Hi all, I'm helping a friend design and build a what we hope will be a good quality and versitile PA reinforcement speaker.
For larger venues, these would be used with a subwoofer setup, but smaller venues they should be able to cover on thier own. High SPL and high quality are important factors, it would also be desireable to be able to expand the system to cover large outdoor crowds without moving to full line array setup, if this would be possible.
We have selected some drivers we think will be suitable for this design. They are:
Woofer - RCF MB15N351
Compression driver - Beyma CD10Nd
Horn - Beyma TD194
The horn + CD for thier high sensitivity and ability to crossover quite low, and the woofer for its relatively high sensitivity and extended response. The lightweight neo motors are an advantage.
A logical crossover point is going to be around 1.5kHz. There are a few issues I can see. Obviously, the CD is quite lot more efficient, so an L-pad will need to be used. In a high power setup I am concerned this could waste a lot of power? However, a lower efficiency CD is going to limit max SPL, which is worse.
The woofer has a bit of breakup around 2kHz, it won't be massively attenuated by a 1.5kHz crossover (probably 12dB/octave). Would it be applicable to run a 1.2kHz (or whatever frequency is determined best) crossover on the woofer and 1.5kHz on the CD, thereby the rising output caused by the woofer breakup is smoothed out by virtue of it having a crossover beginning earlier. Or does doing this ruin the phase response of a crossover?
On phase, naturally the CD sound eminates from a point further back than the woofer, ie, the throat of the horn. What can be done to help preserve a coherent wavefront? A whole wavelength at 1.5kHz is ~23CM, the whole horn path is 14.5CM, so it's possible the spacing between acoustic centres of the woofer and CD is around 1/2 wavelength at this frequency. This means the drivers should hence be wired in phase, not out of phase as is typical in 12dB filters? The exact crossover point should be adjusted for phase coherence?
The TD194 horn is chosen for its 90x40 degree coverage. With relatively limited vertical coverage it is hoped that multiple speakers could be used to get more crowd coverage without treble cancellation issues. How will such a horn interact with the 15" woofer though in terms of dispersion?
The woofer will logically be installed in a vented enclosure. Around 45L tuned at 70Hz looks nice. Not much low bass, but suitable for crossing to subs and enough for smaller venues I'd think:
With regards to the cabinet, how important is shape? Many commercial designs are trapezoid, is this very important for reducing resonance? I guess not so much stuffing material can be used as with Hi-Fi since the heat build up could be significant. Bracing too I guess will remain important.
Baffle step correction I assume is best taken care of before the amplifier, actively, to save power. An EQ or the active crossover should facilitate this. Unless someone knows of a neat way to integrate this into the speaker?
That turned out to be a bit of a long post! Any input is appreciated. This design must exceed the JBL MRX515 in quality and SPL as it will cost more, and if a full design can be developed they shall be built ASAP.
For larger venues, these would be used with a subwoofer setup, but smaller venues they should be able to cover on thier own. High SPL and high quality are important factors, it would also be desireable to be able to expand the system to cover large outdoor crowds without moving to full line array setup, if this would be possible.
We have selected some drivers we think will be suitable for this design. They are:
Woofer - RCF MB15N351
Compression driver - Beyma CD10Nd
Horn - Beyma TD194
The horn + CD for thier high sensitivity and ability to crossover quite low, and the woofer for its relatively high sensitivity and extended response. The lightweight neo motors are an advantage.
A logical crossover point is going to be around 1.5kHz. There are a few issues I can see. Obviously, the CD is quite lot more efficient, so an L-pad will need to be used. In a high power setup I am concerned this could waste a lot of power? However, a lower efficiency CD is going to limit max SPL, which is worse.
The woofer has a bit of breakup around 2kHz, it won't be massively attenuated by a 1.5kHz crossover (probably 12dB/octave). Would it be applicable to run a 1.2kHz (or whatever frequency is determined best) crossover on the woofer and 1.5kHz on the CD, thereby the rising output caused by the woofer breakup is smoothed out by virtue of it having a crossover beginning earlier. Or does doing this ruin the phase response of a crossover?
On phase, naturally the CD sound eminates from a point further back than the woofer, ie, the throat of the horn. What can be done to help preserve a coherent wavefront? A whole wavelength at 1.5kHz is ~23CM, the whole horn path is 14.5CM, so it's possible the spacing between acoustic centres of the woofer and CD is around 1/2 wavelength at this frequency. This means the drivers should hence be wired in phase, not out of phase as is typical in 12dB filters? The exact crossover point should be adjusted for phase coherence?
The TD194 horn is chosen for its 90x40 degree coverage. With relatively limited vertical coverage it is hoped that multiple speakers could be used to get more crowd coverage without treble cancellation issues. How will such a horn interact with the 15" woofer though in terms of dispersion?
The woofer will logically be installed in a vented enclosure. Around 45L tuned at 70Hz looks nice. Not much low bass, but suitable for crossing to subs and enough for smaller venues I'd think:
With regards to the cabinet, how important is shape? Many commercial designs are trapezoid, is this very important for reducing resonance? I guess not so much stuffing material can be used as with Hi-Fi since the heat build up could be significant. Bracing too I guess will remain important.
Baffle step correction I assume is best taken care of before the amplifier, actively, to save power. An EQ or the active crossover should facilitate this. Unless someone knows of a neat way to integrate this into the speaker?
That turned out to be a bit of a long post! Any input is appreciated. This design must exceed the JBL MRX515 in quality and SPL as it will cost more, and if a full design can be developed they shall be built ASAP.
Attachments
Last edited:
A lot of questions!
For a design like this I would probably opt for a 2" compression driver. It will allow you to cross lower and/or it has more dynamic headroom. This comes at the expense of some top end dispersion but the lower crossover point will give you better directivity in the all important midrange. And please do use a bigger horn. The bigger the horn, the lower the frequency to which horn-loading works and the lower the frequency to which directivity control is maintained. I would use the biggest horn that still fits on the baffle.
Padding the tweeter shouldn't be much of a problem.
The relative delay difference between drivers has to be dealt with in the crossover. This might take some effort to get right. Do you have experience with designing crossovers?
Cabinet shape is not very important. There is no need to be afraid of the inside of the box getting too warm due to acoustic damping. The sound power is small in comparison to the electrical power dissipated in the voice-coil.
If you have EQ, I think it's a good idea to do most corrections that way.
Keep in mind though, that it might be difficult to build something a lot better than a JBL MRX515. Those are pretty good speakers at a good price!
For a design like this I would probably opt for a 2" compression driver. It will allow you to cross lower and/or it has more dynamic headroom. This comes at the expense of some top end dispersion but the lower crossover point will give you better directivity in the all important midrange. And please do use a bigger horn. The bigger the horn, the lower the frequency to which horn-loading works and the lower the frequency to which directivity control is maintained. I would use the biggest horn that still fits on the baffle.
Padding the tweeter shouldn't be much of a problem.
The relative delay difference between drivers has to be dealt with in the crossover. This might take some effort to get right. Do you have experience with designing crossovers?
Cabinet shape is not very important. There is no need to be afraid of the inside of the box getting too warm due to acoustic damping. The sound power is small in comparison to the electrical power dissipated in the voice-coil.
If you have EQ, I think it's a good idea to do most corrections that way.
Keep in mind though, that it might be difficult to build something a lot better than a JBL MRX515. Those are pretty good speakers at a good price!
Last edited:
By what criteria did you choose to cross at this point? I ask as I think it would answer some of your other questions.
You want to match phase between the drivers, obviously, and it can be done. Would you be planning to measure? to simulate?
Thanks for your response!
To change to 2" CD and larger format horn without going over budget leaves only 1 real choice of components:
P-Audio BM-D750 Compression driver
PH-2380 Horn
These are presumably not such high quality components but I suppose this may be outweighed by being able to use a lower crossover point. High frequency extension also takes a hit.
I don't have experience with passive crossovers so would need some guidance on this. Am I thinking along the right lines with adjusting the crossover frequency until we can use the driver in reverse polarity to achieve a coherent wavefront?
I'm not concerned with the wadding absorbing acoustic energy and gaining heat, only that it may impede the cooling of the driver motors by insulating the enclosure. I suppose with it being vented though that this circulates cooler air.
I suspect you're right with the JBL speaker. Where there are savings to be made in building your own PA subwoofers, this may not be the case for full-range speakers.
To change to 2" CD and larger format horn without going over budget leaves only 1 real choice of components:
P-Audio BM-D750 Compression driver
PH-2380 Horn
These are presumably not such high quality components but I suppose this may be outweighed by being able to use a lower crossover point. High frequency extension also takes a hit.
I don't have experience with passive crossovers so would need some guidance on this. Am I thinking along the right lines with adjusting the crossover frequency until we can use the driver in reverse polarity to achieve a coherent wavefront?
I'm not concerned with the wadding absorbing acoustic energy and gaining heat, only that it may impede the cooling of the driver motors by insulating the enclosure. I suppose with it being vented though that this circulates cooler air.
I suspect you're right with the JBL speaker. Where there are savings to be made in building your own PA subwoofers, this may not be the case for full-range speakers.
I don't think that driver is going to help you. I think the other poster was referring to the $200+ 2" compression drivers that can cross over at 500Hz. This driver can only cross over at about 1 kHz.
Also, my limited understanding is that for maximum power output, you're going to have to go active, with 24 dB/octave filters. This of course significantly increases your cost.
Last edited:
Actually, I think a 1khz crossover might be about right for a woofer this size. Some 1" drivers can handle it reasonably well, but a 2" driver might be better suited. If you want to use a 2" down to 500 hz, you'd need a huge horn!
The P-audio's look well suited at first glance.
I would go with this:
http://profesional.beyma.com/ingles/pdf/CP750Nd.pdf
on this horn:
http://profesional.beyma.com/ingles/pdf/TD590.pdf
crossing over at 800 Hz.
1.5kHz was only chosen preliminary as a point that the CD+Horn could comfortably run down to.
I think simulation and calculation is more likely than measurements, though I may be able to take some measurements.
The horn for the P-audio is about the largest one could fit onto a design like this. Achieving a lower crossover point in this size format is probably not going to be possible. So, is 1kHz with this large 2" CD and horn a much better choice than 1.5kHz with a 1" CD and smaller horn? Will the 1" CD "bottom out" at its max 70W toward the lower end?
I think simulation and calculation is more likely than measurements, though I may be able to take some measurements.
The horn for the P-audio is about the largest one could fit onto a design like this. Achieving a lower crossover point in this size format is probably not going to be possible. So, is 1kHz with this large 2" CD and horn a much better choice than 1.5kHz with a 1" CD and smaller horn? Will the 1" CD "bottom out" at its max 70W toward the lower end?
Last edited:
The horn I picked is 250x 280 whereas the one you originally picked is 202 square. Is that a large difference? Both are far smaller than a 15" driver.
Sorry, you must have posted while I was typing. Those are nice components, but a stretch too far budget wise. Looking at around £340 on drivers per cab; that CD+Horn leaves 0 budget for a woofer unfortunately.
I'm not sure about bottoming out a 1" CD. The problem with the higher crossover point is the change in directivity: the 15" woofer beams a lot at 1.5k, while most horns beam a lot less. At the crossover you want to match the directivity of both drivers.
Sounds logical yes. The JBL system uses a 1.7kHz crossover interestingly. Not too sure why it has 70x70 radiation horn either, surely wide horizontal and narrow vertical is always more desireable in this sort of application?
With no off-axis graphs or polar plots for the RCF 15", how can we determine what a good crossover point should be? Would you say 1kHz is a good general frequency for a 15" to match the 90x40 horn?
With no off-axis graphs or polar plots for the RCF 15", how can we determine what a good crossover point should be? Would you say 1kHz is a good general frequency for a 15" to match the 90x40 horn?
They are cheaper here in America. Maybe someone can smuggle them into the UK for you?
1kHz is a good start, probably not any higher. If you don't want to go down with the horn, you could think about smaller woofers.
Many manufactures use symmetrical pattern horns to avoid "pattern flip", where the narrow pattern actually becomes wider at lower frequencies.
For a 90x40 to maintain vertical pattern control to 1000 Hz, the mouth needs to be about 12" high (a full wave length of 1K), which presents other design difficulties and compromises.
All front loaded cone speakers begin to beam at a frequency dependent on the piston diameter.
A 15” has a -6 dB 90 degree dispersion at about 1200 Hz.
Dispersion varies with frequency, a 15” speaker is about -6 dB at 60 degree at about 1.8K, a 12” at 2.2K, a 10” at 2.8K.
A 15” has a 180 degree -6 dB beam width at around 600 HZ, a 12” at about 800, and a 10” at about 1200 HZ.
A 12” speaker is around 180 degree at 900 Hz, 90 degree at 1200 Hz, 50 at 2400 Hz, and 30 degrees at 4000 Hz.
Thanks for all the input, it's been very useful! Currently, my friend is having trouble raising money for this project. He needs to sell a pair of 3015LF drivers to get some funds. I say it's a shame they aren't the standard 3015 model as those would probably work in this design!
- Status
- Not open for further replies.