I'm currently collecting the remaining parts for my speaker project (WD 25T, dynaco clones from world designs) and would like to drive them active.
I've all but eliminated digital crossovers as I still run a record player. I'm currently tempted by offerings by Marchand and would consider other options if they were worthwhile. I know of the old KlarK-Teknik (dn800 iirc) which is quite rare and the Bryston units too. I'm looking for a 2 way XO with 2KHz crossover frequency.
The XO has to have balanced inputs/outputs. I would also like notch filters if that makes any sense and possibly step baffle correction if needed.
The amps are ARC D-130 (sand amp) for the lows (sensitivity for full output=1.71 volts) and ARC VT-50 (tube amp) for the highs (sensitivity for full output=1.95 volts).
My budget hovers around 2K$ for the active XO and any recommendations are appreciated.
I've all but eliminated digital crossovers as I still run a record player. I'm currently tempted by offerings by Marchand and would consider other options if they were worthwhile. I know of the old KlarK-Teknik (dn800 iirc) which is quite rare and the Bryston units too. I'm looking for a 2 way XO with 2KHz crossover frequency.
The XO has to have balanced inputs/outputs. I would also like notch filters if that makes any sense and possibly step baffle correction if needed.
The amps are ARC D-130 (sand amp) for the lows (sensitivity for full output=1.71 volts) and ARC VT-50 (tube amp) for the highs (sensitivity for full output=1.95 volts).
My budget hovers around 2K$ for the active XO and any recommendations are appreciated.
Last edited:
Unless you have a reason for not going digital other than idealistic, I would use a ULTRADRIVE PRO DCX2496 from Behringer, you can get one of these for about $400. It uses some of the best DACs and ADCs available and has every concivable adjustment you could ever need. It has an ADC so will take your LP input and convert it to digital.
Then spend the rest of your $1500 on a decent measurement microphone, a preamp and a good speaker meausrement program - CLIOWIN would be a good option. You could also get a speaker simulation package there are some free ones such as speaker workshop or you could buy LEAP which is the industry standard (They are all difficult to use, but leap is a bit less difficult than Speaker workshop). You may get away with not buing a simulator as with the berhinger xover you can make changes so quickly you could work in the "real world".
I belive you will get a better result this way as a very significant part of a speakers sound is the quality of the alignment you achieve. It is very unlikley that a generic PA crossover will work without measurement and even then it may be very difficult to get the drivers to intergrate as you often need to include the acoustic roll of in with the crossover roll off (i.e the two electronic slopes are not the same).
If you must go analogue, I had a look at the Manchard crossovers and only the top of the range one alows you to have different slopes on the LP and HP sections, I consider this essential for a HiFi speaker set up unless your region of overlap is massive (Which is unlikely in a two way design). It has to be configured by swaping out the resistors which will take alot of time. You will still need a measuremnt system to get it right.
Alternatively a possible brute force method might work if you went for the 48dB per octave filter Manchard make. This might be fast enough that the region of phase problems would be sufficiently small that it wouldn't upset the sound much. I have never tried this approach so can't say for sure. Maybe someone else who has coulds chip in here with a opinion
. Opinion on very steep filters is split many people don't like then some rave about them.
Sounds like a fantastic project though and I envy you your budget!
Regards,
Andrew
Then spend the rest of your $1500 on a decent measurement microphone, a preamp and a good speaker meausrement program - CLIOWIN would be a good option. You could also get a speaker simulation package there are some free ones such as speaker workshop or you could buy LEAP which is the industry standard (They are all difficult to use, but leap is a bit less difficult than Speaker workshop). You may get away with not buing a simulator as with the berhinger xover you can make changes so quickly you could work in the "real world".
I belive you will get a better result this way as a very significant part of a speakers sound is the quality of the alignment you achieve. It is very unlikley that a generic PA crossover will work without measurement and even then it may be very difficult to get the drivers to intergrate as you often need to include the acoustic roll of in with the crossover roll off (i.e the two electronic slopes are not the same).
If you must go analogue, I had a look at the Manchard crossovers and only the top of the range one alows you to have different slopes on the LP and HP sections, I consider this essential for a HiFi speaker set up unless your region of overlap is massive (Which is unlikely in a two way design). It has to be configured by swaping out the resistors which will take alot of time. You will still need a measuremnt system to get it right.
Alternatively a possible brute force method might work if you went for the 48dB per octave filter Manchard make. This might be fast enough that the region of phase problems would be sufficiently small that it wouldn't upset the sound much. I have never tried this approach so can't say for sure. Maybe someone else who has coulds chip in here with a opinion
. Opinion on very steep filters is split many people don't like then some rave about them.Sounds like a fantastic project though and I envy you your budget!
Regards,
Andrew
The above is an excellent suggestion, if you are looking for the ultimate in control I would suggest adding to this the Behringer ULTRACURVE PRO DEQ2496 as well which will provide you will real time analysis and eq correction capability and still be well within your budget
Last edited:
I would seriously consider the DCN23 kit from Ground Sound
See:
Ground Sound
The DCN23 is a digital crossover with a lot of very user freindly features. For instance you can change crossover configurations "on the fly", that is, while the system is 'on', which I find very convenient! The excellent software also allows to incorporate driver measurements directly as a basis for the filter configuration.
It is well within your budget. It is more expensive than the Behringer, but it also easier to implement, and I would asume that the sound quality is better too....
Best regards
Peter
See:
Ground Sound
The DCN23 is a digital crossover with a lot of very user freindly features. For instance you can change crossover configurations "on the fly", that is, while the system is 'on', which I find very convenient! The excellent software also allows to incorporate driver measurements directly as a basis for the filter configuration.
It is well within your budget. It is more expensive than the Behringer, but it also easier to implement, and I would asume that the sound quality is better too....
Best regards
Peter
If you get a DCX2496 make sure that you leave budget for modifying it... when used with analog in, the one i had did not sound very good at all.
For maximum analog goodness, a buffered version of the passive line level XO from Marchand might be worth asking after,
And since you are doing the WD set, let me point out the Classic TL for these drivers at the bottom of this page:
Transmission Line Speakers
dave
For maximum analog goodness, a buffered version of the passive line level XO from Marchand might be worth asking after,
And since you are doing the WD set, let me point out the Classic TL for these drivers at the bottom of this page:
Transmission Line Speakers
dave
I have a Behringer CX3400, an analogue active xover which has balanced inputs and outputs. Its easy to use but you're stuck with 4th order (24 dB) Linkwitz-Riley filters. However it doesn't have provision for notch filters or baffle step correction. I considered the Marchand route but it was way out of my price league and you're still stuck with LR filters that have fixed filter slopes (in Marchand's case either 24dB or 48dB per octave).
To get the kind or flexibility you seem to be looking for you have to go digital. I thought about buying the DCX2496 at the time but had a really difficult time getting my head around what the system would look like. With the CX3400 (or other analogue xovers), the xover is between the preamp and the power amps. With the digital DCX2496 I think you have to have a multichannel volume control after the xover and then some type of input selector (in your case with a LP input) ahead of the DCX2496. For me this additional cost (and complication) scared me away.
To get the kind or flexibility you seem to be looking for you have to go digital. I thought about buying the DCX2496 at the time but had a really difficult time getting my head around what the system would look like. With the CX3400 (or other analogue xovers), the xover is between the preamp and the power amps. With the digital DCX2496 I think you have to have a multichannel volume control after the xover and then some type of input selector (in your case with a LP input) ahead of the DCX2496. For me this additional cost (and complication) scared me away.
Last edited:
Thanks for all comments guys, I really appreciate it. Basically, I want to stress flexibility is not all that important. I may not need step baffle correction. As far as notch filtering is concerned, what has me interested in it is the possibility of emulating the BBC dip in the 4KHz region (as per Fletcher/Munson research for Bell Labs) and simply controlling the possibilities of the hump in the mid bass region (80-160 Hz). If a notch filter can't accomplish any of these functions, then I don't think I'd need it. Particularly in the treble frequencies, I can see the advantage of such a filter to tame not only the tweeters but also the different presentation of tube/sand amps. If that makes sense.
I still want to keep it simple and once "honed in" leave the controls alone.
The scary part with digital XO's is how an analog input that is transformed into the digital domain and then reconverted back to analog may create more problems and losses than a basic analog XO. I'm afraid that all the micro details could be lost in the transformations of the original signal. I'm really looking for transparency over flexibility.
I still want to keep it simple and once "honed in" leave the controls alone.
The scary part with digital XO's is how an analog input that is transformed into the digital domain and then reconverted back to analog may create more problems and losses than a basic analog XO. I'm afraid that all the micro details could be lost in the transformations of the original signal. I'm really looking for transparency over flexibility.
I use a couple of BSS FDS360s.
With those it is possible to use all sorts of different slopes, including asymetrical and overlapping ones, there is also provision for shelving or parametric eqs.
The problem is that they are only available s/h and the original TL072 op amps should be replaced by something better. I use BB 2134s with great success although there are better (and more pricey) alternatives these days.
The thing that scares me away from Behringer products is that they are almost invariably badly built cheap rubbish and no professional would be satisfied with the sound quality.
With those it is possible to use all sorts of different slopes, including asymetrical and overlapping ones, there is also provision for shelving or parametric eqs.
The problem is that they are only available s/h and the original TL072 op amps should be replaced by something better. I use BB 2134s with great success although there are better (and more pricey) alternatives these days.
The thing that scares me away from Behringer products is that they are almost invariably badly built cheap rubbish and no professional would be satisfied with the sound quality.
Ouch -- "The thing that scares me away from Behringer products is that they are almost invariably badly built cheap rubbish and no professional would be satisfied with the sound quality."
With the kind of outlay Charles Darwin is talking about (almost 10x the cost of the cheap CX3400) you'd be better off going with the dbx DriveRack 260 (roughly $1000 US). It has in addition to being a crossover, it alos provides alignment and zone delay, graphic and parametric EQ, and a fulltime RTA.
With the kind of outlay Charles Darwin is talking about (almost 10x the cost of the cheap CX3400) you'd be better off going with the dbx DriveRack 260 (roughly $1000 US). It has in addition to being a crossover, it alos provides alignment and zone delay, graphic and parametric EQ, and a fulltime RTA.
I found the generic Analog Behreinger to be better than I expected, fine low pass for the subs, but not up to really high end snuff. Same with my DCX. I only use it on my test bench. They are really cheap, but designed well. With the new National Op-Amps, then the boards from ESP designed for 5532's become a really good choice, provided you know what points you need to start with. You can do a really good active ground power supply and go crazy.
Look at the values. Passive line level is not a solution. I have seen modified amps where the input buffer stage was made into a filter. DIY you can do all kinds of funny things!
Wasn't that bad: I bought two BSS 360s for £100ea. and than got lucky on a tube of 50 non-RoHS compliant BB2134 for just under £50.
How about Ashly analog..XR series...?
I am going the XR4001 in a week or so. The op-amps are socketed, so you could change them if not happy. You can get a new xr4001 for 400.00 if you look. the XR2001 or even the xr1001 are much cheaper. I heavily thought for DCX & BBE ds48, but on pro boards the instilled un-natural artifacts they are reported to introduce has scared me off. I can not afford high end digital so analog it is.
Joe
Joe
I am going the XR4001 in a week or so. The op-amps are socketed, so you could change them if not happy. You can get a new xr4001 for 400.00 if you look. the XR2001 or even the xr1001 are much cheaper. I heavily thought for DCX & BBE ds48, but on pro boards the instilled un-natural artifacts they are reported to introduce has scared me off. I can not afford high end digital so analog it is.
Joe
Joe
I've used an XR-1001 for several years. Works quite well, though the op amps are soldered in and run off 18V. DIY mod would be to change the regulators to 7815s, install sockets, and drop in LME497x0 parts to replace the ancient ones. I've never gotten around to it though.
The caution I'd make is no box crossover I'm aware of offers linear phase. If you're interested in transient perfect it's either DIY from a rather limited set of analog options or DSP with FIR phase corrected IIR.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.