Which tweeter? Peerless HDS or the Vifa xt

[SlimLim]

Hi Guys,

I'm looking to make my own set of bookshelf speakers. I'm still inclined to use the revelator, but i'm willing to explore other brands as well. The price difference is also quite considerable. The tweeter will be used with the 6.5" Scanspeak midwoofer revelator.

The Peerless 810921 or the Vifa XT25, which one is a better tweeter? In respect to detail, naturalness, etc. If it makes it easier, i hate metallic and harsh speakers and cant stand sibilance.

Finally, is it better to have a good tweeter with a great midwoofer or a great tweeter with a good midwoofer.

- Nick

[Aslan]

Hi Nick, IMHO its better to go with the XT25..............it seems to find its way into all my mid range offerings these days. Fantastic value for the outlay.

Your second question is a curly one...............you need BOTH................if you do have to compromise it should be at the bass end of the spectrum.............pick the mid/woofer with the flattest response in the 200-2.5K spectrum................midrange vocalisation is critical.

And dont skimp on crossover bits either.

[Jake]

Cant help you myself, but there would be no better place for advice than from Aslan.

I just want to see the pics as you build them, sounds like a great speaker is on the way.

[rob323]

Nah, Andrew has it all wrong (oh, my Elsinores have the HDS tweeter so consider anything I say as totally biassed )

I reckon, if you are going to be listening to the speakers on axis, then go the Vifa and save some money. But if listening off axis, go the HDS.

For more info on the two tweeters, check out http://www.joerasmussen.com/elsinore/elsinore_15.htm

I would not consider either of those tweeters to be "metallic".

You noted that you don't like "sibilance", I too find this irritating and on some recordings, the Elsinores do show this trait. But I think it is more that the tweeter is doing it's job properly and showing up a bad recording because on well recorded stuff, there is absolutely no hint of it what so ever. Yet through on a 90's pop cd and you can hear every tch and sss sound that got spat into the mic.

Either way, you still get a way above average tweeter, or, go for a pair of Hiquphons.

[Pulse-R]

I found the XT25's are fine, but there are a few tricks. If you're running passive, I think it's important to impedance correct the hump at Fs and add the Zobel too. I have noticed a fair variety (to me, 'fair' is 5%) in the measured T/S specs of XT25's. They are a nice tweeter, but not to be overestimated.

If you're in Melbourne I can lend you a pair to play with fi you like

[sfreak]

They are both equally great tweeters. I would say the Peerless over the Vifa myself, but both good at producing the upper frequency range, the Peerless has better distortion figures than the XT25, but in most situations doesn't really mean much. Only way you can find out yourself is to get a pair of each. And end up with 2 pairs of bookshelfs

And I agree in Aslan regarding the in the 'which is more important'. Midrange is critical and you can always live without the bass section. Well. You could, I wouldn't like to though heh.

[SlimLim]

Thanks for the responses guys.

The link is interesting rob323. I've also been reading reviews on zaphaudio.com

Pulse-R, it looks like you know heaps more than last time i was on an audio forum. Are you dabbling in passive crossovers now? And unfortunately i'm not in melb.

Aslan, appreciate the comments. Do you simply like it because of its bang for your buck nature, or do you think it is superior to other tweeters around the $200 mark?

I will document my project. It's actually my first custom passive project ever. It should be an interesting learning curve.

[Pulse-R]

experience is better to have behind you, than over you.

[SlimLim]

I'm curious as to what you designed pulse-r. Care to share?

[Nude]

Still one of my faves

[SlimLim]

The Scan-speak 3004/660000, to me, on paper looks better than the revelator in regards to spl vs frequency, on axis. At 30 and 60 degrees off-axis it doesn't seem that far off the mark. Has anyone had any experience with this, or even done a back to back with the revelator D2905-99000 or the D2904-70000?

[Pulse-R]

a few half-done things so far - bookshelfs using cheap drivers, work ok but cheap drivers aren't the best

half-way into a pair of floor standers... just waiting to make some more time (that takes ages to make, and never lasts long enough)

the main thing I learned is that trying to design crossovers from published specs will, in 99% of cases, not yield acceptable results. you really need to take the time to measure each driver and work out the best solution for each - High pass/Low pass/ etc. - impedance correction is just as important as tailoring the crossover curve to match the drivers' response. woofers - haven't dared to try ported enclosures yet.. maybe one day.

[Aslan]

Hi Slim, I regularly use the XT25, D2905 and R2904.

The XT25 is fairly forgiving of crossover points...............depending on application anywhere from 2200 to 3300Hz...........it doesn't seem to change its character much.

These three are pretty good examples of "you get what you pay for", although the XT25 punches above its price point.

BTW I use the 970000 version of the D2905 as it has a smoother slightly rising response than the other versions..............you lose 2dB of efficiency though.

The R2904 is a weird beast.............a brilliant tweeter but it completely changes character with different crossover points...............I personally love it.

[SlimLim]

Thanks for the input guys. Through the advice, ive taken a step back and reselected my drivers. I've chosen to use the scanspeak D304/6600 and the Peerless HDS 830884. Why 8" you ask? Well i want the bookshelfs to still have some oomph without the use of a sub. I'll be going ported.

I've just started designing my crossovers. I don't have an electronics background, so i'm still getting my head around the design aspects. I've decided i'm gonig to go with a crossover at 2khz, 12dB

slope and Linkwitz-riley type crossover. Does this sound good?

[Pulse-R]

I have found that the best way is to design the box first, and then measure the in-box response to determine the best crossover specs. Then design the crossover - build it with cheap parts to get the values right, then rebuild the crossover with better parts for the final part.

There is quite a bit of difference in driver perfomance in reality compared to on-paper.

Personally, I prefer a crossover point above 3kHz - but acoustic and electrical crossover points are also often different based on in-box response of both woofer and tweeter - partly due to baffle step and phase/time alignment of the drivers.

Good driver choice

[SlimLim]

Thanks pulse-r, at least someone agrees that ive selected good drivers.

Building the enclosure first sounds logical, as yes, the response will be different in an enclosure, and real world sure beats paper theory. However even if you test it with this method first, then wouldn't your environment alter your results too? The test room, the distance you perform it at, your equipment, the angle both width and height of the test location, to me will all play significant roles in altering your results.

I acknowledge your advice with making a mock cheaper crossover, however even if you do that, replacing the inductors, capacitors and other components with the same value, might not yield the same results. I've been looking up through different brands and quality of parts and they vary in resistance slightly from others. Is this enough to alter the calculations? Gut feeling says yes, but i'll let the experts chime in on this one.

The crossover point, i would like to go higher, but the peerless hds seems to get a bit ugly beyond 2k.

Actually you know what yould help me, a step by step process in designing a crossover. eg. impedence match the speakers first, then secondly choose crossover point etc etc

- Nick

[Pulse-R]

a relatively "dead" room has little effect on measurements at 0.5m measurement distance.

you're right on with the step-by-step.

once you get the impedance matched, then you can measure the acoustic response, and determine the best filter design for each driver (may be asymmetric, etc.). Then you put it together - the cheaper inductors aren't too bad for mock-ups, I use 1mm wire and hand-wound for initial tests, then Solen indusctors for "good" version. not too much difference.

[Hugh Dean]

I have offered two commercial speaker kits in five years. The first, the AKSonics, designed by Ron Newbound, use the XT25 and the M18WO-09-08 Vifa. The second, the VSonics, designed by Laurie Menogue, use the Peerless 884 HDS woofer and 921 tweeter.

This is our opinion, and it could be wrong. Andrew and others here have a different opinion, and that's cool. That just shows you need to do more homework, and find the details yourself, and properly 'own' the final choice.

We feel the 921 is much superior to the XT25. It just seems more natural, and clearer. Fs is even lower than the XT25, so it's easy to design with too.

Cheers,

Hugh