Yeah, if you aren't using the speakers for sound reinforcement, the output outside the pattern doesn't make too much difference.
If you are trying to add some 300-600Hz that you aren't hearing enough of in a microphone and a +6dB speaker side lobe is picked up by it, it does matter as a direct reduction in gain before feedback.
Hi,Here is my one and only mutli-band compare so far. 3-way on left; 4-way on right. I don't know what i'm looking at !!!
REW has a very robust looking multi-tone section in its Signal Generator....ready to explore that.
yeah I guess it doesn't matter which program you use in this case
The JC tests made sense to me like so: noise floor rises as SPL goes up because the driver/system generates more distortion products. From the multitone test it seems rather easy to look where the dynamic range is enough and at what SPL level it starts to deteriorate and at what frequency range it goes first. Perhaps giving away some info where the distortion originates from. On your case I'd look around at the low crossover of the compression driver or perhaps above, if there are hints of less dynamic range at some SPL which would perhaps be at least part of the sound difference you hear between the 3 and 4 way setups. Dynamic range would be height of the peaks from noise floor between.
I wonder if you was able to hear at what dynamic range the performance starts to go away (40db/1% or something else), or if the sound is clearly different already at low SPL level if it shows up in this kind of measurement etc. Many interesting things. Perhaps the back radiation of the mid drivers makes the difference?
I don't knowWell, anyway I guess it doesn't matter how you measure as long as it makes sense to you. Keep posting!
Interestingly, with both sine waves in that 500 to 900Hz range, IMD looks very similar for both the small mids and CD. Not tons of fuzz for either.Since both can cover the range of 500 to 900Hz, using two test tones both in that range would give a better idea of the “fuzz” each creates
The biggest change i see in IMD comes in the 2-4kHz range, when the CD is asked to pull duty down to 500Hz, and F1 is under 700Hz. Like in previous RTA's.
It's kinda like bottom end excursion screws up top end response... Or at least that's my thinking...
Hi Steffen, not sure there's any problem other than getting the CD and mids to fit into a corner to begin with.
I wish i could figure out how without having to build a room's corner around them (for a 90 deg MEH).
Would be cool to have the room corner be the horns walls
They can always be enclosed like i did on syn9 v2. However, that was much easier to do on a 60 deg H, than would be on a 90 deg.
Only real complaint i have with the open back mids is lack of weather proofing for outdoors.
Hi and thx,Hi,
I wonder if you was able to hear at what dynamic range the performance starts to go away (40db/1% or something else), or if the sound is clearly different already at low SPL level if it shows up in this kind of measurement etc. Many interesting things. Perhaps the back radiation of the mid drivers makes the difference?
Well, anyway I guess it doesn't matter how you measure as long as it makes sense to you. Keep posting!
The clarity improvement has been noticeable at about any SPL. I really don't know if the improvement is due solely from reducing the low end load on the CD, or maybe also from not having any back radiation from the mids. These 90 deg guys would be a bit of a bitch to enclose to see.
Yep, i can see that could be an issue for live.
Here's some pink noise SPL-LeqZ measurements...unfortunately only indoor at around 1m.
Entire 5-way: 90dB
Mid section only, 300-700Hz: 75dB
Mid section only, but off to side: 67 dB
Mark,
Timuikku's comment that "dynamic range would be height of the peaks from noise floor between distortion products" is an interesting way of looking at distortion being a reduction of dynamic range which increases with SPL.
THD and IMD are both non-linear, increasing more in % as SPL increases.
IMD also increases more as bandwidth increases, so in your five-way system crossing at 120, 300, 700,& 6300 Hz with the compression mid covering nearly three times the bandwidth as the 4" and tweeters, not surprising it would be the "weak link" in that regard, especially when bandwidth is increased even further.
That said, without first reaching the "tons of fuzz" level, still don't know which has more dynamic range
.Art
The mids on mine would work fine to 1kHz for sure, probably 1.1kHz no problem.
Their 7/8" diameter ports are 3.5" from CD flange, along the horn wall. I figure that could be tightened up to 3", perhaps 2.75" to the flange, to raise freq even higher if needed.
See my post #5 for my "sleeper" idea of a low cost killer build....
Hi Charles,
The biggest factor in my using the BMS, is i have 3 of them to pull off LCR, and only 1 of the B&C ...!!!!
Purely subjective sound quality......?
(i'll call the lower sections HF, and the higher sections VHF, on both of them.)
I'd give a small nod to the B&C when comparing the low sections. Seems a little more open and natural.
And then, a small nod to the BMS on the high sections.
There's a little more difference between the high sections, than the low sections i think.
The BMS all together ends up sounding a little more airy, with a little more sparkle than the B&C. But sometimes i wonder if the extra air and sparkle are supposed to really be there.
The clincher for me however, even if i had 3 of the B&C, is my unusual way of using them to adjust tonality. I keep a volume slider on each section to adjust for tonality, indoors vs out, low SPL vs high, track /album tonal variances, etc.
With the B&C's xover point in the mid to high 3kHz range, both HF and VHF have to be adjusted in tandem. I simply cannot move one without also moving the other nearly equally.
With the BMS's xover at 6.3KHz, the HF and VHF can be adjusted with much greater independence. Just makes dialing in tonality easier for me.
It would be really interesting to try a pair of those with 1" throats and my BMS4550 CDs crossed at 1 Khz. I've convinced myself that the 4NDFs would be good down to 150 Hz at my listening levels and so I would sit it on a sub crossed at that 150 Hz. Pretty simple and as you've shown effective! Unfortunately, I'm plagued with the challenge of getting the same/similar response from a much small footprint using cardioid techniques
Hi Art, yes...very interesting.
The idea of a many mutli-tones testing, producing both harmonic distortions and intermodulation distortions all through the spectrum, that essentially sum up to a noise floor, and have the real effect of reducing dynamic range, .........................sounds oh so plausible to me.
And goes a long way to describe why speakers sound shitty when overdriven.
Hat's off to all those who have led to this trail of investigation !!!!
Last edited:
I think that could be a very nice speaker. My only concern might be intermod using the 4NDFs down to 150Hz.
My gut instinct (which i truly know isn't worth a damn) says there is something about excursion ratios, low freg to high, that we need to pay attention to / keep under an unknown level.
I stay very interested in you cardioid explorations. Looking forward to a build.
you are right. sealed won't make it but vented tuned to 75 hz or so would. my HR sim shows .46 mm excursion at 105 db output with 150 Hz XO. At 100 Hz, its still under .6mm. My normal listening level is 85 db so it would work for me. I'll have to move to a bigger house before I think about building one any further.