Flex Your PCD Mettle:

[Zilch]

Re: Flex Your PCD Mettle:

Indeed an interesting similarity on the compression drivers. Did you not need the spacer/adapter for both of these. I had thought it might have been problematic. Do either of these drivers have cleaner CSD than the DE250?

This is the 1.5" throat variant of the standard JBL eWaveguide, so no adapters required.

Here are CSDs and wavelet transform for the 2452H-SL load:

[NEO Dan]

Re: Flex Your PCD Mettle:

This is the 1.5" throat variant of the standard JBL eWaveguide, so no adapters required.

Here are CSDs and wavelet transform for the 2452H-SL load:

Oh, I thought the aluminum looking spacer was a throat adapter of some sort.

I was also wondering if you have any HF driver only polars posted anywhere? I'm asking because I'm currious to know if directivity changes appreciably depending on the CD on the WG.

[Zilch]

Re: Flex Your PCD Mettle:

Oh, I thought the aluminum looking spacer was a throat adapter of some sort.

Don't know what's up with that; it's part of the driver, a recent design, and I've never disassembled one. I should do that before posting the design in Lansing Heritage. ;)

I was also wondering if you have any HF driver only polars posted anywhere? I'm asking because I'm currious to know if directivity changes appreciably depending on the CD on the WG.

Only once, with the 6" square waveguide, in Smaller Advent cab with 1" overhanging edges, and on a highpass filter:

http://techtalk.parts-express.com/showthread.php?p=1648182#post1648182

Leading with high likelihood to erroneous conclusions with respect to directivity of that waveguide.

[Most embarassing; I'm going back to that one for a do-over.... :p ]

[Zilch]

Re: Flex Your PCD Mettle:

Do either of these drivers have cleaner CSD than the DE250?

EconoWave HP and Deluxe sound very similar. Deluxe has more extended bass response.

If Skywave's suggestion that waveguides without roundovered mouths are more sensitive to cabinet edge effects is correct, there's more yet to do toward improving HP's performance in the time domain:

http://techtalk.parts-express.com/attachment.php?attachmentid=10421&d=1281758552

http://techtalk.parts-express.com/attachment.php?attachmentid=6999&d=1266189868

http://techtalk.parts-express.com/attachment.php?attachmentid=10422&d=1281758572

http://techtalk.parts-express.com/attachment.php?attachmentid=7000&d=1266189891

[edlafontaine]

Re: Flex Your PCD Mettle:

Over on diyAudio in the "waveguide and horns" thread, there is discussion with a vendor about interest in a variety of offerings.

http://www.diyaudio.com/forums/group-buys/166312-waveguides-horns.html

One of them is referred to as "Iwata". It can be sized for a specific cut-off frequency. If enough interest is generated (6) they can be fabbed for purchase. Pictured below is a 250 Hz Iwata. It is suggested that a 500 Hz cut-off would be a better fit and within the scale of these projects.

[donprice]

Re: Flex Your PCD Mettle:

eWave PCD for dummies, round 4

Kappalite 3015 QSC 152i/ B&C DE250

FRD files from HolmImpulse in-room measurements - measured at center of HF/LF at 44" on axis, mic ~60" above floor and below ceiling. Gate value manually set @ 5 ms ~200 Hz for both. Time

3015 ZMA from Jeff's response modeler
DE250 ZMA from 3012HO/DE250 posted previously - is this good enough?

CSP file copied as text file...have not tried pasting into excel file, renaming as *.csp, and reloading. It should work - may have to copy over existing csp file to force recognition as CSP instead of workbook.

Will try to paste screen shots from PCD next.

[donprice]

Re: Flex Your PCD Mettle:

Not quite sure what to do about peak between 4 & 5 kHz.

Second opinions greatly appreciated.

Woofer vertical offset should be -0.156 m.

[Pete Schumacher]

Re: Flex Your PCD Mettle:

Not quite sure what to do about peak between 4 & 5 kHz.

Second opinions greatly appreciated.

Woofer vertical offset should be -0.156 m.

You need to attenuate the tweeter level a few more dB overall.

Try a shunt notch, a series RLC, in parallel with the tweeter to take care of that bump between 4 and 5K. You can size it to have the correct bandwidth and then vary the resistor to dial in the total amount of cut.

By the way, what does "edited" mean in those FRD files?

[donprice]

Re: Flex Your PCD Mettle:

Thanks Pete!

Can the L-pad take care of the excess HF response or do I need to get closer? Flatter? I really don't have a feel for what "shape" or effect the L-pad has.

Shunt notch? OK, I'll read up on it a little and give it a try. I'm curious why I'm getting that bump with this HF combo when I don't see it in the other runs using the same.

Edited means I took the raw FRD files from HOLMIimpulse and cut out the header, rounded to 2 decimal places, and added a 100 dB offset to bring it up to about 90dB reference level.

I see that playing with the "vertical off-axis angle" spinner shows the null near the crossover frequency and it is down ~6 dB at +6 degrees and -20 degrees. Is this an illustration of the forward lobe concept? It also looks like the null is less sensitive to + vertical angles at the listening distance (3.5 m) than at the measuring distance (1.1 m) though that requires reversal of the tweeter phase (or was it the reversal that put the forward lobe higher). I see that I can drive myself nuts with the model.

[Zilch]

Re: Flex Your PCD Mettle:

Now build it and measure the result. Do they correlate? If not, track down the problem -- I'd be looking to see if that peak is real, or an artifact of some error in the input data.

I see that playing with the "vertical off-axis angle" spinner shows the null near the crossover frequency and it is down ~6 dB at +6 degrees and -20 degrees. Is this an illustration of the forward lobe concept?

Yes, exactly.

See Wayne's video:

http://www.pispeakers.com/misc/Vertical_Nulls.wmv

At this point, find the angle of the actual maximal inverse-polarity null via measurement, invert the tweeter polarity in PCD with the vertical offset set to that angle, and adjust the tweeter "z" offset to achieve maximum null there. That should "dial in" the amount of the actual acoustic center differential, and with the tweeter polarity switched back to your design, the vertical spinner will illustrate the locations of the vertical nulls. I use +/- 6dB intrusion of the nulls as the criterion for defining a usable listening window.

In the course of doing that, you'll discover just how critical getting these values correct really is: a change in "z" of just 1mm has a substantial effect in the location of the forward axis and lobe, and how changes in filter orders and component values may be used to "steer" the axis to the desired location, as Wayne describes.

[But doesn't tell us how to accomplish.... ;) ]

[donprice]

Re: Flex Your PCD Mettle:

Close enough to build? That's what I want to hear. I might have it up and running in a week.

Yup, I've watched Wayne's video. Hooking up the mic to a string pinned at Z=0,0,0 should make capturing polar data simpler. Now all I need is to automate the data collection process. Changing the software is out, the cat is useless for holding a mic, so that leaves the SO or my brother. Sounds like a 3-beer project (6 counting prep and cleanup :D).

Thanks!

[Zilch]

Re: Flex Your PCD Mettle:

Close enough to build?

[A prototype, using wire nuts.... ;) ]

[NEO Dan]

Re: Flex Your PCD Mettle:

[A prototype, using wire nuts.... ;) ]

Ain't that a kick in the nuts:eek:

[Zilch]

Re: Flex Your PCD Mettle:

Ain't that a kick in the nuts! :eek:

Even at this point, I wouldn't build anything final without first confirming the PCD design via measurements of a prototype.

SPEAKING of which, it's a little late in this project to be checking design compatibility across driver manufacturing lots, but here it is, 4 ea. 2452H-SL purchased used, of indeterminate provenance, months apart; #8, red, is the one with which I have been designing:



[Looks like I have already sorted them into matched pairs.... ;) ]

[bruson]

Re: Flex Your PCD Mettle:

Thanks for the info about my questions.

I'm going to delay doing much about my temporary speaker until fall.

Has anyone done a MTM arrangement with the Econowave?
Or does that add to the null problems?

I was thinking about it because my 10" isn't as dynamic as the tweeter.
Would having 2-10" drivers below the tweeter be more dynamic, instead of MTM?

OFF TOPIC: I heard live music at a Mexican restaurant last night and even though I couldn't see them, I instantly knew it was live unamplified music.
No system I've ever heard comes close. (Especially mine).
The effortless dynamics and power were wonderful to hear and feel.