There is not a lot of information in the internet, but I always thought that anyone using a computer and a DAC as a source, should implemente filtering using custom software in the PC.... There is no need for active filtering or aditional hardware IMHO.
I would like to avoid using too many DACs... I think DAC output stages are critical in sound quality (like a preamplifier).
Last edited:
Hi charly
Well you can do this passively although it is harder to work out the solution because the filter you make has to have a real loudspeaker and not a resistor as a load. With passive parts, it usually takes 3 parts to deal with each bump . If you include the secondary aspects of these parts in a computer model, the only down side to complexity is weight, size and cost (but not a sonic penalty) If you get one of these horn drivers and horns, there are folks here who can help.
I will expand on Jacks answer too. Unlike a dome tweeter, no horn driver has a flat frequency response. The reason is what they do requires a different response to the signal. If you examine a compression driver, you see a MUCH more powerful magnet gap than a dome tweeter and very low mass construction and facing about a 8:1 - 10:1 compression ratio.
A dome tweeter usually has a much lower B in the gap, less motor and even at times a damped goo dome, it’s response is mass controlled or acceleration while the Compression driver is ideally a Velocity response.
Bottom line, the acoustic power response or efficiency is limited at the upper end to where a good one inch driver has a knee around 2Khz (TAD 2001 is more like 4KHz) while the larger the driver the lower that point.
That roll off is a first order slope (itself) and the olden days fix was to use a horn who’s radiation got narrower as the frequency climbed.
If the narrowing matched the power roll off, the directly on axis response was flat. The fix required for any kind of wider seating area is constant directivity and if the horn were really CD, then the power roll off would be about 6dB same as the driver. In the real driver, one also has several other low pass corners which come into play so the real response is what you have to fix.
The thing is this, what governs how well it does at the low end, often contradicts what is needed at the top end. What one can build in the real world has a low limit and high limit even using the most exotic materials and has a power or efficiency response curve. When you put a driver on a horn, you change (improve) some of that efficiency curve but you also are confining where it’s going and like a directed light bulb, it seems brighter or louder and it is (within that beam).
Whatever driver you use, whatever horn you use that modifies that as above by loading and or directivity, you end up with the “raw” frequency response. Until it is near flat, there is very little one can judge about the sound.
A proper horn crossover involves both the crossover and compensation for the not flat frequency response AND that same compensation corrects the amplitude and phase automatically.
Like I said before, horns are a lot harder to deal with BUT if you get them right, they really work like nothing else. I love horns I guess you can tell haha.
If you have head phones on your computer, pop them on and watch this video. This was the first public demo of a 4 way loudspeaker that has three of those BMS coax drivers coupled together in it.
https://www.youtube.com/watch?v=pk54IFD4znw
These drivers would give you “headroom” and cover a much wider band and do it more effortlessly than any single driver you can find.
Put them on a good sized CD horn and flatten them out, don’t be afraid.
Best,
Tom
Well you can do this passively although it is harder to work out the solution because the filter you make has to have a real loudspeaker and not a resistor as a load. With passive parts, it usually takes 3 parts to deal with each bump . If you include the secondary aspects of these parts in a computer model, the only down side to complexity is weight, size and cost (but not a sonic penalty) If you get one of these horn drivers and horns, there are folks here who can help.
I will expand on Jacks answer too. Unlike a dome tweeter, no horn driver has a flat frequency response. The reason is what they do requires a different response to the signal. If you examine a compression driver, you see a MUCH more powerful magnet gap than a dome tweeter and very low mass construction and facing about a 8:1 - 10:1 compression ratio.
A dome tweeter usually has a much lower B in the gap, less motor and even at times a damped goo dome, it’s response is mass controlled or acceleration while the Compression driver is ideally a Velocity response.
Bottom line, the acoustic power response or efficiency is limited at the upper end to where a good one inch driver has a knee around 2Khz (TAD 2001 is more like 4KHz) while the larger the driver the lower that point.
That roll off is a first order slope (itself) and the olden days fix was to use a horn who’s radiation got narrower as the frequency climbed.
If the narrowing matched the power roll off, the directly on axis response was flat. The fix required for any kind of wider seating area is constant directivity and if the horn were really CD, then the power roll off would be about 6dB same as the driver. In the real driver, one also has several other low pass corners which come into play so the real response is what you have to fix.
The thing is this, what governs how well it does at the low end, often contradicts what is needed at the top end. What one can build in the real world has a low limit and high limit even using the most exotic materials and has a power or efficiency response curve. When you put a driver on a horn, you change (improve) some of that efficiency curve but you also are confining where it’s going and like a directed light bulb, it seems brighter or louder and it is (within that beam).
Whatever driver you use, whatever horn you use that modifies that as above by loading and or directivity, you end up with the “raw” frequency response. Until it is near flat, there is very little one can judge about the sound.
A proper horn crossover involves both the crossover and compensation for the not flat frequency response AND that same compensation corrects the amplitude and phase automatically.
Like I said before, horns are a lot harder to deal with BUT if you get them right, they really work like nothing else. I love horns I guess you can tell haha.
If you have head phones on your computer, pop them on and watch this video. This was the first public demo of a 4 way loudspeaker that has three of those BMS coax drivers coupled together in it.
https://www.youtube.com/watch?v=pk54IFD4znw
These drivers would give you “headroom” and cover a much wider band and do it more effortlessly than any single driver you can find.
Put them on a good sized CD horn and flatten them out, don’t be afraid.
Best,
Tom
Short answer, 300hz.
I know of people who have used them to 175hz.
They are power rated to 450hz, and distortion rated to 300hz.
I use them to 300hz all the time for live sound.
The main limiting factor will be the size of the horn, as the bigger the horn, the lower in frequency the driver will horn load.
I know of a company that crosses them over at 275 hz on a very small horn, and uses the ground behind the driver to baffle step load the driver. I am going to try this will the SEOS, and a brick wall behind the horn driver, to see how low I can go.
MaVo,
I wouldn't think that the rear chamber would be a limiting factor. More likely it will be the minimum distance from the diaphragm to the phase plug that will cause a limit to any real excursion you could have at that low a frequency and the damage that would occur to the diaphragm. I would think that even with a large enough horn with the correct loading you would have to severely limit the max spl output of the device before you would have damage. Perhaps someone else will chime in on this as I have never attempted to go this low with any compression driver even a large format type.
I wouldn't think that the rear chamber would be a limiting factor. More likely it will be the minimum distance from the diaphragm to the phase plug that will cause a limit to any real excursion you could have at that low a frequency and the damage that would occur to the diaphragm. I would think that even with a large enough horn with the correct loading you would have to severely limit the max spl output of the device before you would have damage. Perhaps someone else will chime in on this as I have never attempted to go this low with any compression driver even a large format type.
Kindhornman, my thought about using it lower was something like this: 118dB@1w that means in a home you would probably never use more than 1w. its rated for 150w above 400hz, which implies a large enough excursion capability to reach this input power. a quick test in hornresp showed me that 100x the watt input gave roundabout 10x the excursion. so, in my 1w usage case the max excursion ever used will be less than 10% of the maximally possible. now, going an octave lower about doubles excursion, so i would imagine we could go up to 3 octaves down from 400hz and still be only at 80% max excursion with 1w input, given a large enough horn. more realististically somewhere between 150 and 250hz would make the horn fit into a room better. (which is probably flawed thinking, thats why i asked initially).
Last edited:
MaVo,
One thing to remember is that often the excursion limit is only given for a small range of frequencies and not for the entire bandwidth. So I would have a good look at what the excursion limit is at the lower frequencies which could be only a portion of what you are thinking. I highly doubt that even at only 10 watts you will find that the excursion is not going to damage a compression driver diaphragm at 150hz or anything near that low. The same goes for the efficiency ratings that are given for most devices, they are often only for a very narrow band of frequency that is used as a marketing tool.
One thing to remember is that often the excursion limit is only given for a small range of frequencies and not for the entire bandwidth. So I would have a good look at what the excursion limit is at the lower frequencies which could be only a portion of what you are thinking. I highly doubt that even at only 10 watts you will find that the excursion is not going to damage a compression driver diaphragm at 150hz or anything near that low. The same goes for the efficiency ratings that are given for most devices, they are often only for a very narrow band of frequency that is used as a marketing tool.
Would you clarify this sentence for me please. Too many negatives for me to figure out what you are saying.
So be sure, the 175hz was for a home audio application, with a very small tube amp. (There were meeting the BMS with a sub or low horn that was limited to 175 on its top end.) I heard that it worked, but of course did not hear it.
I have a very large horn, 48"x33", and can mount a 4592 on it to see how low it will go in frequency. What is the theoretical cut off of this size horn?
What kind of test could I do on SPL without damaging the diaphragm?
Charly,
Earl Geddes has some good information on his site about those two components.
Jack,
What I am saying is that even at only a 10 watt input I would expect that you would damage a compression driver if it was to produce a 150hz tone. I could be wrong about that but I don't think so. Even with a horn that would load a compression driver that low the excursion would be more than the device would handle that low from what I remember. Perhaps at 1 watt you would get away with that.
Earl Geddes has some good information on his site about those two components.
Jack,
What I am saying is that even at only a 10 watt input I would expect that you would damage a compression driver if it was to produce a 150hz tone. I could be wrong about that but I don't think so. Even with a horn that would load a compression driver that low the excursion would be more than the device would handle that low from what I remember. Perhaps at 1 watt you would get away with that.
Jack, Kind, thanks.
bms 4590 are traveling fron Utah to Nevada now. should be in Chile next week.
I send this collegue of mine in a plane with 44pounds in compresion drivers. this guy deserves at least a good bottle of wine.
i realized that i didnt bought the standard passive crossover. it seems to be a 18db/oct second order, at 6.5khz. guess i have to get one so i can use my 1.5watt amplifier in mid and highs...
now plan is
to get/build passive crossover for 4590 coaxial driver
use minidsp to go active inegrating coaxial with a woofer
also i plan to use minidsp to correct issues with the horn/CD combo.
waiting now. I will try to uppload measurments when i get there.
someone has a suggestion about bms4590 and its passive crossover?
or about this general approach?
best
bms 4590 are traveling fron Utah to Nevada now. should be in Chile next week.
I send this collegue of mine in a plane with 44pounds in compresion drivers. this guy deserves at least a good bottle of wine.
i realized that i didnt bought the standard passive crossover. it seems to be a 18db/oct second order, at 6.5khz. guess i have to get one so i can use my 1.5watt amplifier in mid and highs...
now plan is
to get/build passive crossover for 4590 coaxial driver
use minidsp to go active inegrating coaxial with a woofer
also i plan to use minidsp to correct issues with the horn/CD combo.
waiting now. I will try to uppload measurments when i get there.
someone has a suggestion about bms4590 and its passive crossover?
or about this general approach?
best
Since you have a DSP, try crossing the horn/compression driver higher, say, a 1/3 octave at a time, listening to music you are familiar with. If the impression of grain-and-grit suddenly disappears, that's your new crossover point.
"Pushing" a compression driver lower than it wants to go will result in very harsh sound, no matter what the measurements say. Direct-radiator tweeters and midranges will tolerate this abuse better than compression drivers; the lower-frequency limit of acceptable sound has a softer edge, and you can push them a few Hz lower that the manufacturer says you can. Compression drivers are the opposite; you get less (good-sounding) bandwidth than you think you might have. So be willing to cross them higher than you originally planned.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.