Blvdre

I've always liked the look of decorators, and that's a dynamite finish!

I'm really glad you looked into this Josh. I bookmarked that thread in my research of server configs, and since I have a firewire out on my macbook, I wanted to try Phillips suggestion (I've yet to do so, but am inspired now). For those of you who are raising your eyebrows (understandably so), you should go back and read the technical explanation. That may or may not shed light on why this combo sounds so good.

Not an easy feat without some good modeling software and measuring equipment. Yes, too true (many like PCD, and it's free if you already have excel). I would add that the same is necessary for a three way network as well.

...and you can still use passives if you decide on a two-way set-up. They just need to have the appropriate compensation designed into the filter circuit.

I have a pair of those for my 4pi builds. They are regarded as having good bang for the buck, and are popular in the DIY community, as are the BMS CDs.

Spot on Tom. I'm no designer, I don't have experience in taking and interpreting measurements, nor am I any good at designing filter networks. What I do know is after reading a bit here and there about basic design concepts, I have a clearer idea as to what is required for a proper design. Center to center spacing, baffle step, vertical nulls, setting the forward lobe, even power response, proper filter design, etc. It's certainly not as simple as some are lead to believe. Now I wince a bit when someone says, "So I paired so and so together and used this existing crossover design, and this custom box". Not that the result necessarily sounds bad, but it certainly isn't optimized. I think people are underestimating the concept of engineeering a design. With just a bit of research, you can begin to understand what that really entails.

I would recommend only dealing with the original Jolida. Apparently they've won their suit, good for them.

Yes, I got the idea for using a swamping resistor on the autotransformer from ALK. Of course you do know that ALK got the idea from someone else, don't you? g Yes, I would assume he did. I referenced Al's design, because many of us are familiar with his networks. I was trying to point out that the crossover appeared to be a (relatively) constant impedance design, so that moving the taps would not move the crossover point (much). I think Al has addressed the slight variation in impedance (and shifting crossover point) by stating that the movement is negligible because of the shallow slopes involved. All to say that constant impedance/ variable tap design works reasonably well for its intended application.

SF, is that swamping resistor I see? Maybe gleaned from ALK's constant impedance design?

For multiple components soldered to a terminal, the problem is that, if the components were installed properly, there will be a strong mechanical connection to overcome as well. For instance, a cap lead wrapped around a terminal will have been "crimped", or secured, with a pair of needle nose in order to establish a solid mechanical connection before soldering. When removing these components, it's difficult to remove every last bit of solder. Using a sucker and following up with wick helps, but requires longish dwell times which heats up your other parts. My preffered method is to clip the old component out of the circuit (as Speakerfritz already recommeded), leaving just little bit of the lead still secured to the terminal. Heat the terminal and suck out as much solder as possible. Now grab your needlenose and grab on to the end of the remaining lead. Apply heat once more and pull the lead out while the solder is flowing. The dwell times required are short which will minimize the heat applied to the joint and other components.

Ah! You bastard! I got got.

+1 on the dye suggestion. I've built a few drum kits, and lots of singles and found out early on that dye is the way to go for depth and eveness of color.

I think you may be misunderstanding Cask's statements. He's not infering that a ported design has more distortion than an acoustic supsension design (if both designs are properly implemented), but that direct radiators (of which both of those designs are) have inherently more distortion than a comparable horn loaded design.

Superb.

As DJK pointed out, bi-radials do require compensation if being utlized in a two-way, the benefit being even power response.

Good info SF. Those pics remind me of the old Hutron Tracker. That was a handy tool for troubleshooting caps and diodes in circuit.

Don't forget, if you use Al's AP-12 for a two-way, you'll need to add in some contouring if you end up using a CD type horn. Usually just a couple of extra components added to your passive high-end circuit.

I have to disagree here. What a capacitor does is exactly this. Combined with a resistor, they smooth out the signal with a time response. If they do it slower, then the value of the capacitance has changed. So one type of cap with the same value as another type of cap can't be slower or faster than the other. If I had to guess, I'd say that the PIO caps have a higher ESR compared to polys, and attenuate the mids and highs more so.

So you're wondering how the cs1(.5) compares to the standard e-wave, not he deluxe or the high-eff, or the 414 + BMS, etc. (more e-waves than you can shake a stick at). No way of knowing without listening to both. Judging by your posts, I'd say you're leaning toward the e-waves. Nothing wrong with that, many have built them and rave about them. Zilch spent a lot of time developing those x-over networks, tailoring freq and impedances, setting the forward lobe, matching dispersion at the x-over freq. etc.

The originals are very reasonable, and I've heard a great 'bang for the buck'. There are also versions that are more expensive than the 1.5. I guess I'm lost as to what your point is. Do you think the 1.5s are not a good value? Have you compared them to the original econowaves, and what was your opinion?

Wouldn't that be a good thing, and in the spirit of the Econowave? (Zilch would be proud.) The concept of a two-way design with even power response has been around for quite a while.

The PD-5VH is spec'd +-5db from 110-4100. Do your measurements show different? Edit: Nevermind, found the thread.

Yup, there are always PD-5VH drivers being touted as being exactly like the K-55. I e-mailed one seller the difference in specs, he replies "thanks, didn't know that", but didn't change the description, and in fact re-posted with the same "identical to K-55" line. Do your homework.

Good points all around. I believe I ended up using Al's 2nd order crossed at 700Hz, and that worked. I also think the GPA 902s don't reach as low as the original Altecs. Just speculation, but I've heard from others that agree. As far as the high end of the 902, I thought it sounded great up high, better than the k-77 or the Eminence/ Crites tweeters (though those aren't slouches). I did have to apply a series RC to balance the sound. The 902s are a tipped down a bit.

You may be right. I can personally say, though, you do not want to even get close to 500 Hz with the GPA 902. I guarantee you will hear some nasty distortion at pretty moderate levels.