Scan-Speak tweeter upgrade options?

[gfinlayson]

I had a large pair of two way ported floorstanders based on Troels Gravesen's SP95/98. I used the 18W8531G00 mid bass and a D2905/930000 tweeter. I recently moved in with my G/F and needed a smaller caibinet (higher WAF) which was suitable for closer to wall mounting than the rear ported cabinet. I also wanted to use the same drivers and crossover to minimise cost. I considered various options but in the end I opted to go for a transmission line cabinet very similar to Tony Gee's Optimo but with a shorter line length and a slightly narrower throat.

The result was quite surprising. Bass balance is perfect for the room - fast, tight, dry and not overpowering. The best bit though is the vastly improved soundstaging, imaging and detail in the midrange.

Anyway back to my point - the 930000 tweeter now seems a bit lacking. Would the D2905/970000 be much of an improvement, or maybe the D2904/980000 (with a slightly modified crossover), or should I be looking at something a bit more high end (R2904/700009 ? ) Ideally I'd like to stick with a 104 mm diameter Scan-Speak tweeter to avoid having to butcher the cabinet too much.....

[Dansk]

You are going to have to modify the filter no matter what replacement you choose - but perhaps something as modest as the Vifa XT-tweeter could make an improvement...

[gfinlayson]

Are the Scan-Speak tweeters that overpriced???!!!! The XT25 costs less than half the price of the 930000.

[tinitus]

Maybe you could just try adjustments on your 930000

What does "lacking" mean?

Maybe you have altered midrange response by moving drivers to new baffle

[rabbitz]

Originally posted by gfinlayson
The XT25 costs less than half the price of the 930000.

..... and you spend more than the other half of the price on components to tame the XT25 below the crossover point as it gets very nasty. If the ZD5 was still listed you'd see what Zaph had to do. The design I believe was pulled due to some problems with batches of this driver.

There are several other cheaper options and the Peerless 810921 is one of the better but of course would need a crossover redesign.

[jerome69]

Originally posted by tinitus
Maybe you could just try adjustments on your 930000

What does "lacking" mean?

Maybe you have altered midrange response by moving drivers to new baffle

+1

Your should only redesign the crossover.

[TheD]

Originally posted by rabbitz
If the ZD5 was still listed you'd see what Zaph had to do.

That had very little to do with the driver and far more to do with the fact he built a delay network.

[rabbitz]

I'm referring to the LCR which is not required on most tweeters as they don't exhibit the bad impedance spike at Fs which is important when using it in a 2-way. Without the LCR it's more difficult to get it to behave in the crossover so you don't get the garbage. I can't remember his values but in a 2-way with an 18W8531G00 I did, the cap alone in the LCR was 100uF and a good MKP fit for the purpose is worth A$59 for one. Add a good inductor and it adds up. That's for the final values used and does not allow for the tweaking components to get there. I'd rather spend that on a better driver that's well behaved in the first place.

I've often wondered if he would have used a ladder delay with a different tweeter as the ZRT doesn't use one but I think he just wanted to use one in a design.

What I'm trying to get across is the 18W8531G00 is a high quality driver and should be partnered with something equivalent and not a piece of plastic from China.

BTW... I was a DST / Tymphany reseller for years so have sold loads of the XT25TG (heard every one in my test rig) as well as played with it for a couple of years so am quite familiar with it's behaviour. I've watched it go from a Danish made driver to a Chinese made driver and there are differences in quality but at least the price has dropped dramatically from a few years ago and now is very cheap (about the same trade price as the D27TG-35-06). Use it much higher in a 3-way and the problem goes away and doesn't need the LCR.

I know there are masses that love it and also without the LCR and that's fine and dandy but didn't for me.

[benchtester]

Originally posted by TheD



That had very little to do with the driver and far more to do with the fact he built a delay network.

Although, on the tweeter side, he used a notch to reduce resonance since the XT25 doesn't have ferro-fluid. It looks like the 9300 does have ferro-fluid so it is probably not the problem here.

edit: On rereading, this is the same point Rabbitz was making in his first two sentences. Additionally, Zaph used a similar (same?) notch on the ZRT's ScanSpeak 6600 for the same reasons. See Crossover section.

http://www.zaphaudio.com/ZRT.html

[benchtester]

Originally posted by rabbitz

I've often wondered if he would have used a ladder delay with a different tweeter as the ZRT doesn't use one but I think he just wanted to use one in a design.

The ZD5 is LR2 and the ZRT is a LR4. With the ZRT, he reduced the crossover slope on the woofer to help align the listening axis. With a LR2 there is less room for tweeking the slopes. He went with a ladder-delay. (I used a waveguide to recess the tweeter and often people use a sloped baffle.) For shallow crossovers, I think most all tweeters will require some kind of time alignment.