There is not too many ways to create a buzzing problem in the mentioned range. Can one rule out the removable baffle vibration as a potential source of the problem? Driver frame and baffle fitting ok?
Once I had a commercial Focal Chorus smaller model on my desk to improve on the crossover and while evaluating the loudspeaker would occasionally buzz loud enough. Turned out a lack of adhesive at one spot made the cabinet vibrate.
Were it because of vertical air resonaces coming out, the proof would show on a raw woofer impedance measurement and near field raw spl measurement. This frequency woud correlate with internal cabinet height. Damping the cavity should remedy it and mesurements would improve.
Once I had a commercial Focal Chorus smaller model on my desk to improve on the crossover and while evaluating the loudspeaker would occasionally buzz loud enough. Turned out a lack of adhesive at one spot made the cabinet vibrate.
Were it because of vertical air resonaces coming out, the proof would show on a raw woofer impedance measurement and near field raw spl measurement. This frequency woud correlate with internal cabinet height. Damping the cavity should remedy it and mesurements would improve.
one question on your enclosure if I may: you used two layers of plywood, correct? What type and how did you attach the one to the other? To me this construction method seems not without it’s flaws, unless the panels were glued firmly together or even better you would have applied a shear absorbing layer in between (CLD).
Man Idk if I feel comfortable doing that, the headphone output for impedance sweeps is already enough to clip the inputs. I don't want to run a crown xli800 output into a motu m4 input. If I kill an input that's all I got.
I played with driver delay and 0.1ms was the only setting that filled in the crossover dip. I was using 0.4ms before as that's what I came up with doing some math, but this shows that would be wrong.
0.15ms seems to sound even better, but I'm not sure if readelay is applying it. There is a definite audible differnce between .1 and .15 but readelay makes the 5 disappear upon closing the plugin, I can only assume it's still applied since the sound doesn't change.
Last edited:
Your thinking is sound. A prudent measuring jig would typically include series resistance, attenuation, DC blocking and so forth.
So you've established that you're on the wanted side of things phase wise. The proper phase isn't always the one that gives the highest summation, although it's a place to start.
I'm going to settle on 0.15ms, it sounds and measures the best to me. I don't even know how one implements that passively, but yes it was not correct before. Wonder how much that hurt the imaging.
I got the woofer level up to what sounds correct to me, and that resonance is definitely still there. Like a sore thumb it just kinda spoils everything. I'd like to cut and cap the port which would let me stuff a good chunk of dampening in the bottom as right now there is really no room.
I got the woofer level up to what sounds correct to me, and that resonance is definitely still there. Like a sore thumb it just kinda spoils everything. I'd like to cut and cap the port which would let me stuff a good chunk of dampening in the bottom as right now there is really no room.
I should probably hold on resonance comments, I'm sitting on axis with the woofer with a desk in front of the speaker.
Allen is it possible to get usable phase measurements, I notice that usually looks like a mess in REW and haven't paid much attention to it before.
Allen is it possible to get usable phase measurements, I notice that usually looks like a mess in REW and haven't paid much attention to it before.
Are you asking about being able to keep time between multiple measurements or are you asking if REW can handle measuring something which is non-minimum phase?
REW takes phase as it comes, which includes delay and diffraction artefacts. Unless you do something drastic like forcing it to read your measurement as minimum phase, or re-setting the starting time or if you were to screenshot the response and read it into a minimum phase assuming response reader, then you should be fine.
REW takes phase as it comes, which includes delay and diffraction artefacts. Unless you do something drastic like forcing it to read your measurement as minimum phase, or re-setting the starting time or if you were to screenshot the response and read it into a minimum phase assuming response reader, then you should be fine.
It might or might not be there’s more than one flaw to consider. The only big speaker (160l or about 5 cu ft) I ever built from 3/4”-ish material needed extensive bracing as in every 8” or 20cm. In both dimensions that is. That construction got it pretty solid though.
The most likely is an enclosure (TL-like) resonance though. The port giving it some mass loading, MJ King would have figured it out already, I think.
Well this is embarrassing. Apparently this little sheet of polyfill is all I had in the mid range enclosure. Place bets now on whether you think this was at all adequate lol.
Not sure what the ideal material would be here, I've got some memory foam I might try, local place has some denim insulation I might go get.
Not sure what the ideal material would be here, I've got some memory foam I might try, local place has some denim insulation I might go get.
Well, from the example i have ( my 3 ways are commercial units i modified filtering from passive to dsp+multiamp) it seems very few absorbing material for mid.
In mine the whole mid box is stuffed. Not overpacked but in volume it is much more than what you have. I would not say it is mandatory as preference comes into play and some may prefer less stuffing but imo it is worth an experiment with much more volume than you have atm.
If you can afford denim insulation try it, for sealed mids it is perfect material.
In mine the whole mid box is stuffed. Not overpacked but in volume it is much more than what you have. I would not say it is mandatory as preference comes into play and some may prefer less stuffing but imo it is worth an experiment with much more volume than you have atm.
If you can afford denim insulation try it, for sealed mids it is perfect material.
I put some other stuff in there and seemed to have fixed a ~1khz resonance showing on the impedance plot and FR.
So as for the lower resonance port, the only thing that has actually had an effect there was wrapping the port in rubber. Plugging it doesn't change anything. Someone mentioned it could be the woofer reflecting off the port? I'm trying everything I can before cutting a port as there isn't really any going back on that one.
So as for the lower resonance port, the only thing that has actually had an effect there was wrapping the port in rubber. Plugging it doesn't change anything. Someone mentioned it could be the woofer reflecting off the port? I'm trying everything I can before cutting a port as there isn't really any going back on that one.
Thats interesting.
Wrapping the port in rubber apparently have some damping effect to resonance.
But if plugging the port dont change anything it seems like resonance is escaping out in some other way, maybe thru woofers cone.
Its hard to make any conclusions, my guess is some weird structural issue/interactions between port, box and woofer.
Woofer reflecting of the port? In that case I doubt wrap of rubber would have such an effect, you would need some good acoustic barrier in between like layer of thick felt.
Instead of cutting, cant you just remove the port and cover the hole with piece of solid ply or similar?
Wrapping the port in rubber apparently have some damping effect to resonance.
But if plugging the port dont change anything it seems like resonance is escaping out in some other way, maybe thru woofers cone.
Its hard to make any conclusions, my guess is some weird structural issue/interactions between port, box and woofer.
Woofer reflecting of the port? In that case I doubt wrap of rubber would have such an effect, you would need some good acoustic barrier in between like layer of thick felt.
Instead of cutting, cant you just remove the port and cover the hole with piece of solid ply or similar?
Since the port is at the bottom, where the pressure variation is at it’s biggest on any vertical (or horizontal, it’s in the corner IIRC) mode, I can imagine the port itself being a mass-spring resonator actuated by the pressure variation. In some more complex way than I describe, that is.
Applying the rubber changes the mass and spring. And moving it out of the danger zone of the (first) mode. But glueing some weight of any kind to it would help too.
Applying the rubber changes the mass and spring. And moving it out of the danger zone of the (first) mode. But glueing some weight of any kind to it would help too.
You could try to fix both ends…or leave the rubber in situ. But Idk you‘re into any modding anymore.
I do some stuff here and there but ultimately just band aids on a what I'd consider a flawed design. You fix the low resonance, you still have a box with pretty poor dispersion which I do value a lot. I'm likely to just build some subs to accompany a speaker with good dispersion that can also get decently loud, like the VBS 10.2
I cut one port and capped it off, impedance looks the same with a spike at the same point. It appears some aspect of this cabinet is doomed. Not much else I can do at this point other than start over.
I'd like to try a rebuild soon, and so far my check list for fixes is as follows, but having some doubts that this style of speaker will satisfy my needs.
bass problems - Make the cabinet shorter to raise the height dimension resonance to a more treatable frequency, the OSMC seems like a good size there.
port - put a port on the back in the middle
the highs - the nd25fw sounded better than the flat flange domes I tried, other than pattern control I'm thinking that the waveguide loading helped minimize baffle interactions. I'd like to use the wg300 and adapter + a dx25 or xt25. The nd25fw only goes down to 2500hz, I'm assuming using a tweeter that can go lower will provide baffle effects reduction further down.
mid range - not sure about this one. minimizing baffle interactions seems tougher. something like the atc dome or bolt are loaded in a guide that I assume helps in similar way to a tweeter waveguide but those are not feasible for me to afford. I've toyed with the idea of just using a new dayton signature 6 inch woofer (response looks impressive).
I'd argue I value pattern control, even dispersion sounds best to me, even some little c notes provided what I'd consider much more coherent sound than these big speakers. this makes me question if this type of speaker is the right choice and if sunk cost fallacy keeps me locked in to using some parts like the sda315. looking around, something like the vbs 10.2 by Matt Grant seems to offer some impressive dispersion at a price I can afford, I'm not sure I can get a 3 osmc like speaker 3 way to behave on par with that. I don't have much experience with CD and horn in the home though. my goals are really a neutral speaker with good dispersion that can get fairly loud and not lack composure at lower levels.
would love to hear you thoughts.
bass problems - Make the cabinet shorter to raise the height dimension resonance to a more treatable frequency, the OSMC seems like a good size there.
port - put a port on the back in the middle
the highs - the nd25fw sounded better than the flat flange domes I tried, other than pattern control I'm thinking that the waveguide loading helped minimize baffle interactions. I'd like to use the wg300 and adapter + a dx25 or xt25. The nd25fw only goes down to 2500hz, I'm assuming using a tweeter that can go lower will provide baffle effects reduction further down.
mid range - not sure about this one. minimizing baffle interactions seems tougher. something like the atc dome or bolt are loaded in a guide that I assume helps in similar way to a tweeter waveguide but those are not feasible for me to afford. I've toyed with the idea of just using a new dayton signature 6 inch woofer (response looks impressive).
I'd argue I value pattern control, even dispersion sounds best to me, even some little c notes provided what I'd consider much more coherent sound than these big speakers. this makes me question if this type of speaker is the right choice and if sunk cost fallacy keeps me locked in to using some parts like the sda315. looking around, something like the vbs 10.2 by Matt Grant seems to offer some impressive dispersion at a price I can afford, I'm not sure I can get a 3 osmc like speaker 3 way to behave on par with that. I don't have much experience with CD and horn in the home though. my goals are really a neutral speaker with good dispersion that can get fairly loud and not lack composure at lower levels.
would love to hear you thoughts.
It's great you're going for a rebuild (otherwise this thread would have been big blue-balls) 😱
Two thoughts:
1. share your cabinet and bracing plans for feedback before building
2. pick higher quality mid and tweeter; the new Dayton Signature range's low power rating gives me some pause. Morel CAM 558 from Parts Express is a cheaper dome. Folks have said that Faital 5FE120 is good. Maybe Wavecor WF152BD06, or even Seas ER15RLY are worth looking into.
Two thoughts:
1. share your cabinet and bracing plans for feedback before building
2. pick higher quality mid and tweeter; the new Dayton Signature range's low power rating gives me some pause. Morel CAM 558 from Parts Express is a cheaper dome. Folks have said that Faital 5FE120 is good. Maybe Wavecor WF152BD06, or even Seas ER15RLY are worth looking into.
Last edited: