Thought I would post my recently completed FAST project, I would appreciate any feedback or suggestions in particular around the crossover.
System consists of Alpair7.3 in DaveD designed Slim GR cabinet and pair Aurum Cantus AC250/752C2 10" woofers using a miniSHARC DSP for crossover. Crossing over at 125hz 2nd order (phase reversed on woofer) and incorporating Linkwitz transform to bring the woofer F3 to 30hz, I was thinking of going to 25hz but initially protecting the woofers as the xmax is only +/-8mm, they are woofers not sub woofers. At a later date i will be taking some measurements and tweaking.
Re choice of drivers, I am a BIG fan of Mark Audio drivers and having tried 3 other MA offerings the Alpair7.3 is my favorite. The Aurum Cantus drivers came up on a local auction site, 4 drivers, new and unused at about 1/3 the cost of importing from Parts Express, the bargain hunter in me couldn't resist. My thoughts in using 2 per channel (as does the top Aurum Cantus system) were that dynamics may be improved as the drivers only had to move 1/2 the travel that a single one would for same frequency and db. The woofer cab is a 65 litre heavily braced sealed box, without correction the F3 would be 58hz.
Only completed the project some days ago (Alpair7.5 already have 1-200 hours on them) but very pleaded with the sound so far. The MA drivers sound fantastic relieved of their duties below 125hz and the bass is tight and well extended, I have another FAST system with SDX10 subs extending to 25hz and I am not at all sure I can tell the difference between f3 of 30 and 25hz.
I do like to listen to electronic music loudly on occasion and this system really responds well to my musical tastes!
System consists of Alpair7.3 in DaveD designed Slim GR cabinet and pair Aurum Cantus AC250/752C2 10" woofers using a miniSHARC DSP for crossover. Crossing over at 125hz 2nd order (phase reversed on woofer) and incorporating Linkwitz transform to bring the woofer F3 to 30hz, I was thinking of going to 25hz but initially protecting the woofers as the xmax is only +/-8mm, they are woofers not sub woofers. At a later date i will be taking some measurements and tweaking.
Re choice of drivers, I am a BIG fan of Mark Audio drivers and having tried 3 other MA offerings the Alpair7.3 is my favorite. The Aurum Cantus drivers came up on a local auction site, 4 drivers, new and unused at about 1/3 the cost of importing from Parts Express, the bargain hunter in me couldn't resist. My thoughts in using 2 per channel (as does the top Aurum Cantus system) were that dynamics may be improved as the drivers only had to move 1/2 the travel that a single one would for same frequency and db. The woofer cab is a 65 litre heavily braced sealed box, without correction the F3 would be 58hz.
Only completed the project some days ago (Alpair7.5 already have 1-200 hours on them) but very pleaded with the sound so far. The MA drivers sound fantastic relieved of their duties below 125hz and the bass is tight and well extended, I have another FAST system with SDX10 subs extending to 25hz and I am not at all sure I can tell the difference between f3 of 30 and 25hz.
I do like to listen to electronic music loudly on occasion and this system really responds well to my musical tastes!
Attachments
Hi Dave,
The F3 is based on the Linkwitz transform (spreadsheet based) you feed into the miniSHARC rather than measured. From that F6 looks to be about 23hz and F10 about 18HZ
I was assuming twice the surface area thus 1/2 the travel required to move a similar amount of air. Well there is certainly very little observed movement unlike the SDX10 that jumps around in a big way with heavy bass passages.
By the way vaguely thinking of getting a pair of FF85WK drivers to try with the sub unit, thoughts on how well they would work with a crossover around 250hz? they are relatively cheap and seem to be in favor for such an application.
Room gain will affect your end results, one shouldn’t try to go too flat or you end up with excess bass. F3 is useful for filters, not humans, F6/F10 give a much better idea of what you hear.
Are the SDX10 sealed? FF85wk are quite nice, i have been listening to an EnABLed set in the Tysen V2 prototypes. With the passive XO they are crossed at about 450 Hz but when i hook up the PLLXO i use the 240 Hz XO point… we originally tried 180 which sounded good but a little higher was bette. SDX10 are a nice “sub” as they have decent extension on top.
dave
Are the SDX10 sealed? FF85wk are quite nice, i have been listening to an EnABLed set in the Tysen V2 prototypes. With the passive XO they are crossed at about 450 Hz but when i hook up the PLLXO i use the 240 Hz XO point… we originally tried 180 which sounded good but a little higher was bette. SDX10 are a nice “sub” as they have decent extension on top.
dave
Thanks for your comments Dave. Understand what you are saying re F3 etc., given performance of the AC drivers I see no treason to push the frequency response further, they really sound very clean and punchy.
Yes the SDX10 is sealed, may have a play with the FF85WK sometime, I have used Fostex drivers before but found them a bit "shouty", I feel the smaller driver would avoid this.
Last edited:
it even works OB, don't let anyone tell you otherwise It’s overlooking the combination of variability of “room gain “ / boundary reinforcement and speaker placement that can lead to unexpected or unsatisfactory results. The room’s acoustics are ultimately in control of what you hear at any given spot, and few of us have the luxury of spaces as benign as Dave’s, which is absent elaborate engineered treatment.
I did some measuring and tweaking of my system yesterday using REW and the miniSHARC DSP.
Having tweaked the crossover to eliminate a small dropoff I noted a band between about 80 and 600hz where it was some 3-6db down, I assunmed this to be baffle step so added some compensation for this. Above 1k or so it actually looks much like the MA supplied FR graph, I may add some compensation to flatten this out a bit later.
What really surprised me in the attached graph is the bottom end, clearly room gain plying its part as my Linkwitz transform targets an F3 of 30hz. The FR graph has variable smoothing applied, REW call this psychoacoustic smoothing.
Comments much appreciated.
Having tweaked the crossover to eliminate a small dropoff I noted a band between about 80 and 600hz where it was some 3-6db down, I assunmed this to be baffle step so added some compensation for this. Above 1k or so it actually looks much like the MA supplied FR graph, I may add some compensation to flatten this out a bit later.
What really surprised me in the attached graph is the bottom end, clearly room gain plying its part as my Linkwitz transform targets an F3 of 30hz. The FR graph has variable smoothing applied, REW call this psychoacoustic smoothing.
Comments much appreciated.
Attachments
ChrisMmm,
Since you have the miniS to do the XO, it's easy to try different combination.
I have found, especially with a FAST project like yours, that the Harsch XO is quite the performer.
Here's how to apply it:
1. Set the low pass filter for the woofer as a 4th order Butterworth at central frequency, fc for the XO centerpoint.
2. Set the high pass filter for the tweeter as a 2nd order Bessel at fc.
3. Set the delay of the tweeter equal to 1/2 of the period of one cycle at fc.
4. Use all positive phase on woofer and tweeter.
Using this, I would probably move the XO higher, like 200-ish, just so the Alpair gets a little more freedom.
Once set, listen to that bass tighten up. And everything else is much cleaner.
Also, instead of boosting the low end to compensate the baffle step, try lowering the 800kH to the top, applying a BSC, again with the miniS. It's always better to take down peaks than trying to boost nulls.
Since you have the miniS to do the XO, it's easy to try different combination.
I have found, especially with a FAST project like yours, that the Harsch XO is quite the performer.
Here's how to apply it:
1. Set the low pass filter for the woofer as a 4th order Butterworth at central frequency, fc for the XO centerpoint.
2. Set the high pass filter for the tweeter as a 2nd order Bessel at fc.
3. Set the delay of the tweeter equal to 1/2 of the period of one cycle at fc.
4. Use all positive phase on woofer and tweeter.
Using this, I would probably move the XO higher, like 200-ish, just so the Alpair gets a little more freedom.
Once set, listen to that bass tighten up. And everything else is much cleaner.
Also, instead of boosting the low end to compensate the baffle step, try lowering the 800kH to the top, applying a BSC, again with the miniS. It's always better to take down peaks than trying to boost nulls.
Last edited:
I only have a rough understanding of the difference between on-axis and power responses. I do my measurements at approximate listening distance (say 4m away) and height with just one speaker (woofer/full range combo) in order that I get both direct, on-axis, and reflected/room sound. Does this mean I am effectively averaging on-axis and power responses?
Thanks perceval, I will read up on the Harsch XO and give that a go. Re the delay, using 200hz thats a wavelength of 1.7m divide by speed of sound (343m/sec) is 5ms so 1/2 is 2.5ms?
Thanks, then I believe I am really measuring the power response. On-axis response is typically measured at 1 metre I believe, maybe I should be measuring both.
Or, since your setup is two different cabinet, you could do it physically by moving the Alpair a few cm back.
Here's the thread that introduced me to the Harsch XO.
S. Harsch XO
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.