2" Peerless by Tymphany - nice driver, right ?
https://audioxpress.com/article/test-bench-peerless-by-tymphany-full-range-2-pls-50n25al01-08
wait what is this ?
20% THD ? how ???
well ...
"For the distortion measurement, I mounted the Peerless by Tymphany 2” rigidly in free air"
he always does that ... FREE AIR ... WHY ??? he had it mounted in an enclosure for his other tests:
"I mounted the PLS-50N full-range in an enclosure, which had a 9” × 4” baffle and was filled with damping material (foam)."
but for distortion measurement it is now in free air ... "rigidly mounted" ... he says it like it's some kind of an achievement ... as if mounting it flexibly would produce a different result ...
that makes no sense ... of course without enclosure the low frequency of fundamental will get cancelled out by the back wave while the higher frequency of harmonics will not, causing the measured ratio of harmonic to fundamental to shoot way up to something absurd like 20% THD ...
does anybody actually believe that a driver from a major manufacturer would have 20% THD in the middle of the passband when measured correctly ?
why does he keep doing this ? isn't he the guy who wrote the loudspeaker design cookbook ? surely he knows better ?
what is going on ?
https://audioxpress.com/article/test-bench-peerless-by-tymphany-full-range-2-pls-50n25al01-08
wait what is this ?
20% THD ? how ???
well ...
"For the distortion measurement, I mounted the Peerless by Tymphany 2” rigidly in free air"
he always does that ... FREE AIR ... WHY ??? he had it mounted in an enclosure for his other tests:
"I mounted the PLS-50N full-range in an enclosure, which had a 9” × 4” baffle and was filled with damping material (foam)."
but for distortion measurement it is now in free air ... "rigidly mounted" ... he says it like it's some kind of an achievement ... as if mounting it flexibly would produce a different result ...
that makes no sense ... of course without enclosure the low frequency of fundamental will get cancelled out by the back wave while the higher frequency of harmonics will not, causing the measured ratio of harmonic to fundamental to shoot way up to something absurd like 20% THD ...
does anybody actually believe that a driver from a major manufacturer would have 20% THD in the middle of the passband when measured correctly ?
why does he keep doing this ? isn't he the guy who wrote the loudspeaker design cookbook ? surely he knows better ?
what is going on ?
Last edited by a moderator:
Very easily:
For the distortion measurement, I mounted the Peerless by Tymphany 2” rigidly in free air and used a noise stimulus to set the SPL to 94 dB at 1 m (6.77 V). Then, I measured the distortion with the microphone placed 10 cm from the dust cap. This produced the distortion curves shown in Figure 15.
The 20% value, which as you can see is heavily dominated by HD2, is a 1m value normalised to 94dB @ 6.77v input. QED.
It normal practice in order to eliminate any potential loading influence from an enclosure, since by definition they all vary.
The same in reverse. He's measuring FR &c. from 300Hz upward, and some prefer to measure FR above the driver's mass-corner frequency in sealed test boxes when they don't have a very large anechoic chamber available rather than using a test baffle. This is more a matter of preference since it can include some degree of edge-diffraction in the midband depending on the damping used, and a variable amount of baffle step loss. Those are givens, but largely irrelevant since they are clearly seen & can also be back-calculated from the provided baffle dimensions using the standard formulas. Seas for e.g. originally used IEC test baffles, but shifted to this method about 20 years ago.
He says it, like any disclaimer, for clarity & to confirm the conditions. We all know how much assumptions count for. 😉
This one is with approx 1 magnitude better THD, that is ten times:
https://audioxpress.com/article/test-bench-wavecor-s-fr4x6wa01-oval-full-range-driver
Those are even better:
https://audioxpress.com/article/tes...r070wa05-aluminum-cone-full-range-2-75-driver
https://audioxpress.com/article/Test-Bench-SB-Acoustics-SB65WBAC25-4-2-5-full-range-driver
Those are just ~5x smaller THD percentage
https://audioxpress.com/article/test-bench-the-w4-2356-4-full-range-from-tang-band-tb-speaker
https://audioxpress.com/article/faitalpro-2fe24-pro-sound-2-5-full-range
Or this one...
https://audioxpress.com/article/tes...e-subwoofer-from-the-dayton-audio-epique-line
If the speaker cost like s**t, looks like s**t, plays like s**t, measures like s**t - probably it is s**t. Case closed.
Chose another.
https://audioxpress.com/article/test-bench-wavecor-s-fr4x6wa01-oval-full-range-driver
Those are even better:
https://audioxpress.com/article/tes...r070wa05-aluminum-cone-full-range-2-75-driver
https://audioxpress.com/article/Test-Bench-SB-Acoustics-SB65WBAC25-4-2-5-full-range-driver
Those are just ~5x smaller THD percentage
https://audioxpress.com/article/test-bench-the-w4-2356-4-full-range-from-tang-band-tb-speaker
https://audioxpress.com/article/faitalpro-2fe24-pro-sound-2-5-full-range
Or this one...
https://audioxpress.com/article/tes...e-subwoofer-from-the-dayton-audio-epique-line
If the speaker cost like s**t, looks like s**t, plays like s**t, measures like s**t - probably it is s**t. Case closed.
Chose another.
Expected. It is 2”.
Note that in his sudies Gedes suggested measured distortin does not correlate with sonics until something like 25%.
This is a midTweeter, XO well above 250 Hz.
dave
Note that in his sudies Gedes suggested measured distortin does not correlate with sonics until something like 25%.
This is a midTweeter, XO well above 250 Hz.
dave
you're on to something ...
even though the wavecor is slightly larger and was used at slightly lower SPL ( 90db vs 94db ) that still isn't enough to explain the difference in distortion levels.
distortion plot of wavecor looks as you would expect - it keeps increasing as frequency goes down
with peerless there is an enormous peak at 800 hz which seems out of place and looks like some kind of defect of either the driver or measurement ... below the peak distortion goes back to levels you would expect to see ...
even though i would only use the driver above 1 khz anyway a properly designed driver shouldn't have such a massive anomaly in its passband which is what gives me pause about this driver ...
it's a shame because it's just $17 from digikey which makes it really attractive option for a line array ...
EDIT: possible explanation is some spurious noise originating from the back of the driver that would be muffled by the enclosure but that is a risky assumption ...
Distortion is calculated as a ratio.
Harmonics(power)/Fundamental(power) or H(dB)-F(dB).
Harmonics are always above the fundamental, 2x, 3x, 4x etc.
In free air, at least for woofer and midrange, lower frequencies are short-circuit so the F goes down more than H so the ratio (THD) goes up, since F is artificially reduced. And same happens to tweeter due to natural rolloff. So this first argument indicates that THD is expected to be higher in free air as you reduce the frequency.
The second argument, is what you pointed out: sound produced by the back of driver indeed has more distortion.
And here my speculation: you have an inverted cone without the central part, inverted boards, you have the spider and the wires all producing sound.
Nothing similar to the front cone shape well designed and optimized to produce the right sound with minimum distortion.
Take a regular US$15.00 mid-range and put a 100/200Hz tone in free air - the sound might be horrible distorted that you don't even need to measure to say that is well about 10%. But, when you install it in a enclosure you get double gain: no short-circuit and you remove the sound that is produced behind the speaker. I've experienced that with some regular speakers - I cannot say that this happens to all speakers, since I don't work with high end speakers.
For these regular speakers, once installed in the enclosure, you can clearly hear that distortion now is at acceptable levels so less than 2 or 1%.
Does it make sense?
Harmonics(power)/Fundamental(power) or H(dB)-F(dB).
Harmonics are always above the fundamental, 2x, 3x, 4x etc.
In free air, at least for woofer and midrange, lower frequencies are short-circuit so the F goes down more than H so the ratio (THD) goes up, since F is artificially reduced. And same happens to tweeter due to natural rolloff. So this first argument indicates that THD is expected to be higher in free air as you reduce the frequency.
The second argument, is what you pointed out: sound produced by the back of driver indeed has more distortion.
And here my speculation: you have an inverted cone without the central part, inverted boards, you have the spider and the wires all producing sound.
Nothing similar to the front cone shape well designed and optimized to produce the right sound with minimum distortion.
Take a regular US$15.00 mid-range and put a 100/200Hz tone in free air - the sound might be horrible distorted that you don't even need to measure to say that is well about 10%. But, when you install it in a enclosure you get double gain: no short-circuit and you remove the sound that is produced behind the speaker. I've experienced that with some regular speakers - I cannot say that this happens to all speakers, since I don't work with high end speakers.
For these regular speakers, once installed in the enclosure, you can clearly hear that distortion now is at acceptable levels so less than 2 or 1%.
Does it make sense?
because i am looking at this micro full range driver as an alternative to a midrange ribbon in an array i also now looked at his measurements of Radian LM10N ( he didn't review then 8" which is what i actually am interested in )
https://audioxpress.com/article/tes...bbon-transducer-from-radian-audio-engineering
but perplexingly for the LM10N he measured distortion IN AN ENCLOSURE
"For the distortion measurement, the LM10n was again mounted the same enclosure as was used for the frequency response measurements"
this guy is killing me ...
https://audioxpress.com/article/tes...bbon-transducer-from-radian-audio-engineering
but perplexingly for the LM10N he measured distortion IN AN ENCLOSURE
"For the distortion measurement, the LM10n was again mounted the same enclosure as was used for the frequency response measurements"
this guy is killing me ...
although i agree with what everybody has said i think i will go with the interpretation from SVP and here is why ...
when Vance reviews drivers if the driver doesn't suck he concludes by saying something like " after reviewing all the data this is an interesting / well designed driver " etc.
but when the driver sucks he simply finishes all the tests and leaves out his opinion of the driver completely.
in the case of this peerless he left no opinion ... which means he thinks the driver is trash.
thanks everybody.
when Vance reviews drivers if the driver doesn't suck he concludes by saying something like " after reviewing all the data this is an interesting / well designed driver " etc.
but when the driver sucks he simply finishes all the tests and leaves out his opinion of the driver completely.
in the case of this peerless he left no opinion ... which means he thinks the driver is trash.
thanks everybody.
You seem to be assuming Vance is ignorant. The reason why is in the text. The LM10n, being what it is, can [quoting]
...be utilized in both open-back (dipole) and closed-back (monopole) formats.
You'll now understand why the box mounting was retained.
Well, maybe we need some more comparisons, if Mr. Dickasson measures the drivers in different conditions... and those conditions are not in the same baffle.
I don't know if we can trust them at all. In any case 10% THD is -10dB from signal, so than can be heard for sure, especially if this is 3rd harmonic (metal frame?).
I don't know if we can trust them at all. In any case 10% THD is -10dB from signal, so than can be heard for sure, especially if this is 3rd harmonic (metal frame?).
Maybe have a look at the editorial note that's been prefaced to the latest Voice Coil test: https://audioxpress.com/article/tes...xial-driver-from-the-sb-acoustics-satori-line
Well, in this case it has high distortion but the usual appraisal is actually not present. My critique has been that he sings praise in the end for drivers that I don't think has been fairing very well in the presented evaluation.
//
//
How many do you need? There are dozens (probably hundreds) of his Voice Coil tests available on-line, last I checked.
He mounts them in different (appropriately) sized sealed test boxes for relevant measurements -a fairly common practice.
A bit like Seas et al. Well, exactly like, actually. 😉
Makes zero difference other than for baffle step, which is clearly visible in the measurements & can be back-calculated for the supplied dimensions using any of the standard formulas, and are irrelevant anyway for the testing purposes.
Amongst the most 'trustworthy' independent tests available. He's been flying the flag for it for quite literally decades now. Not perfect in the same way nothing is, but it's a whole lot better than many others.
I talked about the same drivers he tests Vs other testers. More testers of this driver, not the other way around. That is some kind of PA or Pro or OEM driver probably not much diy crowd cares about, and not much testers to have their hands on the same driver. Or maybe that driver is that uninteresting that this exact thread and that review is all there is in the www, literally. You can check and see 😉
My idea is - if the testing conditions are the same (?) for all the same class of drivers - then probably that driver is just bad. Or exact batch is bad. It happens. Or built to a price point - it is meant to be bad.
I have no clue how he does that. I listed couple of times all the writeups about drivers, and looked at his measurements tens of times - I do not remember seeing exact setup. If someone know - it would be nice to read about it. Maybe then it will explain why some of the drivers measures worse than others, when we all kinda expect them to be better.
And now looked at his other small drivers measurements, it is easy as probably most of them are tagged with fullrange tag - that OP driver measures worst.
I have only praise of the reviewer, no questions about that.
I have followed VC/VD reviews for a long time too. Still somtimes I am mislead at first glance... We must be careful to read his text about how measurements were taken. Like others I have noticed that he never gives heavy critic about the performance. And yes, most of the drivers tested are well above average performers.
HifiCompass is now the best resource for measurements IMO, but the list is naturally short
HifiCompass is now the best resource for measurements IMO, but the list is naturally short
...but gradually growing. Alas, a certain situation in that neck of the woods isn't helping matters, although in the scheme of things, driver tests come somewhere south of nil compared to 'other issues'.
Yes, generally the test setup is stated but it's either buried in the text or elsewhere in VC, the latter of which isn't always helpful if it isn't published along with the nominal driver test itself. Personally I'd prefer a format change as far as the presentation itself goes so the details are quicker to identify, but who am I?
Yes, generally the test setup is stated but it's either buried in the text or elsewhere in VC, the latter of which isn't always helpful if it isn't published along with the nominal driver test itself. Personally I'd prefer a format change as far as the presentation itself goes so the details are quicker to identify, but who am I?
There have been multiple times where Vance's distortion measurements have been subject to high levels of distortion in his measurement chain. Especially second order being significantly higher than it should be for accurate results.
i remember reading an impromptu "study" where they concluded it takes around 100% THD to hear distortion at lowest frequencies but only about 1% in the upper midrange.
this is why subwoofers are run flat out until they break and nobody complains about distortion but with compression drivers they tend to be run at like 1% of power and everybody still complains about them sounding harsh.
since the driver in question here is an upper midrange driver any excessive distortion is a concern.
Last edited: