TOPIC:

I have used the DDRC 24 in my two active speaker projects were they are the active crossover and then the added DIRAC live filter . This worked very well and there is a more full description in the New projects thread.

However I have now bought a pair of Quad ESL57 speakers and had them fully refurbished to better than new specification. Because of the complication of the passive crossover in the Quads it is just beyond my technical skills to make them an active speaker so I am using the DDRC 24 with crossover and EQ bypassed but using the DIRAC Live . When I used the auto target curve and played the speakers I suffered quite a problem with distortion when the volume was raised and when any music with heavy low bass was played.

With a little help from some other DDRC 24 users on an audio site (www.hifiwigwam.com/forum) I managed to identify that the curve was asking too much of both the speakers and amplifier in the low are up 40 KHZ and was just soaling amplifier power and tryin to make the speakers work in area they were unable to do .

For those interested I have attached the DDRC 24 project files which you can review . Below I have also put a picture of the speakers graph with out any target and then the eventual target filter I used which now works very well and allows me to play these speakers at very loud levels with no strain .

I freely admit that migh ability to read the graphs and then transfer the readings to the best filter target is limited . The filter I use below works and sounds good but it may well be that it could be improved on to give a little extra in the bass without send the unit in to distortion .

Any sugestions of even target filter settings that you can post would be much appreciated . As I have three blank filter options left i can easily load any suggestions in to the DDRC24 give them a try and compare them to the current filter I am using . Thank you for anyone that takes the time to view this and any suggestions offered.

I measured a pair of stacked ESL57s some time ago and realised from the measurement that they have what you might call an "impression of bass" from the bass panel resonance. Yours is higher in frequency than I measured (70 Hz IIRC) but that may be related to the age of the panels. To get much bass, I suspect you will just need subwoofers. You don't say what the default target curve actually did (it varies in the low cutoff) but I'd be inclined to try it again but set the low curtain to just below the resonance.

BTW the files you uploaded are the configuration files not the Dirac Live project. If you wanted someone to look at the latter, you would need to upload to Dropbox and provide a link.

I sometimes regret selling my pair....

@BenCat,

Some great suggestions from John Already.
The rule of thumb when building a target curve is to always stay "around" the natural response of the speaker. Never try to be massively above with the target curve. Else you're indeed over exercising the speaker.. No magic can be done against physics as you learned..

It's a LOT easier to linearize by attenuating than boosting so that's why we always recommend to set the target curve BELOW the response. Have a trial in the future (and also you can click on TARGET curve tick box in the first measurement) to overlay both together. It's very useful information for ppl to review.

DevTeam

How are the speakers placed in the room?
If bass is needed, try placing them right next to front wall (with broadband sound absorption panel wedged between them). This will increase amplitude of your speakers while avoiding SBIR at low frequencies.
Maybe it's just limitation of electrostatic speakers, but can't hurt to try.

I am not concerned or trying to increase the bass output as Quad ESL57 are dipole speakers so deliver the same same sound volume backwards as well as forwards so moving them closer to a real wall even using absortion panels wil just increase the reflected sound reaching the listening position .
What I am trying to do is get a Filter by DIRAC that does not put the MiniDSP DDRC24 in to overload and distortion which is what is happening now . I can and do accept the bass limitations of this design I just want to avoid the distortion that they are currently showing when there is large bass content and high volumes.

Looking at the "before" graph, it should be safe to set -10dB curve down to 40Hz.
Problem is below that frequency is giving severely low response of -20dB
If there is a problem with this unintended +10dB boosting from curve, might need to lower the curve down to -20dB (at 30Hz), or
keep the DIRAC curve and simply add a steep highpass EQ filter in the DDRC-24 program around 40Hz.

Separate EQ on the DDRC might not be what you want, but it'll be worth it if subwoofers are involved.
Personally, I have a subwoofer low passed at 90Hz with my 8" woofer speakers that go down to 47Hz @-3dB bypassed filters.
DDRC-24 can handle dual subwoofers as well, so if concerned about stereo imaging,
each subwoofer can be placed right next to each speaker with Dirac 1/2 assigned separately to left/right subwoofer.
I EQ'd them (and inversed subwoofer phase for my setup) until resulting "before" curve is almost flat around crossover frequency range,
and DIRAC does its magic (both left and right channels can reproduce full range down to 10Hz) without any boosting on the mains .

Thank you @ysungsik appreciate your advice . Unlike many on here I have never wanted or used a Sub Woofer as i have verry little hearing in the lower bass regions .

Also with Quad ESL units they major in speed as the effective driver unit is microns thick and no sub woofer using a standard magnet driven driver has any hope of matching it .

I have done soe additional work yesterday (I am off Friday and Minday each week but my Wife is at work on Mondays so this is the day I can really have good look at issues.) As a result I have modified the curve and paired both left and right speakers . Results are very good indeed . No head room issues as yet and Bass is what I would expect for a Quad ESL clear with lots of information and slam and in proportion to the Mid and Treble. This is exactly what I want and Iam now very happy with what the whole system is producing . I have listened for about thre or four hours using both my own music from USB HD via a streamer and also listened to Radio Paradise Flac version and BBC 6 Music at best mp3 rates. The music is flowing and that Quad trick of making voices sound different to what you are used to is very much in evidence . Spoken word by presenters shows of the naturalness of the speakers and makes you realiase what you normally listen to is coloured by the cabinet.