My first wave guide speaker....lots of pics

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Take a look at Daves own plots, at 45 deg the sound is Consistent off axis vs no waveguide.


The bare tweeter off axis falls off in frequency as you increase off axis angle. The WG tweeter doesnt. Thus the dispersion at higher frequencies is improved with the WG. I fail to see why you post that the two terms mean something different. Yes its consistent off axis due to improved high freq dispersion.
 
Im not sure I agree, take this as an example,

http://www.htguide.com/forum/showth...eguide-testing&p=545877&viewfull=1#post545877

In the first measurement we see consistent dispersion up until about 3.5k, then a drop in off axis spl as frequency increases.
In the second measurement we see a consistent off axis response that mirrors the on axis response right from 2k through to the top octave.

The waveguide you are using displays similar characteristics in that it makes the off axis response more consistent, as Badman stated. The fact that it boosts the bottom end and reduces distortion is a given, no argument from me on that point . :)

Every horn is different as is how they react to the driver used. The XT25 has a phase plug which improves how it works with a WG. Posting a Neo3's interaction with a different horn has no relation to how the XT25 works with this WG. The XT25 has better dispersion at higher frequencies than the bare unit, IE there isnt the high end fall off as fast.
 
Yes, though I'd use it all the way around- your waveguides terminate in an angle rather than a fully flat (or rounded/rolled back) surface. While they're better in this respect than many horn/waveguides, there are several that do this better- the SEOS, the OSWGs from both geddes and the DDS variant, the QSC, etc. Of course, you're optimizing for certain factors, which I get- it has to fit, after all, and your polar performance is nothing if not impressive.

Foam or felt allows you to mitigate any diffraction over that lip while still retaining the same flange size. Basically you want to approximate a forward continuation of the profile, with whichever lossy material you use, so it doesn't screw up what's already good, but acts more like a larger roundover. Pic is the JBL Progressive Transition horn/waveguide as used in the Econowave project, with shaped foam extending the profile and killing the hard edge.

The shaped waveguide you show is pretty awesome way of incorporating it into the baffle. Heck of a job! Don't mind the fugly "test mule" style build, or the pine needles on the carpet- the foam is the key!

An externally hosted image should be here but it was not working when we last tested it.


Foaming At The Mouth Article By Jeff Poth

What I have found, with dome tweeters anyway, is that a radius of about 3" seems to work well. Obviously the depth and mouth size has a lot to do with the radius. I've not worked with compression drivers but it seems they are less sensitive to the shape of the guide. For instance A guide like you show with the four corners would not do well with a dome tweeter at least in my experience.
I've kind of moved towards focusing on getting the best controlled directivity and also trying to match the guide depth as close to the acoustical center of the mid/ woofer it is being matched to. Which is usually a 4"-8" woofer. That normally makes the guides 1"-2" deep. So I guess I quite focusing so much on bottom end boost. I'll have to do some testing with the sharper termination or mouth edge. I would think the stepper/ sharper mouth edge would have less baffle edge diffraction but is it going to be any worse then what you get of the sharp edge of the guide. Maybe I'm not thinking about it right. I guess you would get less room interaction from the side walls.
I get what your saying about the foam around the mouth but other then in your own test room I don't see anyone adding it to a speaker just because of the looks.

Dave
 
Trying to find Pellegrenes Google plus page with all his measurements, it shows what Im talking about. I used the neo 3 as an example, not as a rule. I know the phase plug on the xt series and others can help stop cancelation in the top end when used in a waveguide. In any case I know what you are saying and 99% agree, nice project. Im going to use either the ot19 or the Baby xt25 in a guide soon.
Cheers.
 
Trying to find Pellegrenes Google plus page with all his measurements, it shows what Im talking about. I used the neo 3 as an example, not as a rule. I know the phase plug on the xt series and others can help stop cancelation in the top end when used in a waveguide. In any case I know what you are saying and 99% agree, nice project. Im going to use either the ot19 or the Baby xt25 in a guide soon.
Cheers.

Zaph posted all the results with a XT19 with a 6" waveguide. It was just as impressive. I like the lower res freq of the 25.
 
The phase plugs on these ring radiators are why they work better higher in frequency.

Dave, can you make a WG that has a phase plug shaped like the XT25's suspended in the throat by 2 or 3 tiny pins to accomplish the same thing with conventional dome tweeters so you can start with a cleaner unit?

I've spent many hours trying to use phase plugs. All different sizes and shapes. What I found is that it has to be super close to the dome to work well. Problem is if it helps on axis it makes the off axis worse and if it makes the off axis better it makes the on axis worse in most cases. That and it adds a lot of time to making the guides. So I just came to the conclusion that if a dome needs a make shift phase plug I just won't make a guide for that tweeter. Usually if the top end is a mess without a guide the guide will usually make it worse.

Dave
 
https://lh6.googleusercontent.com/-...200%2C30%2060%2090%20degrees%20horizontal.jpg

Take a look at Daves own plots, at 45 deg the sound is Consistent off axis vs no waveguide.

These measurements are 0, 30, 60 and 90 degree measurements. My goal is to have the guide at least match the 60 degree angle of the flat baffle. Usually from 60-90 degrees the flat baffle holds the top end better. Being I am able to reduce the throat size of the guide to 1" on the XT25 I get boost all the way to 12.5kHz. Between that and the guide depth is only 1.5" the guide does quite a bit better. That is what I've been concentrating on with my guides lately. I like to see from on axis to 90 degrees off axis at 10kHz for there not to be more then a 22-25 db spread.

Dave
 
Dave, what happens to the top end when you make the guide deeper, say 2 inches?

The deeper the guide the quicker the top end rolls off off axis. You can pretty much figure when you look at the tweeter off axis in the guide once you can no longer see the dome the top end rolls off quicker. Also on the smaller guides the extra depth will make the walls steeper which will cause more reflections. On an 8" round guide 2" deep is about it with a dome. That is about a 3.9" radius.
I should add IMO.
Dave
 
This is getting more out of line, but as to discussion about wg edges and horns - an example of how Fountek NeoCD3.5H performs. It has a small horn with 1.5cm flat edges. I use it as nude, no frame!

An externally hosted image should be here but it was not working when we last tested it.


My own outdoor measurements. Violet is nearfiel, others are taken at 1.5m distance (5'), 3ms gating. All with 6.8uF cap in series and LR4 xo at 2500Hz set in minidsp. Not bad at all ;) But distortion rises below 2kHz and we must be careful of not destroying the membrane.

a discussion of it at PE forum Fountek NeoCd3.5H Horn Tweeter
datasheet of NeoCD3.5H http://www.fountek.net/uploadfile/1011/25100645.PDF
 

Attachments

  • neocd35h 0 15 30 60 90¤ 3ms nosmo.png
    neocd35h 0 15 30 60 90¤ 3ms nosmo.png
    88.6 KB · Views: 410
  • neocd35h distort nearfield.png
    neocd35h distort nearfield.png
    76.1 KB · Views: 398
Last edited:
The Fountek CD3.5H has internal cap and propably a coil too (impedance dampening) to make it flat on-axis. Basically it is a CD3.0 with a waveguide. If we compare distortion of CD2.0 (Zaph)and 3.5H (PE forum) we can see what the waveguide/horn contributes.
 

Attachments

  • CD35 vs CD2 distortion-vert.jpg
    CD35 vs CD2 distortion-vert.jpg
    123.3 KB · Views: 343
Last edited:
The Fountek CD3.5H has internal cap and propably a coil too (impedance dampening) to make it flat on-axis. Basically it is a CD3.0 with a waveguide. If we compare distortion of CD2.0 (Zaph)and 3.5H (PE forum) we can see what the waveguide/horn contributes.

I'm not seeing any drop in distortion. In fact, the F2 looks worse on the 3.5. On a conventional dome, the WG and cap is a 30db drop at 1k for F2.

I'm a fan of Ribbons but it seems its tough to get a clean one. I had the Aurum 3si and made my own ribbons for it. I just couldnt deal with the sweet spot being 12" wide seated on my couch.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.