Dialling in active crossovers
- Thread starter spiritofmusic
- Start date
A few remarks.
1. The short time DC transients due to switch on switch of many solid state amplifiers will kill your valuable tweeters or ribbons if they are connected directly, bypassing the crossover capacitor.
2. Go to the Stereophile measurements of speakers that sometimes show response of independent units. Try to guess from this picture what crossover frequency, filter type and slope you will have to pick for your DSP crossover. Remember that you have to control phase and delay.
3. Go the excellent Keith's thread about his Acapella system (first post 03-31-2012, 07:19 PM) http://www.whatsbestforum.com/showthread.php?6209-Keith-s-Acapella-system. Consider if you have the time, expertise and patience to carry such job. As far as I understand we are in 2016 and the project is still evolving at full steam.
4. We should understand the motivation and technicalities behind Keith's project and compare them with the possible ones of an Apogee ribbon's project. IMHO such projects are night and day. YMMV.
1. The short time DC transients due to switch on switch of many solid state amplifiers will kill your valuable tweeters or ribbons if they are connected directly, bypassing the crossover capacitor.
2. Go to the Stereophile measurements of speakers that sometimes show response of independent units. Try to guess from this picture what crossover frequency, filter type and slope you will have to pick for your DSP crossover. Remember that you have to control phase and delay.
3. Go the excellent Keith's thread about his Acapella system (first post 03-31-2012, 07:19 PM) http://www.whatsbestforum.com/showthread.php?6209-Keith-s-Acapella-system. Consider if you have the time, expertise and patience to carry such job. As far as I understand we are in 2016 and the project is still evolving at full steam.
4. We should understand the motivation and technicalities behind Keith's project and compare them with the possible ones of an Apogee ribbon's project. IMHO such projects are night and day. YMMV.
Hello Micro
Yes good point even the M2 has a protection capacitor in series with the compression driver to keep DC out.
That is not a good idea. Those measurements are of the complete system so if you have a mid-range in a small sub-enclosure it includes the roll-off from the sub-enclosure. You don't want that included. What you want is just the electrical crossover slope. You need a copy of the schematic and a simulator to see exactly what is happening. You also need to know the DCR of the inductors to model it correctly. You don't need to use LEAP there are many other less expensive software packages and freeware out there. If you don't know exactly what the voltage drives are you are blind and making assumptions about what the crossover is doing. Not a good idea if you are trying to get better performance
Rob
Yes good point even the M2 has a protection capacitor in series with the compression driver to keep DC out.
That is not a good idea. Those measurements are of the complete system so if you have a mid-range in a small sub-enclosure it includes the roll-off from the sub-enclosure. You don't want that included. What you want is just the electrical crossover slope. You need a copy of the schematic and a simulator to see exactly what is happening. You also need to know the DCR of the inductors to model it correctly. You don't need to use LEAP there are many other less expensive software packages and freeware out there. If you don't know exactly what the voltage drives are you are blind and making assumptions about what the crossover is doing. Not a good idea if you are trying to get better performance
Rob
Suggest that people try to use amps that can be bridged. So, if I have two amps, bridged they give me great power while single amping the divas, and I can always insert a crossover in between to try to bi amp. If it does not work, go back to single amp. No extra amp cost.
Detlof at 80 is managing 2 lamm m2.2 bridged for mid bass, plus atmasphere for mids, plus AVM for bass, plus Jadis for tweeters. He will be ashamed to see the younger generation getting their undies in a twist
Detlof at 80 is managing 2 lamm m2.2 bridged for mid bass, plus atmasphere for mids, plus AVM for bass, plus Jadis for tweeters. He will be ashamed to see the younger generation getting their undies in a twist
Probably Ked
I'm going to investigate Gryphon and NATs if I get to a point where biamping Divas is an option
Rich Murry in the US has no reticence going down this route
Tell Detlof my underwear is untwisted
I'm going to investigate Gryphon and NATs if I get to a point where biamping Divas is an option
Rich Murry in the US has no reticence going down this route
Tell Detlof my underwear is untwisted
I, too, found the jumper system in the XVR-1 with its thousands of possible combinations daunting. Set-up would be much easier with variable controls on the front panel of the cross-over. But, as Jack correctly writes, once all values are established there is no need to change them and variable controls can gum up over time and exhibit lower sound quality than fixed jumpers enabling discrete components.
As a theoretical exercise and as a matter of curiosity I am curious to try to figure out a purely passive crossover set-up for Full Ranges. One option is to find original, stock Apogee 4 ohm boxes and simply have someone replace the original discrete components in them with new, higher-quality components of the same electrical values.
I, like Marc, am not big on DIY. I want the speaker's designer to decide these things. I want a product which was designed, optimized and finalized by a designer.
As Kedar suggests why not start as simply as possible and go from there? Marc -- just start with a passive box (no bi-amping initially).
As a theoretical exercise and as a matter of curiosity I am curious to try to figure out a purely passive crossover set-up for Full Ranges. One option is to find original, stock Apogee 4 ohm boxes and simply have someone replace the original discrete components in them with new, higher-quality components of the same electrical values.
I, like Marc, am not big on DIY. I want the speaker's designer to decide these things. I want a product which was designed, optimized and finalized by a designer.
As Kedar suggests why not start as simply as possible and go from there? Marc -- just start with a passive box (no bi-amping initially).
Graz - APOGEE ACOUSTICS
Active Member
I have done this approach with the Soundlab's. It is very easy to create an unbalanced speaker just changing components, even keeping exactly the same electrical values. Higher quality does not mean better sounding in a particular design, particularly with capacitors.
Fortunately I keep a pair of original unmodified complete backplates (cossovers) that that I can swap in two minutes, so I can compare with the original designer intentions anytime.
The best sound comes from having it set up properly -- which might not come from me setting it up myself.
Surely. DIY is perhaps the stronger audio placebo drug, even higher than price!
And no digital is not plug and play. On the contrary, setup of a digital system is very critical, it needs careful fine tuning of the whole system and acoustics.
I would be tempted to say that the analogue easily reached pleasantness is much more adequate to DIY ...
+1
thus my first post... One has to know what one is getting into... it will not be a walk in the park.
When I was designing and building my BG planar/hybrids, I use a Behringer DCX2496 for design and went through a rather huge number of configurations with Frequency, filter type, relative driver level and phase/time alignment. Once I had completed a few months of testing (and remember this was only a 2-way design) I then sought out a better sounding solution to the Behringer. I bought an Accuphase F25 from Japan with a crossover frequency that was close to what I found optimal and a slope that was preferrable (2nd order). Phase was not controllable but with a non-time coherent filter it was not relevant anyway. With that system I found a 300Hz 2nd order butterworth was optimal to couple with the 10 inch woofer system I had designed. The Accuphase was also level adjustable for each driver...a very nice analog xover.
So, how did I tune the system? I had two systems, one was a computer based system that did both bursts (time converted to frequency) and frequency sweeps and later I had a Behringer DEQ2496, which has a pretty good 61 band digital RTA. The latter was more flexible and allowed faster evaluations. I could get good steady-state FR curves with the built-in pink noise generator. It was easy enough to evaluate through the crossover region. However, listening proved in the end more beneficial because the blending of drivers was not obvious from the measurements once the marriage of the two drivers was flat through the crossover region. Then the slope and individual character of the drivers led to the final decisions. The best was 300Hz 2nd order, but 1st order 500 Hz also had potential and sounded quite coherent. However, allowing the bass driver to do so much midrange led to greater and unacceptable coloration. Going even lower like 180Hz 3rd order led to a discontinuity to the sound and a thinness in the lower mids that was also unacceptable. All-in-all a complex process but a great learning experience
Perhaps an easy way to get this done with relative ease would be an Accuphase digital xover. They are not cheap but you are getting a good xover with good DACs and still the flexibility of digital.
I can only speak to the Pass Labs XVR-1 product. Generally the Pass Labs factory spends time with the end-user and tries to send the unit out of the factory with the "best-guess" starting point documented and established. For units that are out in the field, the same sort of process. However in this case the settings are usually documented and if necessary presented as jpg's for the end-user to set.
Keep in mind these are usually only suggested starting points. With all the software available today it's fairly trivial to sim up a number of likely looking crossover topological solutions, which in practice frequently measure well. How well they actually sound in a real acoustic space is another issue. And there in lies both the fun and challenge of electronic crossovers.
Cyclotronguy
Keep in mind these are usually only suggested starting points. With all the software available today it's fairly trivial to sim up a number of likely looking crossover topological solutions, which in practice frequently measure well. How well they actually sound in a real acoustic space is another issue. And there in lies both the fun and challenge of electronic crossovers.
Cyclotronguy
For a guy working with cyclotrons, perhaps, but not to the average audiophile... At least IME/IMO. - Don
Naw, just like being an audiophile.
RULE OF THE FIRST Keep "off' button in reach and one eye on the meter.
And from there: Push buttons, turn dials till you get what you want.
Stock pile lots of wire and solder and sometimes put your faith in things nobody understands.
Cyclotronguy
RULE OF THE FIRST Keep "off' button in reach and one eye on the meter.
And from there: Push buttons, turn dials till you get what you want.
Stock pile lots of wire and solder and sometimes put your faith in things nobody understands.
Cyclotronguy
This is interesting, Marc. Perhaps Henk or Rich would take on the duty of liaising with Pass and specifying to Pass exactly what Henk or Rich calculates should work, and so you would receive directly from Pass a cross-over configuration which at least puts you on target.
Similar threads
