Sorry, but I don't see why you would say this, and I don't really think that it is true. If you have any objective reason for this belief (as opposed to your subjective impressions) I like to hear them.
I tend to agree.
Theoretically the orders are all uncoupled going higher (higher orders do modify lower orders, but not the other way around.) Hence, I agree that assuming that high THD means high higher order nonlinearity is a dangerous assumption that will not be found to be consistently true even if it seems to be a trend.
Not likely.
The exception is not the rule. Just because you can make up an exception cognitively does not make it true. Compression drivers do have higher 2nd HD because of air compression
But if you must, measure the THD then substract the 2nd and 3rd HD. I can do that with the APX500 software.
Thanks DT
No one said always, consistently or to assume. You are assigning words that were not said.
THD is a good indicator. If THD is high start looking for higher order distortions. If THD is low it is much more likely that the higher order distortions will not be problematic.
@gedlee you created your own distortion scale that never caught on.
If you want to be absolutely sure measure and listen carefully.
Thanks DT
For YOU. For me, no. I find waveguides inferior to a proper horn. With a proper horn I hear the sound, with a waveguide I hear the CD. Not that waveguides are horrible to my ears, but I clearly prefer horns as sounding more like natural acoustic sounds and less like audio gear. Ears and brains are different.
H7 is by far the most objectionable HD component. This is a difficult one to measure in a driver being so high order.
I've looked at this comparison carefully a few times. The DE250 reviewed and tested doesn't behave like the ones I've used before and I believe theirs was defective in some way. I find the DE250 to be very good above 2.5k in a decent horn that is deep enough to load it properly, but not quite as resolving as the higher end Faital HF108 and HF10AK. The issue is mainly the one piece polyimide diaphragm/surround material. It has some strange behavior in the upper mids which other drivers using the same diaphragm materials have as well, like the Celestion 1747. Polyester and PEEK diaphragms sound better to me and alu with mylar surround is slightly better up top ie Celestion 1430. This is a superb driver in terms of HF quality and resolution.
Another massive issue people don't understand is the horn plays a big part in how a driver will sound on the lower end cutoff by means of how it loads the driver. If the driver isn't loaded sufficiently, it greatly increases HD. Drivers like the DE250 need a decent sized horn to sound clean around cutoff. I don't specifically know the technical reason for this, but its been my own observation.
I am completely confused.
What is a proper horn?
Something that may be effecting what both of you may be talking about is the acoustic impedance effects of the waveguide or horn near the "cut off" end of the horn/waveguide frequency range.
Run a electrical impedance plot with the Compression Driver attached to the horn/waveguide . The CD/horn/waveguide assembly will have a impedance curve that is anything but flat or smooth.
Place a small value current sensing (0.1R to 0.01R) resistor in series with the Compression Driver voice coil and input the voltage across the current sensing resistor in your audio analyzer. You will be amazed by the distortion and current modulation caused by the impedance modulation of the variable acoustic loading of the horn/waveguide.
Thanks DT
I agree midrange is more integral than highs, however it is the characteristic of a driver that extends full range from a low crossover point, that is more appealing sonically in my opinion than superior mids in the example of the Axi compared with the 288. It is not the better highs that I value over better mids, but rather the characteristic of a single driver functioning as a point source from a low crossover point all the way to 20khz. And furthermore how detail is presented in this configuration. This is a more interesting listening experience in my opinion and superior in every way unless high spl performance over extended time is a constituent in which case reducing the strain of a single driver by adding more drivers and increasing the number of crossover bands is a good idea -- hence line arrays, synergy horns, and the likes.
Last edited:
I am thinking about specifically a driver like this one: https://www.bmsspeakers.com/fileadmin/bms-data/product_data_2014/bms_4599nd_preliminary.pdf. While I suppose you are correct that one could maybe build a line array style horn loaded speaker that would play at >130db using the Beyma though I am not certain how successful that would be compared with a driver like the 4599nd. I think the TAD diaphragm would vaporize at those levels at the intended crossover point. In summary, while the 4599nd, maybe Radians and other higher power rated compression drivers, apparently do not measure particularly well and are inferior for midfield listening at practical volume levels, I imagine these drivers dust a driver like the TAD or Beyma at a certain volume point. At that point the ladder two drivers would start to fall apart sonically and maybe also physically, whereas the former drivers could presumably easily handle several additional db of volume.
Lansing forum members seem to tend to prefer older Aquaplas driver like 2450SL.
Chris - Merry Christmas! I'll happily send the Axi-2050 back if you'd like to run more measurements with the throat lens - just let me know
And I think that sort of assessment is for luddites. I am not an engineer though I know what I like and it is not added coloration, resonances, and phase offset in the name of added acoustical flavor. A perfect direct radiator is superior to that same driver mounted to a waveguide, which is superior to that same driver mounted to a larger horn.
Last edited:
Then how do you explain that there is zero correlation between a THD measurement and its perception?
Other measures don't suffer from this problem.
Again, a rather blatant comment which I find to be wholly untrue. Direct radiators can never be constant directivity, while horn/waveguides can be. This effect is not a small factor, but a major one IMHO.
I don't oppose the use of waveguides, so not a luddite, if that even applies. I happen to like short horns (waveguides?) on woofers. But I prefer long, deep horns for compression drivers. I've heard 100s and know what sounds realistic to me. Waveguided CDs are nice, but I usually hear the CD diaphragm so much more than in a longer horn. Since that's a sound that I'd rather not hear, I use longer horns. Nothing to do with new vs old, just what sounds real and natural when I hear it.
I'm not afraid of horns, I know how to use them and make them sound right. Not a problem, but that skill did have to be learned.
I appreciate this point and your knowledge which is vastly more extensive than mine and accept that what you are saying is true. I intend to read your publication on the subject to better understand this. One question for you, how is directivity effected by listening distance? Are headphones constant directivity devices? Then would would in ear monitors be a greater degree of constancy? And can a speaker which performs excellent in the near field but poorly otherwise like an electrostatic achieve constant directivity in the near field? Or is it conventionally -- without performing some advanced manipulation of the diaphragm, driver, crossover, filters etc -- only in the mid and far field that constant directivity is achieved?
Hi Mike! And a Merry Christmas to you, too!
I might take you up on that offer, since I can see no one else here using the horn for which the lens/phase plug extension was designed (the K-402). My guess is that if the lens is used in horns with another throat geometry that's a lot different, it's likely going to get significantly different performance results-and likely not representative at all of that for which it was designed.
All of Roy's HF horn and compression driver work apparently makes use of redesigned compression driver phase plugs. Roy indicated that the technology used on the current lens/plug has been "on the shelf for many years", and that the lens/plug makes use of that technology. This is an area that Roy has not discussed over the 15 years that I've been in conversation with him (one-on-one and in a small private discussion group).
_____________________________________________________________________________________________
As far as all the diverging opinions that are expressed here (some more emphatically than others) on compression drivers, I still have to go with Dave Gunness' opinion that if the compression drivers/horns are corrected for transfer function (SPL and phase response), the polar coverage pretty much matches, and the longer time-based response (transient response) is about the same, the combination sounds the same (i.e., NOT harmonic distortion figures). Gunness stated that all of their Fulcrum Acoustic loudspeaker models are interchangeable due to their matching transfer function and transient performance properties in all their loudspeakers. I independently found essentially the same thing at a much later date without knowledge of Gunness' statements beforehand.
When I dialed in and listened to the Axi2050 on a K-402 (without said lens/phase plug extension) for a month in my setup, I used one Axi2050 on the right and a TAD TD-4002 on the left (their transfer functions were that close to each other to enable this configuration). That's how I evaluated the difference between the two drivers on a K-402 horn--over a long span of time (properly dialed in on large format K-402 horns). The difference was quite small, except the highest octave, where the TAD TD-4002 had more "sparkle", which I now attribute to better polar coverage of the 4002 in the 6-20 kHz band.
Hearing the breakup of 2" exit titanium and aluminum dome compression drivers is always detectable to my ears. But the TAD TD-4002 and the Axi2050 don't have audible dome diaphragm breakup modes, as well as the BMS 4592 ND (bi-amped dual ring radiator driver with time alignment of diaphragms and zero electrical phase shift between the two diaphragms). I assume this holds for the other full-range BMS (and perhaps B&C) dual diaphragm ring radiator drivers that are bi-amped and time-aligned/EQed flat. JMTC.
Chris
Last edited: