Jump to content

Crossover mods for the AR4x II


Recommended Posts

  • Replies 88
  • Created
  • Last Reply
What's the difference between the two charts posted test-wise?

#1-#6 and no grille seems consistent between the two.

The horizontal and vertical scales. Harder to tell what's what on the 10 dB plot, but it puts them in a full-range context.

[There's a 20 uF protection cap in series with all of the measurements.... ]

Link to comment
Share on other sites

Discussion:

1) Tweeter uniformity - there is considerable variability among samples, but the general response trend is similar and a single crossover design could work well with all, provided that level adjustability is included. Matching pairs in detail would yield optimum results.

2) Grille - benign, but a bit better with that in place; it seems that most are used this way, as they're not particularly handsome without. Design accordingly.

3) Directivity - Uniformly narrow (~45° beamwidth) above 10 kHz. The on-axis VHF peak is somewhat mitigated off-axis, but there are other issues apparent inboard between 10° and 20° at 2.4 kHz and 3.8 kHz, and HF directivity collapses at 25°. Thus, not suitable for orthagonal 90° deployment; design for on-axis, toed-in. If they were mirror-imaged, there'd be more options, but alas, they're not. Could be better horizontal, but that determination can only be made in combination with the woofer response and resultant driver interaction about the crossover region.

Link to comment
Share on other sites

So is the objective to mod the crossovers to try to get the speaker with the OEM tweeter closer to the E-wave version, or the other way around...?

Neither.

The objective is to see if us Westies can better Dave's design using the stock tweeter(s) (we have a baseline, now), and then see how an eWave variant performs in comparision.... :)

Link to comment
Share on other sites

Neither.

The objective is to see if us Westies can better Dave's design using the stock tweeter(s).

Bring it on guys!

Nice group of curves. Any idea about the asymmetry around 1500Hz? I can't tell if it is a cabinet effect or crossover effect. Do you have any phase curves of woofer and tweeter individually?

Variation in the impedance curves and also of the tweeter LF response seems to indicate that the silicon grease is pretty hit and miss. The 8 ohm L-pad seems to work well for compensating tweeter sensitivity.

David

Link to comment
Share on other sites

Neither.

The objective is to see if us Westies can better Dave's design using the stock tweeter(s) (we have a baseline, now), and then see how an eWave variant performs in comparision.... :)

And what is the criteria for "better" in this exercise...?

Link to comment
Share on other sites

Any idea about the asymmetry around 1500Hz? I can't tell if it is a cabinet effect or crossover effect.

The first two sets of curves were done with just a 20 uF cap in series, and it's there in those. The pattern stays throughout the progression; I'd say it's the cab edge and the tweeter location on the baffle.

Do you have any phase curves of woofer and tweeter individually?

I'm showing normal phase at normal polarity in the inverse polarity plot. That -40 dB notch clearly defines the location of the forward axis, by my view. The vertical polars in Post #12 led me to look there:

http://www.classicspeakerpages.net/IP.Boar...-1277516236.jpg

Color code for those is:

On axis midway between drivers = Red,

Up toward tweeter = Yel, Org, Brn, Gry,

Down toward woofer = Grn, Blu, Vio, Cyn.

The upper null is prominent at +20°, whereas the lower one hasn't yet begun to form at -20°, indicating that the forward axis is down below center somewhere.

I'll be generating FRDs for PCD, which will include minimum phase for both. I'll post those plots here, as well.

Variation in the impedance curves and also of the tweeter LF response seems to indicate that the silicone grease is pretty hit and miss.

That's certainly consistent with what Ken found when he dissected some, as shown in his pictorial report. I hope readers are getting that there's correlation between and among these measurements.

The 8 ohm L-pad seems to work well for compensating tweeter sensitivity.

I have never seen the anomalous behavior others allege occurs when applying mismatched L-pads like this. At "Max," it's essentially out of the circuit.

[There was a time when I'd have made a huge deal of how wrong it was.... ;) ]

Link to comment
Share on other sites

That's certainly consistent with what Ken found when he dissected some, as shown in his pictorial report. I hope readers are getting that there's correlation between and among these measurements.

I have never seen the anomalous behavior others allege occurs when applying mismatched L-pads like this. At "Max," it's essentially out of the circuit.

[There was a time when I'd have made a huge deal of how wrong it was.... ;) ]

Hi Zilch,

...been following along with interest. Variability among the old tweeters is certainly an issue.

So, are you saying that the presence or absence of the series/parallel resistance characteristics of the old pots have no effect on your measurements? At max the old pots are still providing 16+/- ohms in parallel.

I guess it doesn't matter, as long as you are using the same type of control for all measurements (?).

Roy

Link to comment
Share on other sites

So, are you saying that the presence or absence of the series/parallel resistance characteristics of the old pots have no effect on your measurements? At max the old pots are still providing 16+/- ohms in parallel.

It's an interesting question. If the resonance peak of a HF driver is a significant issue, an inexpensive means of taming it in the "view" of the filter(s) is to put some resistance in parallel. I doubt that was a consideration in these vintage AR designs; they simply wanted adjustable attenuation, and a variable resistance divider gets it done. Dave used a variable series resistance to accomplish that, instead, and with an L-pad, I've shown both in combination with his design here, which I chose to see if it would work and avoid the problematic vintage AR pots.

Measuring an 8-Ohm L-pad right now, at 11:00 there's 16.5 Ohms parallel resistance, 26.5 Ohms at 1:00, 35.5 Ohms at 3:00, and none at 5:00. Let's hook up an AR4x tweeter (#1) and see how the DCR as seen by the crossover varies:

11:00 - 8.4 Ohms

1:00 - 7.0 Ohms

3:00 - 5.5 Ohms

5:00 - 5.0 Ohms

How it varies in the crossover region is more significant, of course, and I can certainly measure that as I previously did with AR3a, if desired, but this is less variability than the filter in the original design experienced, in terms of DCR, that being 9.69 Ohms at 11:00 and 3.81 Ohms at "Max" with this tweeter by my quick calculation.

A look at the actual impedances of the several subject drivers in the crossover region as measured and posted above gives further clues as to how much "normalizing" them might benefit the design:

http://www.classicspeakerpages.net/IP.Boar...ost&id=5773

But the bottom line in terms of response uniformity for any particular one of them is shown in the attenuation curve, where it does not appear to matter more than half a whit or less:

http://www.classicspeakerpages.net/IP.Boar...ost&id=5775

[Whether resonance damping and/or impedance normalization via parallel resistance will be an ingredient the "special sauce" applied by another designer remains for discovery here.... ;) ]

Link to comment
Share on other sites

Thanks for the detailed response, Z. It makes sense to me, especially as it relates to this thread. As you know, I'm not a fan of those original controls either. ;)

OK, so where are you taking us from here? :)

Roy

Link to comment
Share on other sites

OK, so where are you taking us from here? ;)

I WAS going to take the easy path and concede defeat 'til I found the phase/forward axis thing, which I quite frankly don't fully understand as yet. To investigate that further, I'm going to generate the requisite files and load them into PCD for a look, which will likely trigger an alternative design. I'm hoping Dave will run some vertical polars to either confirm or contest my finding(s).

I also listened to both Dave's (DLS) AR4x and the eWave variant shown quite a bit over the weekend, along with another critical listener yesterday, and despite the eWave not yet being optimized, the frequency responses are similar, yet they sound significantly different. The waveguide is the same 8" square 90° x 50° used in SpitWad, so we already basically know how the directivity comparison will shake out. Listening in ZilchLab, directivity alone does not account for the clearly audible difference(s) in performance.

There are matters of substance yet to be discovered here, and that's where I'm going with it, presently. I hope a bunch will come along for the ride. Once I'm done, I'll turn it all over to Ken to see if he wants to take a crack at it.... :)

On Directivity - required reading:

http://www.gedlee.com/downloads/directivity.pdf

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...