Big problem with cancellation between two waveguides

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I just "finished" building a pair of Speakers using the Dayton 8" waveguide and a BMS 4550 compression driver. One speaker alone sounds good and the directivity might not be perfect above 10 kHz but when I move around my seating position or even the speaker itself a little, playing pink noise, I can hear no dramatic sudden tonal changes. I used pink noise because I experience the same error as I will shortly explain with music and wanted to figure out, what is wrong.

When I play a mono pink noise over both speakers, I experience drastic changes in tone when only moving my head for a few centimeters. Actually it sounds completely out of phase in one position, perfectly in phase a few cms further and keeps alternating between these two.

I also did some rew-measurements in the seating position and with one speaker alone the frequency response stays pretty much the same over my moving range, wich is small, about 60 cms (2 feet). With both speakers and the real time analyzer of rew, i see correct addition and a linear response in one location and heavy dips a few cms further to either side. So basically this matches what I am hearing.

The speakers are playing in Phase, I checked that and using other speakers in my room does not give me that error. Other speakers are Behringer Truth btw. which are known to have a pretty good constant directivity, so the concept is known to work fine here.

I also tried equipping the waveguide with standard dome tweeters and got the same result, so the issue seems to come from the Dayton waveguide. What puzzles me the most is the fact that one speaker alone does not have any strange effects at all.

Does anybody have an idea what might be going on here?
 
Exactly. As Inductor is implying you're most likely experiencing wavelength cancellation when both drivers are energized. Can be mainly due to distance between the drivers ... toe in and perhaps the waveguide not being able to control directivity in part of the range you've decided to use it. So, xo points would be a help here, too. ... or you have a bunch of furniture between the channels?
 
Last edited:
Try uncorrelated pink noise. You probably have destructive mutual radiation causing dips and bad response (tone.) Some times it's better not to use "waveguide" (or wide radiation domes) and use a directional horn. That dayton waveguide probably is radiating all over the place below 5K
 
You don't mention distance between speakers, distance from speakers to your head position.
Distance between speakers is about 2.50 m, distance from each speaker to listener is about 4.5 m.

I tried towing in the speakers a little, didn't help and then a lot, didn't help either.

There are no furnitures in the way.

I use the waveguides from 2.5 kHz upwards. Measurements I could find on the internet indicated pretty much constant directivity between 2 kHz and 10 kHz. I tried listening to pink noise bandpassed between 2 kHz and 7 kHz and it got a little better but didn't dissappear. As I wrote made some quick measurements around my actual listening position and didn't see much change below 7 or 8 kHz.

Indeed it very much sounds like wavelength cancellation, but I don't understand where it comes from.

I didn't save the measurement, because they were really quick and dirty but I can make some new ones the next days.. I would really like to understand what is going on and not just simply start switching waveguides around, which I guess I will end up doing in the end, anyway.
 
Not going further, is also possible that cancellation comes from nulls in the crossover, between say a mid woofer - tweeter. Check phases if not doing active, if active also check phase = 0 and phase alignment vertical position in space between drivers. (I'm not used to active systems).
LCR suggestions to replace Triad Golds (Classic)?
About ceiling height check for hard surfaces (walls and ceiling) and mention the dimension floor/ceiling and/or speaker/ceiling if on a table.
Understanding Studio Monitors: Room Acoustics | Pro Acoustics USA
 
With mono pink noise you are hearing the standard comb filtering that occurs as you move across the exact center line between the two speakers. This is always there to some degree but it tends to be more audible in deader rooms or whrn speaker directivity is increased.

This is a real cancelation in space and would be measurable. However, it stems from both ears hearing both speakers. You can try spreading your speakers much wider (say 120 degrees apart) and the seperation between ears (HRTF) will increase and the effect will greatly diminish.

Replace your speakers with some direct/reflecting omnis and you'll never hear this.

David
 
I see your speakers are relatively close together at 2.5 m apart and 4.5 m away. This compounds the effect I describe above. Try moving them apart or move in closer to the point where you are on a 60 degree triangle and you will find the effect reduced.

Again, if the left ear hears the left speaker and the right speaker together then it hears the interference pattern between them.
 
For checking out the cancellation effect, I had the waveguides play alone, so I can rule out crossover issues. The system is active btw.

I am aware that there is always some combfiltering when playing mono pink noise, but in my case you cannot even listen to music at all. The cancellation is extreme. When moving the head it actually sounds like one of the waveguides would be totally out of phase in one position and switching to perfectly in phase the next.

Using the rta of rew with pink noise, I could see that it all seems to take place somewhere between 7 and 10 kHz. I will be away from home over the weekend but I can make some proper measurements next week. I just don't really have an idea what else to measure, so I will start with measuring slight variations of the angle of one single speaker again and do the same with both speakers, so I can show you the cancellation.
 
What you have found, I have found too. What you need to do is adjust the system where the two speakers don't interfere with each other as much. A better controlled radiation horn, wider spacing, pointing the waveguides so they don't combine as much, a diffuser between them ECT all are options. I always wonder when I see people usinf wide radiation "waveguides" where they direct them to cross in front or at the listening position - Do they not hear this? I do, maybe some people like it?
 
That little 8" conical waveguide is too small to go below where he's crossing it now. 1" drivers typically are good for treble, when you get below 2.5K go with a medium or large format for better performance. Lower limit of 1.2 K for a 1.5" 3" dome driver and 600-700 for a 2" , 4" dome is the way to go if you want it to sound good.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.