I need to repair two home-built PA speakers. The original speakers used 8” woofers and a dome tweeter (which I recognize is not suitable for PA use). The surrounds of the original 8” woofers have deteriorated. The boxes appear to be well-built 0.75” MDF. I’m thinking of replacing the 8” drivers with Eminence Beta 8 CXs, the Eminence ASD1001S for the HF driver, and the recommended Eminence crossover.
But I note that the same crossover is recommended for three different HF drivers, and this suggests that some compromises are made (but are these significant in my application?). Can anyone recommend a different combination of HF driver and crossover? I might save some money if I build the crossover myself, but I need a design.
I plan to seal the box (1.2 ft3) . I calculate an fbox of about 80 Hz and a Q of about 0.33, and this speaker will not get much low-frequency energy anyway (these speakers have a dedicated amplifier and EQ, and the speaker is used in a side fill application).
Tom
But I note that the same crossover is recommended for three different HF drivers, and this suggests that some compromises are made (but are these significant in my application?). Can anyone recommend a different combination of HF driver and crossover? I might save some money if I build the crossover myself, but I need a design.
I plan to seal the box (1.2 ft3) . I calculate an fbox of about 80 Hz and a Q of about 0.33, and this speaker will not get much low-frequency energy anyway (these speakers have a dedicated amplifier and EQ, and the speaker is used in a side fill application).
Tom
I have a pair of 8" fullrange drivers(very similar to the Beta8a) paired with 3/4" CD supertweeters in small cabs in my living room, they look to be an OEM Eminence driver for another company and these were pulled from some old PA cabs where they were used as mids. I tried using several different compression drivers with them and found that I could not get acceptable results crossing the CDs low.. down near the lowest recommended CF... the mids just didn't sound right. My 8" drivers sound fairly good all by themselves they just needed some help on the highend, so the idea of crossing them higher(5khz in my case) occurred to me and that worked much better, and given that a smaller CD would then suffice and I just happend to have some kicking around.
This is a bit long winded and some of it won't directly apply to your case but what I really want to say is I had to do a fair bit of work with the passive crossover to get these to where they currently are in terms of sound quality, so I don't think you will have much success with an over the counter crossover. It could be a good starting point but you will likely end up changing most of the parts and potentially adding a bunch more, so it's hard to say how much value there is in it. But you should try using a higher crossover frequency... like 2.5-3khz, this kind of goes against common wisdom but these small CDs just sound harsh when crossed low.. especially if you try to get some SPL from them, so I suggest you give this a try. If you do the Eminence CD should work fine I would think.
This is a bit long winded and some of it won't directly apply to your case but what I really want to say is I had to do a fair bit of work with the passive crossover to get these to where they currently are in terms of sound quality, so I don't think you will have much success with an over the counter crossover. It could be a good starting point but you will likely end up changing most of the parts and potentially adding a bunch more, so it's hard to say how much value there is in it. But you should try using a higher crossover frequency... like 2.5-3khz, this kind of goes against common wisdom but these small CDs just sound harsh when crossed low.. especially if you try to get some SPL from them, so I suggest you give this a try. If you do the Eminence CD should work fine I would think.
Thanks for the reference information. I suspect that PA speaker designers prioritize robustness and opt for a higher crossover frequency to protect the tweeter. Designers for home audio can afford to put a higher priority on sound quality and can lower the crossover frequency if desired. In this application, I'd go for the higher crossover frequency (as you recommend) for the sake of robustness.
I note that the Eminence crossover is at 2.5khz (12dB/oct). I wonder if it requires reversing the phase of one driver to get them in phase at the crossover frequency. That could affect how this sound combines with the main speakers.
Tom
I note that the Eminence crossover is at 2.5khz (12dB/oct). I wonder if it requires reversing the phase of one driver to get them in phase at the crossover frequency. That could affect how this sound combines with the main speakers.
Tom
Tom,
As is typical for a second order crossover, the tweeter polarity is reversed on the Eminence crossover, as can be seen looking at the back of the PXB2-2k5CX circuit board- the input "-" , woofer "-" and tweeter "+" are all connected together.
If the drivers are connected as labeled on the crossover, the tweeter's polarity is reversed.
Considering you plan to EQ the monitors, the additional passive crossover components and complexity required to flatten the response more than this passive are not required.
Art
Thanks, Art.
It seems that the most practical and expedient solution is to get the Eminence recommended crossover and with the beta 8cx and the ASD1001S.
I'll post a follow-up.
Tom
It seems that the most practical and expedient solution is to get the Eminence recommended crossover and with the beta 8cx and the ASD1001S.
I'll post a follow-up.
Tom
Hi. Thought I share some experience with the Eminence coaxials. I have not tried the 8” but I have the 10” and the 12”. I also have the CX crossover. That crossover works quite well in many combos it is a good crossover damping the mid a bit more than you would expect giving a smile curve response which is easy on the ear. The inbuilt L-pad is about -3dB I think so for most compression drivers you would need some additional padding or eq otherwise the treble will be a bit bright. Consider this though: instead of the ASD1001 consider the PSD2002 or PSD2013. These have much lower distortion more power handling and flatter frequency response and higher sensitivity. All in all better performance in my experience. I think the ASD1001 works better in three way setups from 3,5kHz while the PSD2002 can probably cut as low as 2kHz. With the CX crossover you will have no problem but you might need additional padding, depending on the midrange performance of the 8” which I assume is much better than the 12”. Do you plan to use a sub?
Thanks!
There is definitely a sub (<80 Hz). Since these are essentially side fill, lots of low end from the mains can be heard in this area. I suspect that there will be some additional high-pass filtering (probably first-order) with a corner frequency somewhere in the hundreds of hertz.
Is there a recommended or ready-made crossover for the PSD2002 or the PSD2013?
There is definitely a sub (<80 Hz). Since these are essentially side fill, lots of low end from the mains can be heard in this area. I suspect that there will be some additional high-pass filtering (probably first-order) with a corner frequency somewhere in the hundreds of hertz.
Is there a recommended or ready-made crossover for the PSD2002 or the PSD2013?
Assuming you have eq and an active sub I think you might get better mids using a more traditional crossover like the PXB2:1k6 but again I haven’t really tried the 8” coax and this would be mostly guessing. The 1,6kHz crossover is what is recommended for the higher priced kappalite neodymium coaxials. I have this crossover as well but it is less smooth sounding require more l-pad and possibly more eq. You could also build you own by measuring.
A side note: you might get more responses talking about crossovers in the loudspeaker multi way section. PA section is probably more on the system design for complete rigs with amps eqs subs tops and what not. Also think about the application you plan to use this. Home or outside etc.
A side note: you might get more responses talking about crossovers in the loudspeaker multi way section. PA section is probably more on the system design for complete rigs with amps eqs subs tops and what not. Also think about the application you plan to use this. Home or outside etc.
Last edited:
I am certainly interested in crossover design, and I'd like to acquire the measurement tools to do it someday. I can do active and passive filter design, and I think that with the right tools (and with the consultation of the multi-way forum) I could design a decent crossover. I don't know if I can invest the time to learn on this project, though.
I'm looking into the PSD2002 and PSD2013.
Thanks!
I'm looking into the PSD2002 and PSD2013.
Thanks!
IF Eminence themselves suggest using that 8" speaker, that driver/horn and their crossover, that´s it .... they "should" know.
It’s not that simple. That CX crossover was build for the 10”. It just happens to also work nicely with the 12” and the 8”. But it’s not perfect. It takes away quiet some mid as the LF is actually much lower than 2,5kHz. For most a lowered mid sound pleasing but to others it may sound dull and dark. Further the HF padding in my experience is not padding enough making the treble slightly to dominant. It very much depend on the application to be honest.
Here’s the some chart and technical stuff. This is pretty old now. The new Kappalite coaxials and the PSD2013 with 1,6kHz xo gives more perspective to it all.
https://cdn.shopify.com/s/files/1/0270/8665/1462/files/Eminence_Coax_Use_v1.pdf?v=1613789488
Some more on the crossover here
https://eminence.com/blogs/blog/great-uses-for-coaxial-products
https://testing.eminence.com/great-uses-for-coaxial-products/
Here’s the some chart and technical stuff. This is pretty old now. The new Kappalite coaxials and the PSD2013 with 1,6kHz xo gives more perspective to it all.
https://cdn.shopify.com/s/files/1/0270/8665/1462/files/Eminence_Coax_Use_v1.pdf?v=1613789488
Some more on the crossover here
https://eminence.com/blogs/blog/great-uses-for-coaxial-products
https://testing.eminence.com/great-uses-for-coaxial-products/
If that 10” is anything like other 10” Eminence I’ve used, it requires a gentle roll off starting much lower than 2.5 kHz, increasing to 12dB/oct at around 2.5 kHz. This counters the linear breakup rise that starts below a kHZ and gives 4th order acoustic at 2.5k. The 12’s have a different profile which is actually harder to make flat with a passive X/O. Easy enough with a drivreack or DCX2496, though.
This is the RAW response of the 10CX in a 27L sealed box.
wg_ski is right about the cone breakup.
The highest peak is at 2.5kHz, followed by smaller ones.
Crossing passively at 1.5kHz is ideal but that would mean using costly comp drivers like the JBL 2425.
An alternative is to cross at 2kHz with an active 24dB/oct crossover.
Hopefully, it's enough to suppress the breakup peaks.
wg_ski is right about the cone breakup.
The highest peak is at 2.5kHz, followed by smaller ones.
Crossing passively at 1.5kHz is ideal but that would mean using costly comp drivers like the JBL 2425.
An alternative is to cross at 2kHz with an active 24dB/oct crossover.
Hopefully, it's enough to suppress the breakup peaks.
Last edited:
Just to check my understand of one breakup: Michael's graph shows a peak at 2.5kHz which is about 6dB above a reference level established around 1 kHz. If this peak is due to cone breakup, it could (and I think probably does) become higher relative to the reference at higher signal levels that that used to make this graph. That is, cone breakup is non-linear. Then is it also true that cone breakup should be reduced as much as possible by a higher-order crossover? Or with active crossovers, use notch filters (for the woofer) at the frequencies that excite cone breakup (the lower frequencies).
Thanks for the information.
Tom
Thanks for the information.
Tom
Here is the 10CX response with a 18dB crossover at 1.5kHz.
Black is RAW, Blue is with passive xover.
Interesting. The size of the 2.5kHz peak seems about the same for the black trace and for the blue trace. And for the blue trace, the speaker is driven with a lower amplitude at 2.5kHz. So it seems that the relative size of the breakup peak is not signal dependent; at least not significantly and at least in this case. That seems to recommend a higher-order crossover. Perhaps also a notch filter if an active crossover is used.
BTW, the time frame for the project in my original post has been extended, so I'll have time to consider the information in this post.
Thanks all
Tom
BTW, the time frame for the project in my original post has been extended, so I'll have time to consider the information in this post.
Thanks all
Tom