Has anyone ever built a copy of an MT Audio Design XO-2001 crossover?
It was made in Sweden by Matts Tornqvist but is no longer available, the perfect crossover for the QUAD ESL-63/Gradient SW-63 set up.
I do not have the schematic but there are seven ICs in the subwoofer section and two in the main ELS section.
It has many useful functions, two of which are: a pot for setting subwoofer volume level and another (a delay circuit) for phase shifting the sub. A very elegant and noticable function.
It was made in Sweden by Matts Tornqvist but is no longer available, the perfect crossover for the QUAD ESL-63/Gradient SW-63 set up.
I do not have the schematic but there are seven ICs in the subwoofer section and two in the main ELS section.
It has many useful functions, two of which are: a pot for setting subwoofer volume level and another (a delay circuit) for phase shifting the sub. A very elegant and noticable function.
I already did in the past and offered to pay for it but Mats said no. I posted this topic with the hope that the DIY audio community may be able to find a genius to recreate Mats' printed circuit board so that it can be built, before its complete demise.
The XO-2001 is far more superior to the Gradient crossover.
The XO-2001 is far more superior to the Gradient crossover.
The block diagram of that unit looks very straightforward and it's obviously implemented with an op-amp scheme.
I can't imagine there is anything innovative or proprietary to the point of being a big secret.
I have constructed similar active/EQ's a number of times.
Is a DSP unit out of the question?? The type of implementation the XO-2001 appears to have is easily done with a DSP unit in just a few minutes.
Dave.
I can't imagine there is anything innovative or proprietary to the point of being a big secret.
I have constructed similar active/EQ's a number of times.
Is a DSP unit out of the question?? The type of implementation the XO-2001 appears to have is easily done with a DSP unit in just a few minutes.
Dave.
Dave, I understand your point but ESL63 is not the owner of the schematic and he is correct not publicly disclosing it.
A friend of mine with an ESL63-Gradient SW-63 set up and whose ears I trust compared the XO-2001 with a mini DSP and also with the original Gradient crossover. The XO-2001 performed best in his system. I also have QUADS.
Horacio
A friend of mine with an ESL63-Gradient SW-63 set up and whose ears I trust compared the XO-2001 with a mini DSP and also with the original Gradient crossover. The XO-2001 performed best in his system. I also have QUADS.
Horacio
I didn't suggest that he disclose it. But possibly he might hint/suggest to Mats' that Mats' disclose it. If not, that's certainly fine....but then it's just a mystery and warrants no further discussion.
I don't know what "performed best" means, (subjective evaluation I guess) but would suggest that if there was a noticable audible difference then probably the Gradient and miniDSP crossovers were not set up to exactly duplicate the transfer function of the XO-2001 unit.
Cheers,
Dave.
I don't know what "performed best" means, (subjective evaluation I guess) but would suggest that if there was a noticable audible difference then probably the Gradient and miniDSP crossovers were not set up to exactly duplicate the transfer function of the XO-2001 unit.
Cheers,
Dave.
I just talked to Mats.
If you look at the block schematic it´s pretty straight forward, Op based Linkwitz 24dB filters 100Hz crossover frequency.
The dipole compensation is a high pass filter with crossower at 12Hz and then amplified to match the high pass QUAD section.
There is a switch for subsonic but you can skip that part.
The delay adjustment is an all pass R/C filter with a potentiometer, Mats is using a linear pot but that was a mistake! Use a LOG potentiometer instead! It´s quite difficult to adjust the delay with the linear pot...
Voltage stabilisation of course +/- 15-18 volts depending on OP amps.
There is not many secrets. Double sided circuit board.
If you look at the block schematic it´s pretty straight forward, Op based Linkwitz 24dB filters 100Hz crossover frequency.
The dipole compensation is a high pass filter with crossower at 12Hz and then amplified to match the high pass QUAD section.
There is a switch for subsonic but you can skip that part.
The delay adjustment is an all pass R/C filter with a potentiometer, Mats is using a linear pot but that was a mistake! Use a LOG potentiometer instead! It´s quite difficult to adjust the delay with the linear pot...
Voltage stabilisation of course +/- 15-18 volts depending on OP amps.
There is not many secrets. Double sided circuit board.
Go to following link:
Homepage MT Audio Design
On the left and down side, press "MT Audio kits" then you press the blue button XO-2001 and there you have all information on how it was designed.
Homepage MT Audio Design
On the left and down side, press "MT Audio kits" then you press the blue button XO-2001 and there you have all information on how it was designed.
The dipole compensation is a high pass filter with crossover at 12 Hz (?) and then amplified to match the high pass QUAD section. I do not know what this means.
Do not skip the >subsonic switch<, its important to protect the woofer drivers that cannot take the very long excursions that can happen at subsonic frequencies. Will a DSP have this?
The delay adjustment is an all pass R/C filter with a potentiometer, Mats is using a linear pot but that was a mistake! Use a LOG potentiometer instead! It´s quite difficult to adjust the delay with the linear pot. I too do not like log pots, linear pots can be simply modified with resistors to behave similarly to a log pot.
Volt stabilization is two 78xx ICs.
Do not skip the >subsonic switch<, its important to protect the woofer drivers that cannot take the very long excursions that can happen at subsonic frequencies. Will a DSP have this?
The delay adjustment is an all pass R/C filter with a potentiometer, Mats is using a linear pot but that was a mistake! Use a LOG potentiometer instead! It´s quite difficult to adjust the delay with the linear pot. I too do not like log pots, linear pots can be simply modified with resistors to behave similarly to a log pot.
Volt stabilization is two 78xx ICs.
I've owned and sold this setup. Unless you're using the highpass function to drive your ESL63 (which I don't recommend) the DSPeaker 2.0 is audibly superior and offers some measure of room correction.
DSPeaker-Anti-Mode 2.0 Dual Core
I ran my Quads full range - without additional processing from my preamp. Both the preamp and DSPeaker 2.0 were fed a (split) optical input signal.
The DSPeaker drove a separate amplifier for the Gradient subs (which integrated poorly in my room) and later IB subs which had smoother and slightly deeper response.
I have since sold the Quads and Gradient subs to use higher efficiency drivers. I still use the DSPeaker to drive subs.
It's a terrific device. I recommend it below 300 Hz, only. As a full range preamp, it was a step backward in resolution.
Using XO 2001 for the QUAD´s gives you approx 10dB more headroom. Since you cut away the low frequencies the get much more dynamic and the midrange (which is among the wery best if not THE best) really get another dimension. I did a modifikation on XO 2001 and changed the crossower frequency from 100Hz to 150Hz and I can play extremely loud! Something that you really don´t expect when you have electrostatic speakers. I would say the that you gain in resolution, dynamics, 3D, BASS for sure, and pure volume.
I hope someone makes a clone
maybe with some new fancy OP-amp?
I hope someone makes a clone
maybe with some new fancy OP-amp?
DSPeaker 2.0 costs $1,200 new or around $650/700 used, I cannot afford that.
esl 63: this is exactly my point, that the Do it Yourself audio community may be able to find a genius to recreate Mats' printed circuit board so that it can be built, before its complete demise. A long shot, I know.
esl 63: this is exactly my point, that the Do it Yourself audio community may be able to find a genius to recreate Mats' printed circuit board so that it can be built, before its complete demise. A long shot, I know.
My experience was that the ESL63 may have presented less demand on the amp, but the output was veiled. In my opinion, it was not an improvement.Using XO 2001 for the QUAD´s gives you approx 10dB more headroom. Since you cut away the low frequencies the get much more dynamic and the midrange (which is among the wery best if not THE best) really get another dimension. I did a modifikation on XO 2001 and changed the crossower frequency from 100Hz to 150Hz and I can play extremely loud! Something that you really don´t expect when you have electrostatic speakers. I would say the that you gain in resolution, dynamics, 3D, BASS for sure, and pure volume.
I hope someone makes a clone
While this dedicated unit provided proper EQ for the (dipole) Gradient subs, it sounded no better (or worse) than commonly available sub crossovers.
I would look to the NHT crossover mentioned earlier before sinking time into orphaned designs.
Is the DSPeaker expensive? Certainly.
It has also cleaned up the presentation in my little music corner.