn between ripping my vinyl, woodwork and hedge cutting. Regards, Alan
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 13:12:48 GMT
Alex,
I hear you loud and clear, for many 'speakers this is the case.
However, for good or bad reasons the Naim SBLs are VERY upgradable by going active. This may, of course, equally mean they are just crap passive ;D
Anyhoosen, I've previously run an active set and they were quite stunning, so active it will be again.
Like Chong, I had discrete in my mind but have thus far failed to find a compatible design.
|
|
XTRProf
Fully Modded
Pssst ! Got any spare capacitors ?
Posts: 5,689
|
Post by XTRProf on Jul 12, 2012 13:19:50 GMT
Hi Alan, Usually those speaker XOs use values less than 10uf. So change all to PP as Alex had advised will be good. I had changed mine and there is a big difference. Very obvious is the less grain. I was about to change the inductors too to self wound OCC copper ones when I decided to go triamp XO. The resistors too to metal film. But now that's over as I wanted triamp XO now. If you are adventurous enough and bank proof without robbing it, you can also try film teflon caps.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 13:42:47 GMT
Alan and Chong, Yes, I've also been down the better components on the PAXO. Standard SBL PAXO Tinkered version (not mine) My (but not my work) super dooper, pooper scooper versions A definite improvement but not even close to what I know active can deliver
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 14:45:43 GMT
How hard can that be ? Simply replace each opamp of the ESP design with a simple diamond buffer circuit....
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 16:01:57 GMT
^^^^^^
Do you have a version in mind that would work well in this application as well as physically fitting?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 16:52:39 GMT
If you want it on the available PCB only simple emitter followers would be easy to do. output coupling caps will be needed.
each opamp section must be replaced by a transistor with Emitter resistor. You can even go for J-fet source followers. Opamps, however, will always outperform unless you don't think they will, then they won't.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 17:57:36 GMT
He builds us up; and then smashes us down; Frans, you are a big meany The reason I had discrete at the back of mind was my old active crossover, an Exposure V, was discrete and part of the previous active system I had which I absolutely loved. These are now like rocking horse poo and change hands for alarming sums of money. When John Farlowe departed Exposure he, apparently, took all the designs and rights with him. I do wonder how many have since been reverse engineered by the Exposure lads but they are staying mum
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 19:00:47 GMT
In any case an active XO, be it discrete or opamp based, will measurably/audibly outperform every passive XO, no matter how well made the passive one is for a number of reasons. Even with emitter followers a distortion level of 0.01% can be reached, with the better opamps this can 100x (40 dB lower THD) to 1000x (60dB lower THD) better. Bandwidth of the highpass will easily reach in the MHz zone. Ofcourse there will be no magic designers/designs/topologies or components in the 'stock' ESP filter, as Elliott doesn't believe in those. These you will have/can to add yourself though. The filter itself can be seen as 'more than adequate' even with emitter folowers.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 12, 2012 21:42:29 GMT
The audible part is a belief held by those EEs that believe all opamps sound the same, and are transparent. Many people will say that different types and brands of opamps have their own audible signature. Don't forget they need additional power supplies too,as well as extra interconnects and sockets, which MAY result in low level hum if not very careful.That's quite a bit more expense when cased up properly too, and the potential to adjust for a more "pleasing" sound that may no longer be accurate. P.S. Are you able to accurately measure the the effects of the existing Xover's frequency response dips and peaks , and then adjust the new electronic Xover to correct this, other than relying on your ears ? With cheap speakers that may not matter that much, but with expensive speakers using high quality Xover components, you may be foolish to think that you can do better than the original designer, without REALLY knowing what you are doing.
|
|