I'm looking into low mid spherical horns using cone drivers, and I wonder;
What are the pros and cons of a high vs. low compression ratio?
Note that I will use the horn under into the bass under the horn cutoff. I only want higher efficiency in the low mid to mate better with the mid-tweeter horn.
Cheers 🙂
What are the pros and cons of a high vs. low compression ratio?
Note that I will use the horn under into the bass under the horn cutoff. I only want higher efficiency in the low mid to mate better with the mid-tweeter horn.
Cheers 🙂
low compression ratios sound smooth, easy on the ears.
Efficiency becomes a moot point.
Don't waste your time, check into a 2" community m200a (4:1 compression driver).
I'd be using one now, but I'd prefer to run a 2 way system than a 3-way (less phase stuff). You could get the m200 but it has a resonance that was fixed in the m200a (parallel in the throat)
Using the 400hz stereolab horn (1' deep and across), it can cross it at 600hz and 4khz (LR 24db active crossover).
But you want to run the driver under the horn and boost it, you may need a 3" or a 4" that arn't cheap.
Norman
Efficiency becomes a moot point.
Don't waste your time, check into a 2" community m200a (4:1 compression driver).
I'd be using one now, but I'd prefer to run a 2 way system than a 3-way (less phase stuff). You could get the m200 but it has a resonance that was fixed in the m200a (parallel in the throat)
Using the 400hz stereolab horn (1' deep and across), it can cross it at 600hz and 4khz (LR 24db active crossover).
But you want to run the driver under the horn and boost it, you may need a 3" or a 4" that arn't cheap.
Norman
You mean a 3" or 4" throat horn? I was thinking of using a cone driver, around 6-8 inches and maybe a 2:1 compression ratio. It will be used up to 800 hz.
ahhhh, ok.
remember mass corner is real. The compression of the throat and maybe a phase plug help a driver get a little higher. The high end mass corner is (Fs / Qes) * 2 .
I could be wrong, but you may not need to run a throat smaller than the cone.
Norman
remember mass corner is real. The compression of the throat and maybe a phase plug help a driver get a little higher. The high end mass corner is (Fs / Qes) * 2 .
I could be wrong, but you may not need to run a throat smaller than the cone.
Norman
I use a ev dh7 compression driver (10:1).
I can only quote a friend (djk) who is into this stuff.
He buys stuff just so he can listen to it.
"Dave tells me he could live with an M200 on one side and a TAD4001 on the other and is going to keep the M200s as spares.It should be pointed out that he is using them from 400hz~4Khz with a TAD2001 as a tweeter...... I have used pairs of the Eminence up to 500hz crossed to the CLS M200 with a LR24, they work well....... I think the 2440 sounds better below 800hz, the 2441 better above 5Khz. For a 400hz~4Khz dedicated midrange I think the 2" exit Community Light and Sound M200 in many aspects sounds better...... The M200A is not available, and has not been available for some time. The M200 is useable to 4Khz, but must have a real crossover to roll it off above there (peak around 6.3Khz). How does it sound? Due to the lower compression ratio it sounds more 'relaxed' than high compression ratio drivers (BMS, JBL, etc).... Many of my friends went 2" years ago on the Klipschorn. Most of these would have been JBL. At any given volume level there is an order of magnitude reduction in distortion. Going from a JBL 2" to a low compression 2" from CLS a big improvement, and the CLS M200 is a lot less money too.... I wouldn't do it with a JBL driver.A Community Light and Sound M200 sounds great at 400hz on a 2397.A 375/2440 rocks and scrapes and a 376/2441 makes oil-can sounds.The only horn loading on a 2397 is provided by the 2328 throat adaptor.... "why does Avantgarde in their Trio's use the midrange driver without phase plug?" Why, does it need one? It goes to about 4Khz and then drops like a rock. The very low compression ratio sounds less 'squeezed' than the typical 10:1 ratio unit. Avantgarde uses a different impedance coil so it requires no attenuation in the network.... The best way to remove the 6Khz peak is to use the M200A, which rolls off smoothly needing no parts. Having said that, I also have some of the older M200s with the peak. With a 24dB electronic crossover one may ignore the peak.... Lastly, consider the stock version of the driver used in the Avantgard Trio, the Community M200A. The difference is in the voice coil impedance. The M200 has a peak at 6Khz, and would require a crossover to notch it out, the M200A has a smooth roll-off at 4Khz without a crossover (as does the 290, but at 6.3Khz). The M200A has a very different sound due to the low compression ratio (about 2:1 vs the more typical 10:1), much more 'relaxed' rather than 'hard or compressed'...... The JBL2445 on the 2380A has a higher peak/trough from 500hz~4Khz (about 8dB) than the Community does (about 6.5dB), and the Community has much less distortion too (JBL over 10% at 10W at 3.5Khz vs the Community hitting about the same 10%, but with 75W in! Even adjusted for dB down from actual output (the JBL has a peak where the Community has a dip) the Community has less distortion.... The JBL is only 105dB at 500hz, the Community is 110dB. In the 1Khz region the JBL is 108dB, the Community is 112dB."
Norman
I can only quote a friend (djk) who is into this stuff.
He buys stuff just so he can listen to it.
"Dave tells me he could live with an M200 on one side and a TAD4001 on the other and is going to keep the M200s as spares.It should be pointed out that he is using them from 400hz~4Khz with a TAD2001 as a tweeter...... I have used pairs of the Eminence up to 500hz crossed to the CLS M200 with a LR24, they work well....... I think the 2440 sounds better below 800hz, the 2441 better above 5Khz. For a 400hz~4Khz dedicated midrange I think the 2" exit Community Light and Sound M200 in many aspects sounds better...... The M200A is not available, and has not been available for some time. The M200 is useable to 4Khz, but must have a real crossover to roll it off above there (peak around 6.3Khz). How does it sound? Due to the lower compression ratio it sounds more 'relaxed' than high compression ratio drivers (BMS, JBL, etc).... Many of my friends went 2" years ago on the Klipschorn. Most of these would have been JBL. At any given volume level there is an order of magnitude reduction in distortion. Going from a JBL 2" to a low compression 2" from CLS a big improvement, and the CLS M200 is a lot less money too.... I wouldn't do it with a JBL driver.A Community Light and Sound M200 sounds great at 400hz on a 2397.A 375/2440 rocks and scrapes and a 376/2441 makes oil-can sounds.The only horn loading on a 2397 is provided by the 2328 throat adaptor.... "why does Avantgarde in their Trio's use the midrange driver without phase plug?" Why, does it need one? It goes to about 4Khz and then drops like a rock. The very low compression ratio sounds less 'squeezed' than the typical 10:1 ratio unit. Avantgarde uses a different impedance coil so it requires no attenuation in the network.... The best way to remove the 6Khz peak is to use the M200A, which rolls off smoothly needing no parts. Having said that, I also have some of the older M200s with the peak. With a 24dB electronic crossover one may ignore the peak.... Lastly, consider the stock version of the driver used in the Avantgard Trio, the Community M200A. The difference is in the voice coil impedance. The M200 has a peak at 6Khz, and would require a crossover to notch it out, the M200A has a smooth roll-off at 4Khz without a crossover (as does the 290, but at 6.3Khz). The M200A has a very different sound due to the low compression ratio (about 2:1 vs the more typical 10:1), much more 'relaxed' rather than 'hard or compressed'...... The JBL2445 on the 2380A has a higher peak/trough from 500hz~4Khz (about 8dB) than the Community does (about 6.5dB), and the Community has much less distortion too (JBL over 10% at 10W at 3.5Khz vs the Community hitting about the same 10%, but with 75W in! Even adjusted for dB down from actual output (the JBL has a peak where the Community has a dip) the Community has less distortion.... The JBL is only 105dB at 500hz, the Community is 110dB. In the 1Khz region the JBL is 108dB, the Community is 112dB."
Norman
A high compression ratio is neither bad nor good, it is simply one variable in the horn throat distortion equation. Throat length and flare rate are the other variables. In other words, if you use a high compression ratio behind a rapidly expanding waveguide, you may find you have less distortion than a low compression ratio feeding something like an exponential horn. This was a major motivator behind JBL's Optimized Aperture line of waveguides.
Also take Tom Danley's unity/synergy horns for example. Very high compression ratio in the cone midranges, but they feed a conical throat that flares rapidly. Many consider them a benchmark for openness and clarity.
If you really want to quantify throat distortion, I highly recommend the Hornresp simulator. After you run a calculation and you're in the SPL window, click on tools / sample, and it will tell you the 2nd harmonic distortion contribution of your compression ratio/throat system. No need to guess or generalize.
Also take Tom Danley's unity/synergy horns for example. Very high compression ratio in the cone midranges, but they feed a conical throat that flares rapidly. Many consider them a benchmark for openness and clarity.
If you really want to quantify throat distortion, I highly recommend the Hornresp simulator. After you run a calculation and you're in the SPL window, click on tools / sample, and it will tell you the 2nd harmonic distortion contribution of your compression ratio/throat system. No need to guess or generalize.
Attachments
A good concise guide, Bjorno.
Another thing I forgot to mention worth considering: total distortion rises with the bandwidth you demand of any driver. Everything's a trade-off.
Another thing I forgot to mention worth considering: total distortion rises with the bandwidth you demand of any driver. Everything's a trade-off.
So if I understand it correctly, the horn wont load above the high end mass corner of a driver? The driver I'm looking at has a mass corner of 188 given the equation above.
You can't buy a Community M200A...they only exist for Avant Garde, and CLS even then says they don't really exist.
At the low power levels used in home audio, I've never seen any real issues from a high compression ratio. I built some horns last year which seemed to be suffering from a compression ratio which was too high, but in hindsight I believe the real problem was simply harmonic distortion.
Here's a pic from one of my recent horns. Note that the PVC is basically pressed right up against the cone.
I'm a big fan of offset horns, and you can see that this is another one. I generally don't bother with coupling chambers.
Is there any special qualities in a driver I should look for, that works well in an horn application like this?
I thinking of using the horn below, which can be shortened to a max 4 inch throat. It has to work well from 50 hz up to 800 hz where it will be crossed to B&C DE250 compression driver (thus it will also be used under horn cf) If I use an 8 inch woofer it will then have a compression ratio of 2:1
I thinking of using the horn below, which can be shortened to a max 4 inch throat. It has to work well from 50 hz up to 800 hz where it will be crossed to B&C DE250 compression driver (thus it will also be used under horn cf) If I use an 8 inch woofer it will then have a compression ratio of 2:1
Attachments
Last edited:
Mathematically higher compression ratios only have advantages, but practicaly speaking there are limitations. Since a comprehensive test of distortion in compression drivers was unable to find any relationship between perception and THD I have to doubt any non-linear effects that would be a concern. At high CRs there will be high velocities and viscose effects. This are the limitng issues.
Thanks for the reply! What about high freq extension? Is a 2:1 CR going to work up to 800 hz? What about high end mass corner? (not knowing exacly what that is...)
Higher CR will always extend the usable bandwidth because it raises the impedance seen by the diapraghm. But higher CR need phase plugs and these are hard to get right. Basically the higher the CR the more precise the phase plug needs to be and the more specific the shape of the diaphragm. With a fixed diaphragm shpe and little possibility of doing a complex phase plug one is forced to use lower CRs and hence limited bandwidth.
You call that a 50hz horn?
lol
An externally hosted image should be here but it was not working when we last tested it.
THIS is a 50hz horn