I’m having a hard time getting my head around the final baffle losses of dual opposed drivers on a three way speaker that is shallow, ie narrow baffle as seen by the side firing drivers. I’m assuming it’s a complicated relationship between standard baffle losses and driver directivity, room gain and potentially each driver’s waves combining at the front and rear and adding for higher efficiency at some point? That last point is the one I’m particularly unsure of.
The plan is a wide baffle, shallow three way that may be suitable to table top and near wall positioning. It uses the GRS Neo 3 and 10 which I already have sitting on a shelf for more than a year and Dayton ND105’s with matching PR’s. I took some inspiration from the Lyngdorf Model S as far as form factor goes.
I might being trying to force something that just isn’t worth it and won’t work that well, but I’m trying to do something different and interesting with the GRS drivers that I now have laying around since I scrapped the project I bought them for. For one, the combined SD of the woofers will only be about 102 cm, sensitivity 88db on a really good day and I’d be crossing way higher than normally advised for side mounted. I’d love any feedback one way or the other. Thanks in advance for any insight!
The plan is a wide baffle, shallow three way that may be suitable to table top and near wall positioning. It uses the GRS Neo 3 and 10 which I already have sitting on a shelf for more than a year and Dayton ND105’s with matching PR’s. I took some inspiration from the Lyngdorf Model S as far as form factor goes.
I might being trying to force something that just isn’t worth it and won’t work that well, but I’m trying to do something different and interesting with the GRS drivers that I now have laying around since I scrapped the project I bought them for. For one, the combined SD of the woofers will only be about 102 cm, sensitivity 88db on a really good day and I’d be crossing way higher than normally advised for side mounted. I’d love any feedback one way or the other. Thanks in advance for any insight!
Imagine cutting the box in half - between the woofer baffles. Place one half against a wall. You'll get bass without a step and some midrange will be subjected to comb filtering.
Ok but why? Is the bass without a step do to the wall/ table placement or is that in free space? And what causes the comb filtering and what determines the frequencies?
My simple thinking.
Number of drivers will not affect baffle step on the same sized baffle. Sensitivity gains are all that you change (+/-3dB if in series or +/+3dB if wired in parallel).
The biggest issue I have is directivity with these woofers at 90 degrees. Are you expecting a smooth transition to the GRS NEO10? What is the recommended minimum crossover point and slope for the Neo10 and is this below the point where your 90 degree woofer response is really dropping off?
Number of drivers will not affect baffle step on the same sized baffle. Sensitivity gains are all that you change (+/-3dB if in series or +/+3dB if wired in parallel).
The biggest issue I have is directivity with these woofers at 90 degrees. Are you expecting a smooth transition to the GRS NEO10? What is the recommended minimum crossover point and slope for the Neo10 and is this below the point where your 90 degree woofer response is really dropping off?
With the cabinet mirrored due to the symmetrical bass arrangement, my description shows exactly what each side acts like. It's up to you to imagine that in free space, so for bass this would mean omnidirectional.
The lower mids would have comb filter potential from sound that goes backward and bounces of the imaginary wall. The reality is simply that it's coming from the back.
The lower mids would have comb filter potential from sound that goes backward and bounces of the imaginary wall. The reality is simply that it's coming from the back.
So two baffles still acts like one?
Yes, it’s an issue I have as well, but I have a feeling that in practice it’s less of an issue than it seems. I’m thinking that the directivity of small woofers, shallow slopes, room effects and the drivers mounted near each other on the same plane will work ok. It seems to work just fine on the Kef Blade, they even crossed higher on the second iteration. Measurements look perfect out to 45 degrees or so. I may have to account for some loss of output in the 200-400 range in the XO. FS is around 200Hz for the NEO10, so a 2nd order crossover above 400Hz seems reasonable. I’ve only measured them on a narrow open baffle, but if I remember correctly distortion increased rapidly under 200 Hz.
Ok I got it now Allen. The comb filtering only applies to the mids coming from the woofers correct? I thought you meant the front firing drivers.
Do I see +6db sensitivity and no step in the bass or does it act more like a .5 situation after the crossover is applied? So do two 82db drivers behave like a single driver below baffle step without a step in the bass or two wired in // without a step, ie. 88db below baffle step? That would sound like a free lunch, but I’ll take it if you say so!
I lived for three years with Allison Four speakers, they were my first speakers, and I selected them between others...I mean, nowadays I would dismiss them at first sight, but nonetheless they were superb speakers.
Now, you...ha ha ha (jocking, sorry...)
Just wondering why they are not upfiring.
But, yes, design & vision of the whole and experiment is a fact.
Now, you...ha ha ha (jocking, sorry...)
Just wondering why they are not upfiring.
But, yes, design & vision of the whole and experiment is a fact.
Since woofers were the topic, I mentioned the below normal baffle step region and the transition baffle step region. As far as the above normal baffle step region, the highs will radiate more like a dipole, except they would also be in phase with the acoustic centre half way between them.
You will see +6dB and no step with just woofers on the back.. ideally. The exception will be discrepancies due to bass room modes and midrange comb filtering. It should go without saying that if you add this rear radiation, you should take care to ensure it doesn't run in to it's own troubles.
A .5 is primarily the same as a woofer on the back, only the lesser details vary.
That’s an idea, but I hadn’t considered upfiring. I’m not shooting for a diffuse Omni type of sound here. I’m actually counting on some reduced vertical dispersion due to the long planar drivers. I’m hoping that results in less comb filtering from a potential table top set up
When you say rear radiation, are you talking about the side firing woofers and the opposing woofers are rear firing to each other?
I was really referring to anything. Ideally we'd keep it all between the walls but that's usually not an option. Often you'll see simpler results away from walls or flush with them, but less so in between. Not that you can really avoid it, I'm just pointing out the issue I'm referring to, at whatever frequency and whatever angles apply with your speaker.
Yes. What is teh XO to the midrange. It should be below the point at which the woofer output is omnidirectional. Say 450-500 Hz.
dave
Awesome Dave! Thank you! I understood what Allen said eventually, but this is still quite helpful. And btw, he said exactly that. What does the graph key say, I can’t quite read it? It looks like it says “On wall, X mm to wall”. If so can you explain that because I don’t get it.
Can you tell what magnitude of bipolar dip I would get based on my design? The enclosure is roughly 12”w by 6”d, so about 2:1.
Can you tell what magnitude of bipolar dip I would get based on my design? The enclosure is roughly 12”w by 6”d, so about 2:1.
About 400 Hz. I assume XO slope is also important. I’d think that a steep slope for the woofers would limit issues, but my research mostly points to the opposite being the case.
Nearly every answer to the question of XO point of side firing woofer says under 200 Hz, with many people saying as low as 80. I’m postulating that it doesn’t need to be nearly that low, but I don’t know for sure. Can you explain why you say under 450-500? Pure Omni behavior probably ends around 200 Hz.
That the graphs are the same for a monopole speaker on the wall or a bipole out in the room that is twice as deep. In the first case the wall is acting as a second source, in the bipole another woofer.
Allenposted this a couple times.
The behaviour of this arrangement is more important, to my mind, for the active reaction force cancelation, no BS is just a bonus.
dave
To me the baffle step benefit is at least as important. Any small driver with decent extension in a small box won’t be very sensitive, so every bit helps. And small drivers shouldn’t cause nearly as much vibration as a larger driver might. Maybe all those PR’s will cause some heavy vibrations? Who knows
Normally you'd try to identify those problems. Assuming you've also chosen the crossover frequency for some reason, that defines the steepness needed. Otherwise we're all guessing.
It's typically preferable to use the most shallow slope that ticks all the boxes, but no shallower.
In this case of course, your main concern is the radiation of the multiple drivers and also the matching of the upper and lower ranges. Otherwise the 400Hz region give or take is quite an easy region for woofers and mids, eg no breakup or lower resonance to navigate around.