Assorted driver tests

[augerpro]

Results have been posted fellas.

Images not available

Please read the READ FIRST folder. Just to remind you guys, the wiggle in the impedence from 10k-20k is from my soundcard, it's not real. Just thought I'd mention it again since some of these Z's wiggle a bit already. Also the harmonic distortion plots were generally 96-98dB under 1.5k Hz, and 98-101 over 1.5k Hz or so, due to the rising response of these drivers. Yeah, these can scream alright.

[JamesK]

bah. I'm all excited to read the results and photo-bucket is on maintenance.

[joecarrow]

Results have been posted fellas.

Images not available

18 Sound 8NMB420
B&C 6NDL38
B&C 8NDL51
Beyma 6G40Nd

Please read the READ FIRST folder. Just to remind you guys, the wiggle in the impedence from 10k-20k is from my soundcard, it's not real. Just thought I'd mention it again since some of these Z's wiggle a bit already. Also the harmonic distortion plots were generally 96-98dB under 1.5k Hz, and 98-101 over 1.5k Hz or so, due to the rising response of these drivers. Yeah, these can scream alright.

Having had a look, my first impression is that the overall distortion looks quite low compared to the other drivers you've tested, but unfortunately it's the EVEN order products that are a good bit farther down! If the proportions of each product were more in line with home hi-fi drivers, the pro drivers do look like their measurements would stack up. Anyway, that's just my snap judgment- I have to head to work so I can't get too deep into reading those graphs right now.

Oh, and after a quick look at multi-tone, it does look like the B&C pulls ahead of the other pro drivers.

Very interesting stuff! Thanks, Brandon!

[mazurek]

Very interesting, it certainly looks like it would be worthwhile to pursue testing of some of the other drivers in those families. I will have to wait until I get back from my trip to recheck some of my testing results, and see how they compare with yours. I believe your graphs tell me similar things. I think I ultimately passed on my BC6NDL38 because of the trend towards third instead of second order HD, but its bigger brother seems to test differently at higher frequencies.

[cjd]

That 8" B&C looks like it could handle a surprisingly high crossover point (the shape helps that too!). Might be superb for open baffle work. Hmmmmmmmm...

C

[augerpro]

I agree the 8NDL51 looks like a superb driver-by any standard. The two 6" are also very nice, although the compromises are more apparent...I'd love to see a carbon fiber cone on that Beyma motor & suspension! The 8NMB420 doesn't impress that much, although the bass performance is very good for a PA driver. I bet the larger ones in that family are very nice for woofer duties.

The results were definitely eye opening for me. On one hand, distortion is generally not as good the better hifi drivers, and FR and Z wiggle a bit, some a lot, some not so bad.

On the other hand, they don't take a lot of power to be loud. And their peak SPL is probably much higher than a hifi driver, and probably better behaved HD at those high levels too. And of course their is the oft cited but little quantified "higher dynamics" that these PA drivers are associated with. Lower power compression and frequency modulation due to changes in impedence should also be much lower.

And with the 8NDL51, it seems to do quite well with HD, better than Dayton, not as good as Excel or Revelator. FR is nice too. But at high level I expect the 8NDL51 to pull ahead of it's hifi competion, driver for driver. Anyway that's my take from my current understanding of the compromises. The search for the perfect driver goes on...

[augerpro]

That 6NDL38 deserves a closer look I think. Kind of in a middle ground between PA and hifi. Good distortion, particularly in the lower frequencies, highish sensitivity, neo motor and open basket, and I would expect lower power compression and high peak SPLs like all good PA drivers. Only downside I see is an FR that would take some finesse.

Does the distortion at low frequencies give some indication of how the driver will behave at a higher frequency but also higher excursion? Like at 300Hz at high SPL, or OB?

[joecarrow]

Distortion at lower frequencies is partially an indication of distortion at higher drive levels, but not totally. As far as I know, modulation of Le and BL will rise more sharply at some point that is not indicated by low frequency low level performance. I would be really interested to see how these drivers fared another 10 or 20 db higher, if the hi-fi drivers could even keep up.

There are a lot of sounds and noises that just have a lot of visceral impact, and you'll never really be fooled that they're in the room with you if it isn't loud enough.

[Ludvig]

On the other hand, they don't take a lot of power to be loud. And their peak SPL is probably much higher than a hifi driver, and probably better behaved HD at those high levels too. And of course their is the oft cited but little quantified "higher dynamics" that these PA drivers are associated with. Lower power compression and frequency modulation due to changes in impedence should also be much lower.

I have made quick comparison between a 8" PHL 2460 and my RS180. At lower levels the RS180 is slightly ahead in distortion performance, but when the SPL go up, the RS180 curves are rising considerably while the PHL just take a walk in the park so to say. That is also clearly audible. It is not comfortable to the ear to push the Daytons, but the PHL's just play it louder with almost similar ease as with lower SPL's. If I remember correct, the second order distortion from the PHL is not very low, but the rest is. It is also evident that the PHL does not clip the dynamics as most of the hifi-speakers out there do. Midbass transients hit my chest in a way that I have never experienced with "hifi-drivers".

However, the PHL do suffer from som low midrange wiggles in both frequency and impedance, so they are probably not the last word in midrange resolution. I'm eventually going to get a pair of BMS 8N215 to compare with. These have tripple demodulation rings in their neo motors (single in the PHL) and according to datasheet a very flat, well behaved impedance curve with low inductive rise. No driver can measure that bad in frequency response as in their datasheet (large wiggles already in very low frequencies), so I'm excited in how these measure in real life.

[300Z]

Just in case this haven't been posted here before, the 8NDL51 was tested on the Klippel by Npdang earlier this year.
The results can be seen here!

Leo

[JoshK]

Hi Brandon,

I just noticed the measurements you did on the pro mids. Thanks a bunch! I've long wanted to see measurements done that were compared to more standard hi-fi drivers. This is great stuff!

I have a pair of JBL 2123H 10" mids. I was wondering if you have any desire in testing these? They are a bit harder to come by now, mostly obtained off of epay. I'm interested to see how they would stack up to a more modern pro mid. They have a pretty strong reputation so hence my curiousity. If you are at all interested I'll send you one or both to test out.

My intention was to use them in a hi-eff OB 3 way as a mid. I bought them purely on reputation to give them a try. I wouldn't be hurt if they sucked, in fact might be pleased as I can easily flip them on ebay for at least what I paid and buy the cheaper modern pro mids. I have been reading a lot on hi-eff forums and it seems the more unobtainium a driver is the more its praised. I'd just like to put a ruler on that.

[augerpro]

PM sent Josh

[augerpro]

Some new drivers at PE. Dayton has new cheapish neo tweeter that may be interesting.

Dayton ND28F-4

Morel ET338 tweeter

Morel EM1308 mid-dome

Jantzen JA-2806 tweeter

EDIT: I ordered the Dayton tweeter.

[cjd]

The Dayton tweeter is unfortunately likely to be like their classic line "silkie" - definitely not a bad tweeter but clearly budget. However, I have hopes that it sells well and they consider an RS "version"

[---k---]

Yeah, a Neo RS Tweeter would solve some problems.

[Brian Bunge]

I sure as hell am ready for one...

[gitarretyp]

Thanks for making these measurements, they're very helpful. Would it be possible to get some on versus off-axis (45 degrees is fine) plots along the lines of what eighteensound publishes with their drivers?

[augerpro]

Thanks for making these measurements, they're very helpful. Would it be possible to get some on versus off-axis (45 degrees is fine) plots along the lines of what eighteensound publishes with their drivers?

Well I don't have all teh drivers on hand anymore. I do have the 8NMB420 and 8NDL51 still though. If I get a chance maybe I'll measure this, if I don't sell teh 8NMB420 before then.
While the cone shape can contribute to the off axis response, the driver will still be basically bound by cone diameter vs. wavelength. So I don't know how valuable it would be...after all I think most of us would use other yardsticks to decide between, say, two 8" woofers for a mid. If off axis response was a priority I'd say you'd be better off dropping to a 6", compared to any small difference that may exist between the two 8". IMHO.

If someone sent me the Ciare 8.64 NdW perhaps I could be talked into dragging the others out and doing an off axis measurement *hint hint*

In other news Creative Sound has finally released the SDX7. It supposed to a kinda sorta replacement for the Adire Extremis.

[gitarretyp]

The project i'm considering would use an OB mid and a waveguided tweeter with the goal of an even power response by crossing the mid where its dispersion matches the waveguide. Cone shape and cone modes significantly effect the off-axis response. The wildly different off-axis responses from 18 sound's various 12" neodymium drivers illustrate this pretty well. Also, looking at ~1.5-2kHz for the 6nmb420 vs the 12nmb420, the 12 is beaming about the same as the 6.

I'm interested in the pro drivers because high'ish efficiency and output are also some of the other goals i have in mind. As such and having seen that most pro drivers start having problems much above 2khz, i'm primarily interested in 8 or 10" drivers to keep the even power response.

[augerpro]

Man I can't believe I haven't posted in this thread for so long! I knew I worked a lot of overtime the last nine months but jeez...anyhoo I need some advice from the big brains here. I believe I can do Klippel's performance based tested outlined here: http://www.klippel.de/pubs/Klippel%2...nt_XMAX_02.pdf
I may be able to do the parameter based method too but I'm not sure of SE's capability there. Anyway I have a couple questions. Klippel mentions relatively cheap laser measurement devices to measure cone movement. Does anyone know where I could find this?

Also he mentions the test steps here:
"1. Measure the resonance frequency fs of the driver.
2. Excite the driver under voltage drive with a two-tone
signal at f1=fs and f2=8.5 fs with an amplitude ratio of
4:1.
3. Perform a series of measurement while increase the
input amplitude and measure the peak voice coil
displacement and the sound pressure in the near field of
the driver. Perform a spectral analysis of the sound
pressure signal and determine the total harmonic
distortion and intermodulation distortion according IEC
60268.
4. Search for the minimal value of the peak displacement
where either dt, d2 or d3 are equal to the threshold d.
5. State the peak displacement Xmax, and the type of
distortion limiting the excursion."

What does he mean by driving the amplitude at a ratio of 4:1? I'm assuming he's not talking about the input voltage. So should F2 be 6dB higher than F1?

[Dennis H]

It's not a laser but SL shows how to measure ballpark P-P cone movement for free.

http://www.linkwitzlab.com/faq.htm#Q29

[JonP]

I know how you feel... been pretty busy on my end, too.

Congrats on getting back into things... been fighting my way back as well.

So... he mentions "voltage mode drive" and "amplitude ratio" so that does sound like a 4:1 voltage ratio between the F1 and F2 signals. Seems to be the intent.

I was looking around for laser distance devices, a year or so back. Did some asking around of a few ME types at work, and it seems that "low cost" was several hundred, maybe a few thousand, for something that can measure down to the fractions of a mm we would need. Maybe I didn't find the cheaper stuff, but I kind of gave up at that point. Then again, lots of innovation going on out there, and a small and cheap one would be a desirable device for the manufacturing industry...

I did find SL's great little "mechanical micrometer"... that thing just impressed the heck out of me. A wonderful, innovative, different idea. Perhaps he didn't invent it, but I'll take it as a example of his brilliance.... fold a piece of paper to solve a difficult and sensitive measurement problem.

[augerpro]

Golly Dennis that's pretty brilliant! I suppose it could work well enough-Jon I'd be happy with 0.5 millimeter accuracy I think. I found this posted on the SE user forum: http://www.acuitylaser.com/AR200/index.shtml
Not sure on the price though.

[Piotr]

Another way of measuring excursion of LF drivers;

Use a mick stand or maybe one of those thingies with alligator clips that you use as extra hands when soldering small parts. Attach something to this rig/stand.. a drill, a pencil, a screw or something. Place the tip of the device so it points to the apex of the cone. Measure the distance and slowly increase the level to the driver. When you precisely can hear the cone bump into the "thing".. press paus, remove the device and press play again and measure voltage or spl or whatever.

Or work the other way around.. Play a certain level and slowly move the device closer to the apex until they make contact. Press paus and measure the distance.


/Peter