Planar Speaker Asylum

Welcome! Need support, you got it. Or share your ideas and experiences.

Return to Planar Speaker Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Subwoofer crossover points with ESL57

208.46.240.4

Posted on March 7, 2016 at 10:26:38
FlyCast
Audiophile

Posts: 65
Location: New Haven
Joined: September 5, 2015
Looking to add a pair of subwoofers to the rebuilt ESL57 I just received. Any suggestions for crossover points/slopes? Thanks

 

Hide full thread outline!
    ...
RE: Subwoofer crossover points with ESL57, posted on March 7, 2016 at 11:27:00
pictureguy
Audiophile

Posts: 22597
Location: SoCal
Joined: October 19, 2008
Subs with built-in plate amps will have a low-pass crossover built in. Most of THOSE will be 24db/ octave.

I would also recommend, as a GENERAL RULE, to ALSO restrict the lows TO the panels. Set the crossovers maybe 10 to 15 hz apart, so there is both minimal overlap AND you make an effort for flat response THRU the passband. This point depends on the natural roll-off frequency of the panels. Let the panels handle the overtones from the bass. I've found panels and cone drivers to produce 'mush' thru the passband and above, due to phase problems maybe. My system with the above system is very musical.

If you are a COMPLETE DIY guy, willing to go thru some trouble, something like the MiniDSP might come in handy. On the mini? Each input channel and each OUTPUT channel will include 5 bands of parametric EQ. So it might be possible to 'notch out' an annoying room node? Using the 2x8 would also set you up for a later Biamp path.
This is MY next step.
Too much is never enough

 

RE: Subwoofer crossover points with ESL57, posted on March 8, 2016 at 12:41:44
used-hifi
Audiophile

Posts: 1100
Location: Surprise AZ
Joined: March 18, 2003
shame your far away i have the perfect subwoofer for your 57's

RH labs rs1

 

Have a look through this thread ... (nt), posted on March 8, 2016 at 15:15:43
Posts: 2794
Location: Orange Co., Ca
Joined: September 19, 2001
nt

 

RE: higher than you think, posted on March 10, 2016 at 09:29:46
BigguyinATL
Manufacturer

Posts: 3475
Joined: April 10, 2002
The first is to play with with your ESL's. You are going to need a HP crossover anyway - start with a 6dB octave slope and creep the frequency up and then down(along with your volume control) to find out where the dynamics of the main panel start to seem OK & not excursion limited. I like to use a single loudspeaker and a Mono signal for this - so you eliminate variables. You can switch to the other speaker and compare anytime. Then up to a 12dB slope and try the sane thing - Go to 18dB if you but most rooms support enough that 6 or 12 dB are usually enough.

Now you need your subwoofers - you are going to use two (or more) and they don't have to be the same! I build my own, but you can pick any price point you like. The subs will need an 18dB or 24dB LP filter. (Most do) and a continuous frequency and phase adjust. The sub can be anywhere - but side wall placement is my favorite with dipole panels - placing them at the "Null" wall reflection point. Start with one and adjust with the highest crossover point - use some pink noise (I band limit the pink noise to 3kHz or so) and play it into both the single ESL and the Sub at a moderate level (something that feels comfortable to your ears and the speakers - you are going to be doing this for a while). Raise the sub level until you just hear it contribute to the level level at your listening seat. Tweek the phase control until you get a maximum level. (if the bass sounds a little tonal, try moving the sub a foot or two along the wall) Now back down the frequency until you don't hear the sub anymore. You can iteratively do this several times.

Switch to the other ESL and Sub and repeat. Finally, with a mono signal pink noise, play all the speakers. Make sure all signal seems to be coming from between the two ESL's. if the relative bass jumps a lot or fall off from the individual left and right tests, tweek the phase of the left and right subs a little in opposite directions.

Now you can play some music!




"The hardest thing of all is to find a black cat in a dark room, especially if there is no cat" - Confucius

 

RE: Subwoofer crossover points with ESL57, posted on March 12, 2016 at 15:47:16
JimL
Audiophile

Posts: 3773
Location: New Mexico
Joined: November 24, 2002
I used 63 Hz for the crossover point on mine with 24 dB/octave to the sub and 6 dB/octave to the Quads. The Quads roll off at about 18 dB/octave so in combination with a 6 dB RC filter it behaves similar to a Linkwitz-Riley 24 dB/octave. My sub is a Hartley 24" woofer in a dipole configuration and some custom bass boost to flatten out the response.

 

RE: Subwoofer crossover points with ESL57, posted on March 15, 2016 at 18:30:42
IanLane
Audiophile

Posts: 35
Joined: May 22, 2000
Hi Jim,

I am indebted to you for originally posting this idea. I have used basically the same approach as you with my dipole subs crossing at 80 Hz rather than 63Hz. I guess that your crossover point is undoubtedly more technically correct since mine is very close to the Quad bass panel resonance. I think that the teflon cap giving a 63Hz crossover point for the high pass was not available at the time. I still get a virtually seamless blend even so. With the Quads you benefit with the most transparent high pass caps you can get.

Ian

 

RE: Subwoofer crossover points with ESL57, posted on March 30, 2016 at 17:43:21
sqlsavior
Audiophile

Posts: 55
Location: New Mexico
Joined: June 25, 2013
I run my original ESLs with two modest (HSU STF-2) down firing subwoofers located directly behind each main. The mains run full range, and the subs are crossed over (low-pass only) at about 45 Hz. Each sub runs only its respective channel, so bass is still stereo. The connections are at speaker-level. Blending is superb.

BTW, I tried an Anti-Mode DSP, summed to mono, with them and found that in my room, it didn't help enough to keep it in the system. I'm in an open floor plan, so modes and nodes are minimized naturally.

 

Page processed in 0.025 seconds.