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.
No mention of what crossover is being used. Maybe both drivers are playing at the same time in the same frequency range?
Crossover is a Neville-Thielle generated by Acourate and run in brutefir. There is some correction applied to make the speakers flat in the nearfield. I applied no further room-correction at this point to make sure I don't get any strange effect from that. One speaker alone sounds as it should. The cancellation effect is definitevily not related with the crossover since the effect is present even when running the waveguides alone. The effect remains when I swap the compression drivers with standard dome tweeters run through the waveguide. The effect is NOT THERE when I use my Behringer Truth monitors at the exact same location. The Behringer as measured by Dr. Geddes show pretty much the same polar response (above 2-3 kHz) as the Dayton waveuide measured by Zilch (I think it was).

Comb filtering from the combination of two distant sources, exactly what is the problem? Only way to get rid of it is to buy a pair of earphones i guess :)

BR,
Anders

When playing mono pink noise through both speakers, the effect you are hearing is normal. Dave and others are correct, you need to move closer to an equilateral listening triangle.

So, you have just revealed the intrinsic fault of 2-ch stereo ;)

Some old threads had addressed this issue. Try searching "3-ch linear matrix", "flooder" ... etc.

I know the effect of practically any speaker-pair when playing mono pink noise through them. That is exactly the problem. What I hear here (actually there, I am not at home ;) ) sounds like that effect pushed to perfection. At one head-position I get perfect coupling of the two speakers and the sound comes from a tiny spot in the middle and a little move of my head to the side results in a sound like someone flipped a switch to invert the phase of one speaker. I never heard that effect SO strong, it is actually not possible to listen to music with it. That is what is puzzling me so much...
 
Just a small suggestion, perhaps take them outside to see if it is a room interaction issue. Could be early sidewall reflections making two 2-source interference patterns. These threads where the asker doesn't really give context always seem to devolve into spirited guessing games....
 
Taking them outside might be a good idea. That would have to wait a while, though, because I live in the middle of the city.

The side walls are treated with absorbing foam on the left and right wall. It's not really a lot but it improved things quite a bit when i installed it. It is Bastoect mats of 1/2 m height, 2 m length and 7 cm thick. The rest of the room is untreated. The ceilings are nearly 4 m high, so there should be no disturbing reflections. The floor is wood, no carpet.

All in all not a very good room to listen to music, but once I threw out my old B&W speakers and installed the Behringers with their directivity most of the issues I have had until then disappeared. Then I decided to build the speakers I am fighting with right now.
 
Normal comb filtering. Learn to live with it. All stereo speaker pairs will do the same. It's also why 99% of cables sound "different" Why We Believe

It is practically not possible to listen to music with the speakers and not all speakers do it in my room, I do have other speakers that play just fine in the same location. As I wrote:

I know the effect of practically any speaker-pair when playing mono pink noise through them. That is exactly the problem. What I hear here (actually there, I am not at home ;) ) sounds like that effect pushed to perfection. At one head-position I get perfect coupling of the two speakers and the sound comes from a tiny spot in the middle and a little move of my head to the side results in a sound like someone flipped a switch to invert the phase of one speaker. I never heard that effect SO strong, it is actually not possible to listen to music with it. That is what is puzzling me so much...
 
Dayton 8" waveguide.
 

Attachments

  • Dayton 8 waveguide.jpg
    Dayton 8 waveguide.jpg
    105.2 KB · Views: 159
No, no other waveguide tried, yet. Actually I am rather sure the waveguide is causing trouble, since other speakers work very nice in my setup, but I'd really like to understand what is going on here. The Donovo Waveguides here look really interresting. Getting them to Germany is a bit difficult, since the shop doesn't ship to Europe: http://www.diysoundgroup.com/waveguides/other-waveguides.html
 
I actually did some quick and dirty measurements, but didn't keep them. I just quickly made them to check polarity and how a single speaker changes under the (small) different angles at my listening position. Polarity was fine. Changing the polarity of one speaker made things even worse so the measurement was correct. I will make some better set up measuremnts and post them next week.
 
This is not a speaker problem other than the fact that higher directivity tends to exagerate the effect. It is standard interference between two sources. It is worse in your case because you have a relatively narrow angle between your speakers. (2.5 m seperation and 4.5m distance away)

Have you tried a wider spacing and closer listening position (as several have suggested multiple times)?

As an experiment toe the speakers to the point of the nearly facing each other. Walk up between the speakers to the point that you are wearing them like headphones. Put on mono music and sway left and right. You can find the mid point by the position where the music is centered in your head. But the comb filtering will be totally gone! This is beacuse at these wide angles you actually have very good seperation between your left and right ears. Each ear hears one speaker only, so no comb filtering.

Conversely, if you want to hear worst case comb filtering, then place your speakers side by side and walk across in front of them.

This is your problem. Nothing else.

David
 
Ok, but if that is the case, why is the problem much stronger with the recently built speakers that with the Behringer, when they have pretty much the same radiation pattern? We are not speaking about a subtile difference, we are speaking about: Behringer - totally fine to listen to, even when moving around - and - new built speakers - cannot be used to listen to music without going nuts after 1 minute.

I can try a wider spacing, but the speakers will end up being very close to the sidewalls.

I did toe the speakers in to a point where they nearly faced each other and the problem got milder, but actually remained.

I am building speakers for over 10 years now, I do play around with car hifi quite a lot, I have built quite a lot of speakers with fullrange drivers, which beam like hell and I am well aware of the comb-filtering-effect of two speakers. I would not open up a thread asking for help if it was anything I ever came across. This issue here has a quality of its own, I never came across anything like it. It renders my hifi-setup totally unusuable and I never had that happen to me ever since I started playing around with building speakers of my own.

It's not a normal phenomenon you just "have to live with". It is not possible to enjoy music with these speakers. Something is really wrong here and I just can't figure it out. Perhaps it will have a simple explanation and hopefully a simple solution but at the moment I just cannot see it.

It takes quite a while before I open up a post in a forum, but in this case, I just cannot figure it out. I tried quite a lot of things over the past weeks, believe me.

Let me get home and make some measurements and hopefully we will find what is causing this. I have heard (and built) so many speaker setups in my life and this is something that I would hardly believe if I didn't have it sitting in my home. This is not a newbie-call, otherwise I woudln't have chosen this forum.
 
So, some more questions:

1. Do you have to stick with such positioning (2.5m separation / 4.5m distance) and the room acoustics ? (I guess it's probably heavily damped?)

2. Do you have to stick with the overall design of this speaker?

In addition to a wider seperation, I'd suggest flip the speakers and make them upfiring. Then EQ the response to your liking and compare. (you don't have to break anything in this trial, you know)

As you mentioned, and same as what I was thinking, your new speaker and setup let the intrinsic fault of 2-ch stereo perfectly presented. If this is the case, I'd think 2 solutions:

1. Adjust it to a condition that is not so 'perfect' as a 2-ch stereo.
2. Give up the conventional 2-ch stereo and find some other way.
 
So, some more questions:

1. Do you have to stick with such positioning (2.5m separation / 4.5m distance) and the room acoustics ? (I guess it's probably heavily damped?)

The side walls are treated with absorbing foam on the left and right wall. It's not really a lot but it improved things quite a bit when i installed it. It is Bastoect mats of 1/2 m height, 2 m length and 7 cm thick. The rest of the room is untreated. The ceilings are nearly 4 m high, so there should be no disturbing reflections. The floor is wood, no carpet.

2. Do you have to stick with the overall design of this speaker?

In addition to a wider seperation, I'd suggest flip the speakers and make them upfiring. Then EQ the response to your liking and compare. (you don't have to break anything in this trial, you know)

As you mentioned, and same as what I was thinking, your new speaker and setup let the intrinsic fault of 2-ch stereo perfectly presented. If this is the case, I'd think 2 solutions:

1. Adjust it to a condition that is not so 'perfect' as a 2-ch stereo.
2. Give up the conventional 2-ch stereo and find some other way.

Yes, I want to stick with the overall design, but I might have to find another waveguide. Sadly this will perhaps lead to having to rebuild the cabinets.

I will try what happens when setting them up wider apart as soon as I am home and find some time.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.