Very good questions. Let me try to address them.
You can't go by Radian's posted curves for several reasons. Those were taken with different cabinets, different port tunings, different HF elements and different crossovers (the old stock 322s?).
We don't get our crossovers from Radian, they are hand made by Frank Fazallari of Coherent Speakers in Canada. He also makes high end crossovers for Radian in some applications.
He has been working with the Radian concentrics for some years now in his own designs, which are sold regionally. You can see one of his designs featured here, just scroll down the page:
The zobel network is his idea, which not only affects the woofer's impedance, it largely affects the high end, taming down some of the "harshness" usually associated with running horn tweeters at high volumes. If he has time, he may swing by here and give some insight. I spoke to him yesterday and he said he would try to post here but was busy building a lot of stuff this weekend.
The reason I have not posted a graph, is that being the manufacturer it will be tend to perceived as "the" graph. And we have measured indoors, outdoors, on axis, off axis and with several different mics.
However, we do not have an anechoic chamber and I am really pushing to get third party measurements done. I prefer having a third party do it as even if the manufacturer's graphs look really good, I feel it will be taken with a grain of salt until independently measured and rightfully so.
We are sending some CX-10s off to be reviewed soon and meanwhile I am searching for someone to do some testing in a real nice chamber. Finding one that won't break the bank seems to be the challenge.
I should point out that few of the passive, horn loaded loudspeakers sold in the ID market have published such graphs that you are requesting. I would love to see some published but again, having a third party do it would also give frame of reference with other loudspeakers tested under the same conditions.
So if you will be patient with us, I think we will be able to accommodate in the coming months.
Before leaving the facility we test each loudspeaker in pairs or in threes to ensure consistency. We take close mic sweeps and make sure there are no inconsistencies with crossover performance, we double check for proper phase. We then listen with program material. Each speaker usually gets played for at least an hour to break in. We find it imperative to listen to each one critically.
I will see if we can get an impedance plot soon as well. The first crossovers had a bit of an impedance dip above 15 khz and Frank was able to remedy that without making any audible changes. The reason for that was that we did not want it to "confuse" any digital amps with protection circuits, which could theoretically happen. So far, we have had minimal issues, but having the minimum impedance swing higher gives some peace of mind. I know there are a lot of four ohm speakers out there, but I feel better having the added stability as a precaution.
Originally Posted by Chu Gai
Jeremy, when I go over to Radian's website and look at some of the drivers that I believe you're using, in addition to providing some basic information, they also provide the frequency response with the driver in a vented box using an appropriate stock crossover (1500 Hz crossover point, 12dB/octave slope). In each of the drivers, there's a broad dip in the FR about an octave wide up to something like 1.4 kHz. Going out on a limb, I'm going to say that's possibly intentional given that these drivers were originally intended for large venue reproduction and that given that as you go up in frequency, there's more attenuation so in the end it kind of works out.
Now, in a home environment, that would not be desirable or acceptable. Earlier in this thread, you made made mention that you were playing around with capacitors to address some FR dips. I take it this is due to to the issues given above, yes? Also, later in the thread you stated that your crossovers incorporate the use of a Zobel (complex conjugate) presumably to make the speaker essentially resistive over its response.
Given all that, can you provide graphs of the FR on axis as well as off axis in something like 15 degree increments up to say 60 degrees? Also, do you have impedence plots to share? Lastly, I think, apart from fit and finish, what sort of tests do you perform to ensure quality and reproducibility?
Thanks in advance.