If Xmax shouldn't be exceeded, why have a Xlim?

Well. Interesting reading. all I can add it that in my opinion a speaker is (usualy) designed or should be designed to reproduce a sound and not alter it. I kinda destroyed a speaker quite badly (shown below) so i wouldn’t purposely aim to over extend a driver on a regular basis. To me it seems to be asking for trouble. 

Edited by wayward91 - 01 March 2011 at 11:21pm

Think there could be something here worth considering which comes from early mathematical models of the 'ideal' Horn loading for a cone loudspeaker and still holds true today.

This is that the ideal loading for a driver occurs with zero cone movement as it models as the ideal drive condition since the throat coupling volumes remain constant.

Now we know that in the real world this isn't very practical; but to best match our driver to our horn loading then the cone displacement should be as small as possible to achieve a result closest to the theoretical ideal....

For greater deviation from this ideal, distortion rises whilst when exceeding Xmax, distortion rises substantially more.

So rather than trying to extract the 'maximum' from a single driver/cabinet design; the use of multiple under-driven cabinets are the better solution in terms of the system being capable overall of reproducing the original program source whilst minimising 'colouration' of the original program material it is reproducing....

Ok - other factors apply but the principle is valid as posted by others in this thread earlier
Mik

greeef wrote: I wouldn't trust this to the letter, there are a lot of different equal loudness contours out there,

That old Fletcher-Munson chart is extremely inaccurate, with the current standard being ISO 225:2003 - see here and here.

(a bit of a trhead hack, but it's on Xmax and i thought it appropriate to post this here)

The other day I replaced one of my beloved PD186 that had started scratching (for whatever reason?) and put in an 18Nd9300, which is a neo version of 18LW1400. Similar T/S, checked for plausible response and mech. movement using hornresp. I'm used to giving my bass bins 500 Watts each whole night long, so they should move about 8...10 mm in each direction which has proven totaly fine. Not too much power here.

Now what happens: that particular bin with the neo driver beats like it's hitting something. But only on high excursion, on loud drum beats. It was in parallel to a PD186 that had no problems. Turning down that particular amp by ~6dB cured the problem.

Do I just have a bad one? It's second hand, but I tested it with a tone generator, no scratching even at large displacement wo I thought it should be fine.

The 18Nd9300 is one of those 'new' drivers with 15mm gap and 24mm coil winding height. PD186 is one of those 'old' drivers with 8mm gap and 24mm vc. But then they promise similar Xmax figures and some say that having a less steep BL-curve leads to a larger controlled displacement anyway. There's also the DDS spider, where the PD has an 'old fashioned' double spider.

Has anyone had similar experience with 'modern' drivers?

Thanks in advance,
Robert

If you really want to know the answer, measure the TS Parameters of both drivers in question. You are overlooking many things that play a factor on how the loudspeaker is reacting in the box. Measure the TS Parameters of both drivers.

Best Regards,

thanks!

@Pasi: Maybe that's why they discontinued a seemingly sound driver? 15"ers and a 12"er are still available with the same magnet structure that the 18Nd9300 has, but no 18"er.

@Elliot: Point taken, i did not measure T/S. Still, such entirely different behaviour is a little bit unexpected, judging the datasheets, isn't it? 500W in any enclosure that properly loads the driver should be ok with this type of driver (35Hz HPF was set, just plain-old techno beats on and on). In the end of the day, the Nd9300 is quite high-tech compared to PD186 and you wouldn't expect 18sound to totally screw up the datasheet, would you? (I admit that PD kind-of did exactly that).

@Mark: They were in 3 stacks of 2 bins each, with only half a meter between stacks. So that shouldn't have been the problem. I'm going to have to do what you suggested though: have a direct stacked comparison between drivers. Because if this behaviour is structural, I can't use such drivers. Hopefully just one bad driver.

I must say I do like this forum for the type of discussion you can get here. Thanks again. This might also be interesting regarding the recent change of T/S on the much-used PD186.

Is there anyone else who can comment on 'new-fashioned' drivers, say 'P300 or 'LW1400 against PD186?

regards, robert