D2430K compression driver information?

[ivica]

http://www.google.co.uk/url?sa=t&rct...52109249,d.ZGU

Download the pdf.

One snippet it directly describes and has a diagram for a 1.5" coil version

Hi pos and macaroonie,

Thanks for the link and the file. In any case it would be interesting when ONE DAY some of our AH-forum member would make "independent" measurements with real available 1.5"-throat horns, either on axis and off axis, THD..... etc.
For me personally, comparison with 2" BMS 4590 or 4592ND especially with 'new-comer' 1.5" BMS 4595ND would be interesting...(745 U$D http://www.usspeaker.com/bms%204595nd-1.htm )

Regards
Ivica

[pos]

Here are some measurements comparing the D2 driver with a 2435 and a 2450SL equipped with Truextent Be diaphragms.

- PT-F95HF horn, laying on the floor, upward-facing
- dayton mic (calibration not loaded!)
- ~50cm distance from the horn mouth (not precisely matched between drivers of different depth...)
- 4th order HP filter at 800Hz using Hypex AS100D plate amp
- no EQ
- all drivers are coming from ebay, but the Truextent Be diaphragm is new

The noise floor was pretty high, with hiss coming from the amp (no lpad or autoformer was used, and this of course impacts the lower impedance drivers more...).
Output level were adjusted in the filter to compensate for differences in nominal impedance.


D2430K vs 2435HPL



D2430K vs 2450SL-Be

[ivica]

Here are some measurements comparing the D2 driver with a 2435 and a 2450SL equipped with Truextent Be diaphragms.

- PT-F95HF horn, laying on the floor, upward-facing
- dayton mic (calibration from cross-spectrum)
- ~50cm distance from the horn mouth (not precisely matched between drivers of different depth...)
- 4th order HP filter at 800Hz using Hypex AS100D plate amp
- no EQ
- all drivers are coming from ebay, but the Truextent Be diaphragm is new

The noise floor was pretty high, with hiss coming from the amp (no lpad or autoformer was used, and this of course impacts the lower impedance drivers more...).
Output level were adjusted in the filter to compensate for differences in nominal impedance.

Hi POS,

Thanks for the measurements You have sent us.
Interesting that on the measurements fro the D2430K there are no deeps around 9.3kHz, 11kHZ, and 16kHz as with M2 horn

( look at the )
http://www.audioheritage.org/vbullet...l=1#post354296
http://www.audioheritage.org/vbullet...1&d=1383303224

I wonder what would be the drivers behavior with PT-H95 horn.

regards
ivica

[1audiohack]

Hi Thomas;

Well it certainly looks like the D2 will be nice to work with on the DIY side.

What is the lower trace on the scale? Does your test rig generate a phase response?

I will get back to you on your PM to me today.

Thanks for sharing!

Best,
Barry.

[Earl K]

Yes,

Thanks POS !

<> Earl

[4313B]

Yes, thank you very much for the post pos. I'm still not going to give up my 4-inch coils ( it's a legacy JBL thing ), but these things are considerably more reasonable in the price department. G.T. did mention a distortion issue with the D2, I think it was around 1 kHz, but I can't remember the gory details. And it sounds damn good in the M2.

[ivica]

Yes, thank you very much for the post pos. I'm still not going to give up my 4-inch coils ( it's a legacy JBL thing ), but these things are considerably more reasonable in the price department. G.T. did mention a distortion issue with the D2, I think it was around 1 kHz, but I can't remember the gory details. And it sounds damn good in the M2.

Hi 4313B,

May be LF ( <1kHz) distortion of D2 is involved by the M2 horn. It seems to me that M2 is a kind of compromise HF bandwidth and HF dispersion. It would be interesting to get some more data about D2 with the PT-H95 horn. I believe that LF distortion would be reduced, but HF section would lost bandwidth and dispersion.

regards
ivica

[pos]

What is the lower trace on the scale? Does your test rig generate a phase response?

Hi Barry

The lower trace is the 3rd order distortion.
The noise floor was pretty high, and the mic I used is also not ideal for distortion measurement (internal 2nd order is always higher than the measured driver for example...).
I should redo this with proper gain structure (an autoformer between the amp and driver) and a beefier mic (1/2" capsule), but I don"t think I will have the opportunity to do this anytime soon...

Regarding phase, it looks like all the drivers exhibit pretty much a perfect minimum-phase behavior over there bandwidth, so no surprise there.

Thanks for sharing!

my pleasure

[pos]

Yes, thank you very much for the post pos. I'm still not going to give up my 4-inch coils ( it's a legacy JBL thing ), but these things are considerably more reasonable in the price department. G.T. did mention a distortion issue with the D2, I think it was around 1 kHz, but I can't remember the gory details. And it sounds damn good in the M2.

Distortion seems to rise pretty quick under 1kHz (and level drops down), maybe that is the issue GT talked about?

[pos]

Hi 4313B,

May be LF ( <1kHz) distortion of D2 is involved by the M2 horn. It seems to me that M2 is a kind of compromise HF bandwidth and HF dispersion. It would be interesting to get some more data about D2 with the PT-H95 horn. I believe that LF distortion would be reduced, but HF section would lost bandwidth and dispersion.

regards
ivica

I think the horizontal HF beamwidth will be better with the PT-H than with th PT-F, thanks to the diffraction devices.

David Smith did an interesting comment about the M2 horn somewhere on the diyaudio forum (and I posted it here also).
He said something along the same lines as you: compared to a 2344 this new waveguide favors off-axis smoothness (relative to on-axis at least) over absolute directivity control.
Certainly a worthwhile compromise.

[pos]

Some random thoughts:

- I did some listening tests comparing the D2430K and 2450SL-Be with careful EQ, and they sounded more alike than different I would say... Almost impossible to distinguish. That said, it was restricted to the 1.2kHz-18kHz range, with no other drivers, and with the horns laying on the floor: not an ideal situation, and quite fatiguing.

- The truextent Be diaphragm has a major breakup above 20kHz that should be attenuated with EQ or it will trigger any preringing nastiness the DAC might have (if not a 96kHz DAC that is), as can be seen in the impulse measurement (preringing just before the impulse)

- The 2435HPL is truly an impressive small driver, with the highest mass break point frequency I have ever measured in a driver. Its phasing plug does not play well with that particular horn though, as we all know, and that ~13kHz hole cannot (and should not) be addressed with EQ (it moves off axis!). The 2432 phasing plug would probably be worth trying : 2432 front cap with 2435 back cap and diaphragm...

- The D2430K has a really smooth response up high with no apparent breakup or resonnace. This reminds me the 2450SL (Ti SL diaphragm) a lot : the mylar ring is probably very well dampened. It also have the same rise in the UHF that can be found on many other ring radiators, like the 2407H.

- The fact that the D2430K is a 32ohms driver (in 16//16 position) is a really nice feature as it will reduce the hiss (and distortion) coming from the amp (-3dB compared to a 16ohms driver, and -6dB compared to a 8ohms drivers). I wish more compressions drivers would be 32 or even 64 ohms devices, especially when used with active filtering: no more lpad or autoformer would be needed with most amplifiers (but it looks like the crown XTI is still not silent enough for this...)

[pos]

Here are some additional distortion measurements with a better noise floor (mic with a 1/2" capsule, and an autoformer to reduce hiss and distortion from the amp), comparing the aforementioned 3 drivers.
Each graph shows a different harmonic distortion order as the lower traces.
The mic is not calibrated, so disregard the (heavily smoothed) upper curve.
The drivers where grossly EQed for CD compensation and midband matching.

blue curve is 2450SL-Be
red curve is 2435HPL
green curve is D2430K

2nd order



3rd order



4th order



5th order



6th order

[ivica]

Here are some additional distortion measurements with a better noise floor (mic with a 1/2" capsule, and an autoformer to reduce hiss and distortion from the amp), comparing the aforementioned 3 drivers.
Each graph shows a different harmonic distortion order as the lower traces.
The mic is not calibrated, so disregard the (heavily smoothed) upper curve.
The drivers where grossly EQed for CD compensation and midband matching.

blue curve is 2450SL-Be
red curve is 2435HPL
green curve is D2430K

2nd order

Hi POS,

Thanks for the THD data You have done.
On the 2nd distortion curve it is 'visible' that D2430K driver is NOT so good for the lower frequency range ( under 2kHz), as 2435 or 2450. I can expect that the reason is much smaller diaphragms (two) total surface ( I can guess the surface is about 60% of the 2450 diaphragm). So I would expect that the larger horn would help there, but then, the problem can arise in the high frequency response.

regards
ivica

[pos]

Yes, maybe that is the issue 4313B talked about.
On the other hand it might also be a problem with that particular driver (from ebay), but I guess it would had showed also with other harmonic orders in that case...
I got a pair and the one you see here measured is the best one of the two: the other one had markedly more distortion, especially around 2kHz for even order harmonics :
https://www.dropbox.com/s/nwbce5k9mbdyfj7/D2430K.zip

Probably a misaligned VC, but I don't want to open that driver
I will get another one as a replacement from the seller and we will see what happens...

According to the AES paper the total radiation area of the two 3" ring (extending inside and outside of the VC) should be similar to a single 3" dome.
And the push-push orientation should reduce even order distortion....

[ivica]

Yes, maybe that is the issue 4313B talked about.
On the other hand it might also be a problem with that particular driver (from ebay), but I guess it would had showed also with other harmonic orders in that case...
I got a pair and the one you see here measured is the best one of the two: the other one had markedly more distortion, especially around 2kHz for even order harmonics :
https://www.dropbox.com/s/nwbce5k9mbdyfj7/D2430K.zip

Probably a misaligned VC, but I don't want to open that driver
I will get another one as a replacement from the seller and we will see what happens...

According to the AES paper the total radiation area of the two 3" ring (extending inside and outside of the VC) should be similar to a single 3" dome.
And the push-push orientation should reduce even order distortion....

Hi POS,

May be some diaphragm destruction....

I do not know the 3" dome exact data, but it seems that two 3"rings would have the same surface as 3" dome,
if 4"dome to be in comparison, then I would expect that for 4" dome surface would be about 85cm2, comparing to about 49cm2 for 3"dome.


regards
ivica