Missing params for XT25

[ct991]

Hi there. Long-time listener, first-time caller. I'm wondering what to do about the missing Le parameter on the specs for the Peerless v-line (Vifa) XT25. Since it's not listed in the specs, do I assume it's an insignificant value or some default nominal value or something? I don't have the driver in my possession yet, nor do I have the equipment to measure this on my own. I'm just planning at this point. Has anyone else run into this? Thanks in advance for the help.

[Dennis H]

T/S parameters aren't really too useful for a tweeter. They are mostly used to design an enclosure and, obviously, a tweeter doesn't need one. The frequency response and impedance curves, preferably measured on your baffle, are what you need to design a tweeter crossover.

[ct991]

But don't I need some ballpark figure as a starting point? If I want to put a CR (impedance eq) circuit in there, how do I go about it? ...or do I even need to put one in there? The reason I want to do this is because I've read that it will remove any 'harshness' from the tweeter output.

On a different note, I see that some people recommend an LCR circuit with this tweeter. However, with an Fs of 530Hz and a crosspoint more than 2 octaves above that, is this still necessary?

[ct991]

I don't suppose anyone has actually measured the T/S params for themselves have they? I see on Zaph's website this tweeter is tested (in the Tweeter Mishmash) but the params are not posted, as they are with his other driver tests. Any help would be greatly appreciated.

[ThomasW]

I took my Sencor 'Z'-meter (it's a capacitance/inductance analyzer) and an XT25 measured 06.8 microHenry. That's not as good as a full blown frequency sweep but it's the best I can do.

[ct991]

I'll take it! Thanks so much for the info.

As for my other question: what do people think about using a CR or LCR filter (or both) on this tweeter? Would either (both) be worthwhile with a crosspoint of 2400Hz?

[Dennis H]

As for my other question: what do people think about using a CR or LCR filter (or both) on this tweeter? Would either (both) be worthwhile with a crosspoint of 2400Hz?

I doubt you'll get much help with that here. It sounds like you want to do the old-school-brute-force crossover where you EQ the impedance flat and then do a "textbook" crossover on top of that. If you review the crossovers in the stickies, you'll see none of them use that approach. Rather, they use actual measurements of the drivers along with crossover-design software (Speaker Workshop is a free one) to come up with an optimized crossover. Not trying to discourage you but your approach to the whole problem is flawed.

[ThomasW]

Since you're in Canada, you have a somewhat limited selection of drivers, unless you plan to import from elsewhere. How about some info regarding the rest of your proposed design?

[ct991]

I'm planning a new center channel speaker using Dayton Reference Series drivers from PE. I am considering a WMTMW config (or possibly a WTMW) using RS180's (7") and RS125's (5"), along with the Vifa XT25.

For the crossovers I was thinking of 2nd order APC (2nd order electrical - 3rd or 4th acoustic), with crosspoints at 300Hz and 2400Hz. I was going to put Zobel filters on all three networks as well as an L-pad on the tweeter (and possibly the woofers). Mids and woofers would be wired in parallel. I considered series notch filters but from what I've been reading I don't think they're necessary with this design.

I've tried modelling this crossover network in x-over pro, but I'm sure this is a pretty simplistic way to do it. I just thought that since I'm new at this, I would be happy enough with the results.

This is really the first speaker I've done from scratch, so I've probably made many mistakes, but I welcome any advice from those who are wise in the ways of such things.

[ThomasW]

but I welcome any advice from those who are wise in the ways of such things.

My advise would be to forget about the XT25 since the RS28A is a much better tweeter (I've used both so this is from experience not speculation).

Build one of the sticky speaker designs. The people who developed the crossovers for those are experienced designers and there have been enough builders that all the bugs are ironed out. You'll learn a lot about design by reverse engineering what others have built and have a better speaker to boot.

[jkrutke]

Hello,

There's an LR2 filter @ 2.5kHz for the XT25 in this thread on DiyAudio. The XT25 does need a LRC network across the terminals to flatten the impedance. ( components C13, L14, R15)

If you can massage the RS125 into a good LR2 slope, (difficult) this should work for you on the tweeter section or at least be a good starting point. It will probably work without the delay network also, (components L6, C8, C9 and L10) since this was designed to go with a larger woofer having a deep acoustic center.

The baffle is 9" wide, the tweeter is 3" from the top and offset 3/4".

Those aren't the easiest drivers to work with. My advice: scrap X-over Pro, as it's nearly impossible to get good results with it. (IMHO) Move up to something better, which would be anything - Soundeasy, LspCAD, Speaker Workshop. Measure your own drivers too, don't rely on anyone else's data.

Good luck,
John

[jkrutke]

My advise would be to forget about the XT25 since the RS28A is a much better tweeter (I've used both so this is from experience not speculation).

I hope you don't mind if I disagree with you there. :B I much prefer the XT25. (the off axis rolloff doesn't bother me, I toe the speakers in) Sounds a little more natural to me. I've used both also.

Though I do have to agree, for a first or second design, it's better to build another proven design and learn from it.

[ThomasW]

I hope you don't mind if I disagree with you there. :B

Not a problem....

Properly XO'ed I prefer metal domes. I think the overtones/top-end sound more like real instruments.

[ct991]

Well I really tried to use a different piece of software. I tried SoundEasy and Speaker Workshop. Both programs are somewhat less than intutitive so I haven't gotten much done with them.

Is X-over pro really so bad? It seems to be a fairly intuitive program and has some nice design features. What is it about x-over pro that makes it an inferior design tool?

I'll try my luck with LspCAD and see what I can come up with.

[Dennis H]

Is X-over pro really so bad? It seems to be a fairly intuitive program and has some nice design features. What is it about x-over pro that makes it an inferior design tool?

As far as I know, it only does "textbook" crossovers. You need something that will use imported frequency and impedance curves. If you have Excel, Jeff Bagby's Passive Crossover Designer is nice. You're going to need frequency and impedance curves from somewhere, preferably measured in a box similar to yours, to use any of these tools.