XT25 in a waveguide?

[spasticteapot]

I've been thinking of building a set of speakers as a challenging academic exercise. (Read: I don't know what I'm doing, but all this stuff looks cool so I'm using it.)

I picked up a set of Vifa XT25TG30 tweeters a while back for next to nothing on eBay. While (not unreasonably) known for being a pain in the buttocks to work with, the major issues - including rising distortion at low frequencies and a gnarly impedance peak - are both well-known and solved in documented ways.

However, because I'm a glutton for punishment (and am paying $40 a month for a shop membership I don't use), I'd like to stick it in a waveguide. Rationalizations include:

1. The boost to the low end will do wonders for distortion in a 2-way
2. A waveguide is a tidy solution to the offset problem without using a crazy Zaph-esque delay filter
3. It's a nice way to match dispersion to a ~6.5" woofer
4. It's a tidy way to create a higher-order crossover without buying expensive crossover parts
5. Giant waveguides are bling.

The Canary II design here looks pretty neat with a ~2khz crossover, though it doesn't appear to be compensating for the giant impedance peak and the frequency response is a bit odd - perhaps the waveguide is improperly mated to the driver? I was figuring on using the SB16PFC25-8 for the smooth and extended response, and crossing somewhere between 2-2.3khz.

[ani_101]

Re: XT25 in a waveguide?

I've been thinking of building a set of speakers as a challenging academic exercise. (Read: I don't know what I'm doing, but all this stuff looks cool so I'm using it.)

I picked up a set of Vifa XT25TG30 tweeters a while back for next to nothing on eBay. While (not unreasonably) known for being a pain in the buttocks to work with, the major issues - including rising distortion at low frequencies and a gnarly impedance peak - are both well-known and solved in documented ways.

However, because I'm a glutton for punishment (and am paying $40 a month for a shop membership I don't use), I'd like to stick it in a waveguide. Rationalizations include:

1. The boost to the low end will do wonders for distortion in a 2-way
2. A waveguide is a tidy solution to the offset problem without using a crazy Zaph-esque delay filter
3. It's a nice way to match dispersion to a ~6.5" woofer
4. It's a tidy way to create a higher-order crossover without buying expensive crossover parts
5. Giant waveguides are bling.

The Canary II design here looks pretty neat with a ~2khz crossover, though it doesn't appear to be compensating for the giant impedance peak and the frequency response is a bit odd - perhaps the waveguide is improperly mated to the driver? I was figuring on using the SB16PFC25-8 for the smooth and extended response, and crossing somewhere between 2-2.3khz.

yay, go for it.... That's about all i can contribute now, but i am rooting for you!!!

[ernperkins]

Re: XT25 in a waveguide?

I'm currently working on a system with the XT25TG30 using Dave Pellegrene's 6 1/2 W X 4 1/2 H elliptical waveguides. He tested several waveguides and posted the results at https://plus.google.com/photos/10163...90832534582817. IIRC the 8 inch circular waveguide had the best results. They were too big for the cabinets I wanted to use so I went with the elliptical guides.

[ryanbouma]

Re: XT25 in a waveguide?

The XT25 does excellent in waveguides. Seems most ring radiators do.

[TN Allen]

Re: XT25 in a waveguide?

"...Giant waveguides are bling."

Bling isn't a word I would use, but they are useful and attractive in an unconventional way, and that they can contribute some remarkable, and somewhat under recognized beneficial effects on sound.

[I]"I've been thinking of building a set of speakers as a challenging academic exercise. (Read: I don't know what I'm doing, but all this stuff looks cool so I'm using it.)"[I]

My approach exactly; let me know if you want help with large WGs.

[spasticteapot]

Re: XT25 in a waveguide?

...Giant waveguides are bling.

And there are some wonderful, I think, under recognized benefits to using them.

In fairness, the others are pretty legit. :P

If you can suggest a waveguide profile suitable for mating an XT25 to a driver like the one above, I'm all ears. I'm still trying to wrap my head around crossover design, but it sounds like it'd be a nice solution to match directivity and the acoustic centers of a slightly odd driver. It seems like a fairly simple second-order crossover at 2khz is within the realm of possibility...

[carpenter11235]

Re: XT25 in a waveguide?

"...Giant waveguides are bling."

Bling isn't a word I would use, but they are useful and attractive in an unconventional way, and that they can contribute some remarkable, and somewhat under recognized beneficial effects on sound.

[I]"I've been thinking of building a set of speakers as a challenging academic exercise. (Read: I don't know what I'm doing, but all this stuff looks cool so I'm using it.)"[I]

My approach exactly; let me know if you want help with large WGs.

[ATTACH=CONFIG]57685[/ATTACH]

Those are really interesting. Any measurements to share? Especially off axis...just curious

[Face]

Re: XT25 in a waveguide?

In fairness, the others are pretty legit. :P

If you can suggest a waveguide profile suitable for mating an XT25 to a driver like the one above

Yes, one of Dave P.'s. ;)

[Pete Schumacher]

Re: XT25 in a waveguide?

Yes, one of Dave P.'s. ;)

If acoustic center alignment is a priority, then it would make sense to first measure the differences between the mid and tweeter on a normal baffle, figure out the offset and cut a guide to give just that amount of depth to compensate.

[TN Allen]

Re: XT25 in a waveguide?

Pete,

You've hit upon an idea that opens up an interesting "can of worms". Designing a shape that provides a specific offset can be a significant challenge complicated by driver dimensions and fastening requirements. Additional problems can be encountered if there are two waveguide contours, tweeter and mid/woofer, and they are to blend as tangent surfaces. Complex surface definition pushes the capability of design software.

If in addition to defining complex WG surfaces, simulation of the acoustic effect of the surface is wanted, the process becomes even more complex, and in so far as I have seen, remains a significant challenge for software designers. There are simulation programs based upon Finite Element Analysis algorithms, but I wonder how effective these are for predicting acoustic effects. Testing, and attentive listening after milling and assembly, is probably a more cost effective and reliable approach.

Fortunately, once defined, complex shapes are relatively easy to mill, or build using an additive process such as 3D printing.

[TN Allen]

Re: XT25 in a waveguide?

[quote name="TN Allen" post=2066153]"...Giant waveguides are bling."

Bling isn't a word I would use, but they are useful and attractive in an unconventional way, and that they can contribute some remarkable, and somewhat under recognized beneficial effects on sound.

[I]"I've been thinking of building a set of speakers as a challenging academic exercise. (Read: I don't know what I'm doing, but all this stuff looks cool so I'm using it.)"[I]

My approach exactly; let me know if you want help with large WGs.

[/QUOTE]

Those are really interesting. Any measurements to share? Especially off axis...just curious

These aren't wired yet. I don't really test, although I occasionally try. I've found listening carefully works better for me. Room dynamics and personal listening preferences play a significant part in my design and building choices, although, I recognize test data might be useful to others. After I have these wired, I'll try to test and post results.

[spasticteapot]

Re: XT25 in a waveguide?

These aren't wired yet. I don't really test, although I occasionally try. I've found listening carefully works better for me. Room dynamics and personal listening preferences play a significant part in my design and building choices, although, I recognize test data might be useful to others. After I have these wired, I'll try to test and post results.

What kind of crossover and drivers might be involved for drivers mounted in a waveguide of this type?

[TN Allen]

Re: XT25 in a waveguide?

This is another experiment, so I plan to try a few different drivers. Currently there is an Usher 9950 in one, and the Vifa OC25SC65-04 in the second. The midrange in both cabinets is currently the Peerless 831735, but I may also try the Usher 8836A, the 7" yellow cone Kevlar woofer. This is another possibility: http://www.parts-express.com/tymphan...aker--264-1134 for the midrange, although, this will require some modification of the waveguide throat opening.

The cabinets recently had the 4 Ohm version of these woofers: http://www.parts-express.com/tymphan...aker--264-1142

I'll put these back in the cabinets, but may also try the 8 Ohm version. I doubt the difference will be noticeable.

Regarding the cross over circuit. It also will be a bit of an experiment. Because I want to try various drivers I plan to use at least two amps. I also plan to assemble a midrange circuit that will allow me to switch various coils and capacitors in and out to vary the midrange response. I already have a similar circuit wired that I use for tweeters. Both of these circuits are to be 1st order. I'll connect a few different coils to the woofers to vary their response.

All of this may seem unconventional, it is a little, but will allow me to try various ideas I find interesting, including tone control at the speakers rather than in the preamp. Volume control will be with attenuators between the source and amps.

[Pete Schumacher]

Re: XT25 in a waveguide?

Will there really be any benefit to putting a guide on the mids? I'd love to see how it modifies the response of the mid.

[TN Allen]

Re: XT25 in a waveguide?

Yes, the primary advantage I hear is in clarity. The first few midrange guides I did were small, and the effect was minimal. However, when I built the first larger double guide, which is pictured in the oval waveguide thread, I also built a flat baffle of similar dimensions and shape. There is a significant improvement in the sound from the waveguide baffle. The drivers were the same in both baffles, as was the volume of the enclosure, and I used the same amp, connecting one cabinet to left, the other to right, and used a mono signal.

I suspect the double waveguides reduce crosstalk to some extent, which makes it much easier for me to hear clearly. Of course this may be an advantage for me that others who have more acute hearing might not need.

On the otherhand, I have had the double guide pictured, aka bedpans, out in public quite a bit in various settings for a few months at a time and people, some musicians, have commented positively on the sound they produce.

I can add more anecdotal evidence, and of course it is meaningless without something like data from Holm or Omnimic. Were I not so far away from the midwest, I would offer to bring the cabinets your way so you might decide if there is an advantage. Of course if anyone plans to be in Maine, I will be happy to have them stop by to listen.

One of the reasons I don't make much of an effort to test is that while the effect of a guide on a tweeter shows up well in the data, that of the effect on the midrange is less dramatic at the lower, 500-1000Hz. frequencies. However, the difference in sound is striking. I suspect the midrange guides interact more with room dynamics than do the tweeter guides, measuring this objectivrly, and in some meaningful way is beyond my resources.

If a waveguide has an effect on a tweeter down to 500-1000Hz., would it not also affect a midrange to the same frequencies?