Interestingly though, the 10' has half the rolloff of the mouth measurement.
So what does "flush at mouth of horn" mean? To me it sounded like flush mounted and distance measured from the baffle.
Last edited:
Flush at mouth of horn is the microphone position. All the distance measurements are from the forward most surface of the hour mouth.
The horn/driver is on a stand with the lip of the horn in free space with the exception of the few inches at the bottom where the edge of the horn is sitting on the edge of the platform.
The horn/driver is on a stand with the lip of the horn in free space with the exception of the few inches at the bottom where the edge of the horn is sitting on the edge of the platform.
Last edited:
Horns need EQ, they're not inherently flat like many piston transducers. Limiting and controlling room interaction, massive efficiency and headroom, low distortion, low diaphragm displacement, all good things.
A smooth horn response is one that doesn't need high-Q notches, lump or tilt equalization is a lot easier and more benign.
An externally hosted image should be here but it was not working when we last tested it.
It seems that all curved wall horns have this kind of general trend with the varying distance.
I am wondering if that glitch at 8KHz will chance if you put foam around the outer lip.
I think I am going to dig up some of my old data and have a look at that scenario again.
I have experiance with both the B&C 250 and the JBL 2420 and what gets my attention is the difference in your measurements and the Azurahorn 550 horn and the JBL. When I measured them both I found the B&C was actually had a flatter response in a 350 hz round wooden tractrix horn then the JBL - but look at this graph of the Azurahorn with the JBL - something ain't right - your measurements look a lot like mine but the tractrix goes a little lower with it's slower flare but pretty close - The Azurahorn graph for the JBL is shown here:
http://www.azurahorn.com/JBL2420_on_550.pdf
I don't know where I got this is a B&C 250 - is it?
http://www.azurahorn.com/JBL2420_on_550.pdf
I don't know where I got this is a B&C 250 - is it?
Last edited:
I was wondering the same thing. That zigzag between 8 and 10khz looks particularly nasty to me, and would be clearly audible as harshness in the treble. (Be interesting to see what the CSD looks like, that would be more telling)
I've gone to the trouble of EQ'ing much smaller (1dB) resonances than that in that frequency range before because they were audible and unpleasant on some music.
My hearing still goes to 17.5Khz so the next major resonance around 15Khz would bother me too.
Some of the frequency response wiggles are native to the driver, some to the horn.
Below two horns with the B&C DE 250 driver are compared, though without using the same actual driver and test set up response will always differ.
Attachments
That one is only a little more than 1dB (and at 1/24 octave smoothing, too).
That one's only about 2.5dB and very high Q. These are "major resonances" to you? You're not easy to please! It is the first part of a truly major breakup in the top octave, though. I would certainly want to kill that.
In any case, these are measurements taken flush with the horn mouth at no particular angles. Wasn't everyone just saying how you should never measure a horn like that, or was that another thread? Anyway, not what we're usually looking at with horn measurements on this site/thread.
Some resonances interact with the amplifier depending on how the XO is designed, so there is no telling whether the solution should be at the driver or at the amplifier. I think what DBMandrake is hearing is probably more an amplifier issue, but solved by decoupling the interaction instead of the amplifier design.
I would still recommend measuring inside the horn as mentioned in the HOM measurement thread.
I would still recommend measuring inside the horn as mentioned in the HOM measurement thread.
Last edited:
Actually, I made the measurements and posted them for discussion of the post(s) earlier in the thread on differing frequency response with distance from the horn.
The series of measurements is taken at different distances from the mouth at 90°. I would consider the first 4 measurements 0", 6",1' and 2' to be mostly useless for anything but discussion of change in FR going from near field to far field.
Gary