Then do a Fourier analysis on that album and see which frequencies actually need the most power. The mid will probably be 10dB down from the sub. And several dB below clipping. This is why speaker manufacturers are able to rate little 64 mm x 15mm coils with fine gauge wire at "350W". But don't take my word for it, measure it.
You use a TMT-M-T-M-TMT setup and restrict the outer drivers? Or TMT-TMT-M-M-T?
With 2 M74As ... you for SURE have enough power at 700Hz. 2 D7608 won't do a lot here.
And you can cross the T34B lower - it's a beast. Looking forward to your measurements and listening experience at different crossover points.
Can you do me a favour - can you measure 8NDL51 at different levels and show THD over frequency? Like I do in my test?
You know not all PA drivers behave at low SPL levels, 8NDL51 is an interesting candidate and I expect good performance.
And SB34NRXL75 - you need at least 4 to come close to that midrange power, maybe 8. You know my measurements, one M74A could easy stand with 2 of them in a linear speaker (no bass boost as used normally, crossover in the 600-650Hz area).
Just in comparison - this is THD of my prototype with 2x 12" close, 1x M74A and 1x T25B:
compared to one of the lowest THD speakers measured at ASR:
At 114dBSpl the LF drivers start to compress but not the single 3". And this are sine sweeps, not music with crest factor. So you can expect 120dB long term SPL from 2 M74A and plenty dB of peak over that.
compared to one of the lowest THD speakers measured at ASR:
At 114dBSpl the LF drivers start to compress but not the single 3". And this are sine sweeps, not music with crest factor. So you can expect 120dB long term SPL from 2 M74A and plenty dB of peak over that.
Attachments
@IamJF I've suspected the M74A would be capable of more than I'm ready to throw at it. I don’t want to cross under 3k to the T34B, jainly because its something that has always sounded "right" to my ears not pushing anything which radiates that wide between 2 and 3.5k
That causes issues in the room unless an effective WG is used to narrow it closer to the midrange. The other reason is the balance in dynamic capability, which is greatly skewed running a low crossover on a tweeter or anything with a thermally less capable VC with thin windings and a poor thermal conductive VC carrier. Looking at percentages of pink noise vs average power distribution of various types of music, the power split occurs at widely varying crossoved points and cutoff slopes. 400 hz used to be deemed as the accepted split point for 50/50 power distribution back in the late 70s.
Now music is much more bass heavy and saturated (partially because of the limitations of vinyl) and the crest factor is higher. I've done analysis on various genres of music using Wavelab 11 pro mastering software. It shows great variations in power distribution, especially with heavy rock (screaming guitars), large big band jazz orchestras and drum solos (crash cymbals will almost always blow HF drivers at those levels). With these types of music played at a natural balance without EQ and very limited compression (the term "natural" with regards to rock is to be taken loosly), the percentage of power the mids and tweeter has to handle are much higher. The other aspect is HP cutoof slopes used. Those will greatly change the percentage of power the tweeter has to take. If i play a realisticly recorded, uncompressed gun shot from a 44 magnum (yes, I've recorded those too) you have alot of concentrated midrange energy from 200 hz on up to 6 k. No, I dont play gun shots just for kicks, but they are surprisingly close to very heavy rim shot on an undampened snare drum with a 2 ply head. So, while I believe you're not far off from your actual modern day power distribution percentages associated with most of today's music, these are different than the typical music and other audio I work with. Some demos I did with ATC monitors, the listeners weren't told what to expect. Its interesting to see how every day people react to surprising, very loud noises. Using very large 2" CDs mounted on medium width pattern horns, these were the only speakers available which could repeatably stand up to the punishment. We blew 2 tweeters on the ATCs, which was getting expensive and needed a more affordable durable speaker.
I worked with Foley artists for a long time building sound libraries for film, research and firearms training. I used to set up tactical training scenarios for LEOs and military. We designed the first speaker setups which could play back gun shots at realistic levels. We first used large fullrange drivers with big VCs to keep them from being damaged by heavy transients, but they would compress badly in the mids, so we ended up with large 2 way sysyems with huge 2" CDs using titanium diaphragms. Playing back such program material takes peaks of a few hundred watts on very efficient drivers which can handle the bursts of audio, including sirens, propeller airplanes and turbo fan engines. I want my setup to tolerate some greater extremes in power distribution so I can reproduce some of these sounds without ruining excursion limited, smaller diaphragm drivers
and its going to he arranged as M-T-M-MB-MB-W-W. The T-M-MB will play its full dedicated range and the upper M, and lower MB will be shaded.
That causes issues in the room unless an effective WG is used to narrow it closer to the midrange. The other reason is the balance in dynamic capability, which is greatly skewed running a low crossover on a tweeter or anything with a thermally less capable VC with thin windings and a poor thermal conductive VC carrier. Looking at percentages of pink noise vs average power distribution of various types of music, the power split occurs at widely varying crossoved points and cutoff slopes. 400 hz used to be deemed as the accepted split point for 50/50 power distribution back in the late 70s.
Now music is much more bass heavy and saturated (partially because of the limitations of vinyl) and the crest factor is higher. I've done analysis on various genres of music using Wavelab 11 pro mastering software. It shows great variations in power distribution, especially with heavy rock (screaming guitars), large big band jazz orchestras and drum solos (crash cymbals will almost always blow HF drivers at those levels). With these types of music played at a natural balance without EQ and very limited compression (the term "natural" with regards to rock is to be taken loosly), the percentage of power the mids and tweeter has to handle are much higher. The other aspect is HP cutoof slopes used. Those will greatly change the percentage of power the tweeter has to take. If i play a realisticly recorded, uncompressed gun shot from a 44 magnum (yes, I've recorded those too) you have alot of concentrated midrange energy from 200 hz on up to 6 k. No, I dont play gun shots just for kicks, but they are surprisingly close to very heavy rim shot on an undampened snare drum with a 2 ply head. So, while I believe you're not far off from your actual modern day power distribution percentages associated with most of today's music, these are different than the typical music and other audio I work with. Some demos I did with ATC monitors, the listeners weren't told what to expect. Its interesting to see how every day people react to surprising, very loud noises. Using very large 2" CDs mounted on medium width pattern horns, these were the only speakers available which could repeatably stand up to the punishment. We blew 2 tweeters on the ATCs, which was getting expensive and needed a more affordable durable speaker.
I worked with Foley artists for a long time building sound libraries for film, research and firearms training. I used to set up tactical training scenarios for LEOs and military. We designed the first speaker setups which could play back gun shots at realistic levels. We first used large fullrange drivers with big VCs to keep them from being damaged by heavy transients, but they would compress badly in the mids, so we ended up with large 2 way sysyems with huge 2" CDs using titanium diaphragms. Playing back such program material takes peaks of a few hundred watts on very efficient drivers which can handle the bursts of audio, including sirens, propeller airplanes and turbo fan engines. I want my setup to tolerate some greater extremes in power distribution so I can reproduce some of these sounds without ruining excursion limited, smaller diaphragm drivers
and its going to he arranged as M-T-M-MB-MB-W-W. The T-M-MB will play its full dedicated range and the upper M, and lower MB will be shaded.
Last edited:
Greetings ! Have you got some documentation somewhere showing your approach on this match-up? Let's see, I think we write @scholl
I use a basic iterative approach; build, measure, theorize, calculate, build, listen. I build the box, load the drivers and measure them without a crossover. Take a look at the measurements then theorize what type of crossover would work. Use a calculator to get the crossover parts values into the ball park, build it then measure again until I have what I want or what works best. Recently I've been exporting the measurements from REW to XSIM then I dicker around with values to get into the ball park then measure, tweak and measure until I get what I want. Then listen to the results. Sometimes less perfect crossovers sound better.
I don't have any other documents but from much, much reading of what others are doing, we all do the same things. Many good writeups show the final product and leave out all of the iterations.
In the old days folks would mostly tweak and listen. They\we didn't have the tools we do today or they were very expensive.
Also, for a new idea I spend a lot of time choosing the parts. But also sometimes I find parts I didn't specifically choose at very low prices and make them work. 2206 midbass is IMO a very good part from 130-1200 in closed box. 70-1200 in BR. A builder could do worse than mating it to a Wavecore TW030wa11 or 12
https://ds18.com/collections/8-loud...nge-loudspeaker-275-watts-rms-550-watts-max-1
https://ds18.com/collections/8-loud...ro-8-mid-bass-loudspeaker-300-watts-rms-4-ohm
https://ds18.com/collections/8-loud...range-loudspeaker-with-bullet-600-watts-4-ohm
https://ds18.com/collections/8-loudspeakers/products/pro-exl-8-midrange-loudspeaker-800-watts-4-ohm
https://ds18.com/collections/8-loudspeakers/products/pancadao-mid-bass-8-1000-rms-4-ohms
https://ds18.com/collections/8-loud...ymium-magnet-full-range-loudspeaker-500-watts
https://ds18.com/collections/8-loud...ro-8-mid-bass-loudspeaker-300-watts-rms-4-ohm
https://ds18.com/collections/8-loud...range-loudspeaker-with-bullet-600-watts-4-ohm
https://ds18.com/collections/8-loudspeakers/products/pro-exl-8-midrange-loudspeaker-800-watts-4-ohm
https://ds18.com/collections/8-loudspeakers/products/pancadao-mid-bass-8-1000-rms-4-ohms
https://ds18.com/collections/8-loud...ymium-magnet-full-range-loudspeaker-500-watts
I really enjoy the broad scope of suggestions here. While they may be off the mainstream path of logic, it helps to look off the beaten path for inspiration. That can lead to finding a solution which would have been ignored using the direct approach.
DS18 may have a few hidden gems, but I'm unfortunately not thrilled with their QC and consistency. Almost every cone driver I bought from them before had some manufacturing debris stuck in the VC gap. Thats not good in any situation.
Having weighed my options, I feel that multiple 8NDL51s will pose the most economical yet functional solution. Its a really good driver with many attributes. One of these isn't sensitive enough and won't keep up if the HP is pushed low enough to work with a subwoofer. Two of them will be plenty powerful for midbass with some extra leeway to experiment with various HP frequencies and slopes. Its just not a pretty looking driver front mounted, but as long as they sound great, cosmetics are secondary.
I'll try getting some distortion measurements. These pro drivers are usually a little hotter on 2nd order HD, but that can make things sound a little more lively in a good way.
DS18 may have a few hidden gems, but I'm unfortunately not thrilled with their QC and consistency. Almost every cone driver I bought from them before had some manufacturing debris stuck in the VC gap. Thats not good in any situation.
Having weighed my options, I feel that multiple 8NDL51s will pose the most economical yet functional solution. Its a really good driver with many attributes. One of these isn't sensitive enough and won't keep up if the HP is pushed low enough to work with a subwoofer. Two of them will be plenty powerful for midbass with some extra leeway to experiment with various HP frequencies and slopes. Its just not a pretty looking driver front mounted, but as long as they sound great, cosmetics are secondary.
I'll try getting some distortion measurements. These pro drivers are usually a little hotter on 2nd order HD, but that can make things sound a little more lively in a good way.
They went MIA here for two years. The distributor is the direct retailer and for that whole time they have been responding with "getting the container ready"
Back in stock now and I have to say that I find the build quality visibly superior to Daytons and even JL Audio
That one really is a 500 hz + driver for horn loading. I need more bandwidth in the lower mids.
Would you please define
What exactly do you mean by that? Specifically spl and low-end extension.
What exactly do you mean by that? Specifically spl and low-end extension.
@profiguy
Fantastic driver for the price:
https://www.parts-express.com/Tymph...-Paper-Cone-Woofer-8-Ohms-299-2265?quantity=1
Not sure if it suits your requirements…
Fantastic driver for the price:
https://www.parts-express.com/Tymph...-Paper-Cone-Woofer-8-Ohms-299-2265?quantity=1
Not sure if it suits your requirements…
profiguy in post #1
"I'm in search of a decent 8" midbass driver. Needs to do mid-90s dB/2.83v and handle the 300 - 900 hz range with ease, preferably 4 ohm impedance.
The main issue I'm running into looking for a suitable driver is sufficient xmax along with efficiency. The driver will need to handle some decent power without turning into mush with loud transients. <snip> Levels I expect are 110 dB peaks in the specified range.
<snip> enough xmax to deal with a 300 hz 2nd order HP at higher output levels."
A driver with as much xmax as the 8NDL51 but with more power handling ie. 8NDL64. I'd choose the 8NDL51 in 4 ohms if it handled the extra power I need running in a smaller sealed enclosure. The issue is the power compression being in a small air tight box. The larger VC would help, but a second 8NDL51 paralleled and shaded would be better for output sake, keeping distortion to a minimum. This would also make the crossover point more flexible.