T34 can go very loud and low so that it should be paired with a larger mid that can keep up, and so to take the advantage of the T34 size. For single 3"-5" mid the tweeter is somehow oversized. But it has very wide dispersion so too large and deep cones might not match well. I had the T34A running together with a 18Sound 8NMB420 as mid. Tweeter was fantastic but it was no match in sound signature. Changed to SB TW29TN, the tweeter was worse on its own but much better integrating with the 18S. So with the T34 I would prefer a large, loud mid with comparable wide dispersion - M142 would be ideal, if you have the money. Or take two M74A, D'Appolito if you like it (me not), or on a small baffle the 2nd Mid as bafflestep filler. Similar price and design is the Accuton C168 CELL Mid (or double C90 CELL Mid), but from data and dispersion I would then prefer the Bliesma. In that class is also Scan Speak Ellipticor or Kartesian 185_vKi, but I would take a Waveguide for the T34 then to match directivity. There are much cheaper (cone) mid alternatives out that might not be worse, but if you are pondering around that price class....
Yepp, this is also not my price range, and I doubt that a Revelator 18M for 1/4 the price performs noticeable worse. But M142 is a unicorn, there is nothing comparable of this driver kind with serious performance. If you're not willing to spend so much, take a Somasonus WG for the T34B and a nice powerful 7" mid(woofer), there are some out like the Revelator for much less but flawless performance.
Yup, my take was 4 x NE149-W (5.5") for a WMMT, big domes from Bliesma too much expensive. Be is dead (few SB29 SB Acoustic avialable here and there).
The only dome mid that was in my budget was the new Tang Band, but I wanted a much lower cut off, wider bandwidth mid : 3 octaves ! So all the Volt, Bliesma were way off my budget.
sorry, a bit off topic.
The only dome mid that was in my budget was the new Tang Band, but I wanted a much lower cut off, wider bandwidth mid : 3 octaves ! So all the Volt, Bliesma were way off my budget.
sorry, a bit off topic.
Thank you, @Kwesi, several of your thoughts are similar to mine- small mid (T74) unable to cross low enough to make sense, also don't really like D'Appolito arrangement, etc.
The waveguides I have seen for the T34B (mabat, hificompass, etc.) seem to have a 120 degree dispersion at about 1.3khz. This would seem to need a large mid/midwoofer/woofer (if a cone) that is becoming directive to match/not have a large change in directivity at XO. How big? PTT8? PTT10? Lom185_vKi?
For the T34B without a waveguide it seems that either a smaller cone mid/midwoofer or the (wider dispersion) M142 would be candidates to match dispersion characteristics at XO(?).
I have been thinking for weeks which driver to buy next and am ready to pull the trigger. Guess I am not too concerned about $ at this point as I have slowly been accumulating funds.... Though wanting to build L,R, and C (music the main emphasis) makes it even more daunting.
Thanks again,
Bill
The waveguides I have seen for the T34B (mabat, hificompass, etc.) seem to have a 120 degree dispersion at about 1.3khz. This would seem to need a large mid/midwoofer/woofer (if a cone) that is becoming directive to match/not have a large change in directivity at XO. How big? PTT8? PTT10? Lom185_vKi?
For the T34B without a waveguide it seems that either a smaller cone mid/midwoofer or the (wider dispersion) M142 would be candidates to match dispersion characteristics at XO(?).
I have been thinking for weeks which driver to buy next and am ready to pull the trigger. Guess I am not too concerned about $ at this point as I have slowly been accumulating funds.... Though wanting to build L,R, and C (music the main emphasis) makes it even more daunting.
Thanks again,
Bill
The T34B is not as wide radiating as the T34A but not as narrowing as the SB teeeters tend to be. You could just treat it like any other tweeter except it will go louder if that is what you need.
PKAudio used it in a waveguide with a narrower dispersion and crossed with a 6 or 7 inch mid scanspeak results look good.
You do not have to match directivity at the crossover point. Complete matches will produce a bump in the DI and then a more significant change in directivity quickly. A good but not complete match can produce a more smoothly changing DI.
Matching patterns at the crossover point is popular with narrower higher DI horns where you want to keep the DI as high as possible for as long as possible.
So it depends on what you want achieve.
PKAudio used it in a waveguide with a narrower dispersion and crossed with a 6 or 7 inch mid scanspeak results look good.
You do not have to match directivity at the crossover point. Complete matches will produce a bump in the DI and then a more significant change in directivity quickly. A good but not complete match can produce a more smoothly changing DI.
Matching patterns at the crossover point is popular with narrower higher DI horns where you want to keep the DI as high as possible for as long as possible.
So it depends on what you want achieve.
Yes. In fact, I have not found a fool-proof guideline to match up the DI of a LF and HF driver to ensure a smooth DI over the whole range. The only real way to know is to load the polar responses of both drivers into VituixCad and simulate some crossovers. Sometimes what seemed like a great match between mid and tweeter will turn out to be rather poor in overall DI, while in another case, what seems like a poor match will blend together nicely.
I'm pairing the T34B with 2 x M74A in MTM or TMM, depending on measurement results. A 2.5 way type of mid/tweeter arrangement can be very beneficial if its done just right, especially on narrow baffles. A single M74A would probably hang very well with the T34B, but I don't want a dynamic range deficit in any specific place in the curve.
My bass-mid drivers are 2x Eminence CB3010N-8. They can cross up to 800 hz 2nd order. I had an MTMWW planned for all the mentioned drivers which may change as the design evolves. I have separate sub enclosures for the 4 x SB34NRXL, each in their own 90 ltr vented box driven by 2 x SpeakerPower 700 plate amps.
At some point there was a thought of using an 18" per side and doing a 4 way with them, along with 2 x 8NDL51 midbass and the M74A/T34B arranged in MTM-MB-MB-W, but thats just obscenely huge.
The Eminence CB3010Ns perform exceptionally as extended range woofers in medium sized vented boxes. That's more manageable in size and format than a 4 way with 2 x 12". Using surround edge coating on the SB34NRXL could result in clean enough midbass for a 500 hz 2nd order LP mating with the M74As. The tweeter choice isn't yet decided between the T34B or T35C002. They're both great tweeters, but the Be dome will stay more composed at the extreme end of the SPL spectrum.
A big part of the final goal is ending up with a highly sensitive speaker. This is something I've always preferred, as high sensitivity usually goes hand in hand with low level detail retrieval and lower stress on the amplification. Of course its at minimum a bi-amped system (even without the subs), consisting of separate amp channels for midbass and mids/tweeters. Its better to have the flexibility of active analog filters than a fixed passive network with a ton of copper and huge, expensive caps. Only the subs get DSP, so its primarily an all analog system.
My bass-mid drivers are 2x Eminence CB3010N-8. They can cross up to 800 hz 2nd order. I had an MTMWW planned for all the mentioned drivers which may change as the design evolves. I have separate sub enclosures for the 4 x SB34NRXL, each in their own 90 ltr vented box driven by 2 x SpeakerPower 700 plate amps.
At some point there was a thought of using an 18" per side and doing a 4 way with them, along with 2 x 8NDL51 midbass and the M74A/T34B arranged in MTM-MB-MB-W, but thats just obscenely huge.
The Eminence CB3010Ns perform exceptionally as extended range woofers in medium sized vented boxes. That's more manageable in size and format than a 4 way with 2 x 12". Using surround edge coating on the SB34NRXL could result in clean enough midbass for a 500 hz 2nd order LP mating with the M74As. The tweeter choice isn't yet decided between the T34B or T35C002. They're both great tweeters, but the Be dome will stay more composed at the extreme end of the SPL spectrum.
A big part of the final goal is ending up with a highly sensitive speaker. This is something I've always preferred, as high sensitivity usually goes hand in hand with low level detail retrieval and lower stress on the amplification. Of course its at minimum a bi-amped system (even without the subs), consisting of separate amp channels for midbass and mids/tweeters. Its better to have the flexibility of active analog filters than a fixed passive network with a ton of copper and huge, expensive caps. Only the subs get DSP, so its primarily an all analog system.
Last edited:
@hifijim You probably already figured out that going from a large cone to a proportionally smaller dome (along with its different geometry in radiating surface) by nature results in a shift in directivity. The two things which can partially fix this either is shifting the turnover frequency slightly from a shared turnover frequency and / or gradually changing (tapering) the baffle width to help get a wider DI with the larger cone. Sometimes it requires more overlap, which has to be compensated for by using a different or modified rolloff slope. I'm not sure that a sim can show this accurately. Tapering the baffle shape is one option, but isn't flexible if the shape isn't as the sim suggests. Diagonally tapering the front baffle towards the mid and tweeter is one way to achieve a more gradual DI from the larger cone. Of course these are just generalizations and some experimentation will be needed, but narrowing the baffle towards the tweeter (among other things) has always worked for me in this regard.
This is the JBL 250ti -
This is the JBL 250ti -
Last edited:
Wow!! Now I gotta ask: please show us your room?! It must be a pretty big space, at least a small hall?
I am curious about the M74A's crossover frequencies. It seems that if you run the M74s up high you don't take advantage of the ability of the T34B to cross relatively low, and in that case could use the T25B? Though perhaps if you do that you are guaranteeing extreme dynamic capabilities up high from the T34.
If you don't cross them higher, then the low frequency capabilities leads the M74 to cover a small bandwidth(?). Or maybe it is a philosophy of where you prefer to have crossovers (and where you don't)(?).
What do you think about the M142A? If given the option would you prefer it to 2x M74s in combination with the T34?
Bill
@mikessi Yes, when its closer to completed, I'll put up some pictures. There's a lot still to finish before that and temporary stuff causes confusion.
The choice of M74A mids to blend with the one and only T34B was made by looking at alot of data from various sources. I still like my D7608s and believe they're some of the best 3" soft domes ever made for their modest price. The dome construction materials and profile are what makes this mid exceptional, despite apprehensive attitudes I come across criticizing its design. I'm here to prove those people wrong and show how good it really is when used correctly within its capable parameters.
I'm still not convinced the M74A and B are straightforward to cross under 500 hz, even at a much steeper slope. Thats why i decided to run a pair of them per side to get the distortion down as low as a larger cone midbass. Despite how much linear excursion these things are capable of, I'm not a fan of pushing them that hard. Small drivers shouldn't be played that way mechanically.
The choice of M74A mids to blend with the one and only T34B was made by looking at alot of data from various sources. I still like my D7608s and believe they're some of the best 3" soft domes ever made for their modest price. The dome construction materials and profile are what makes this mid exceptional, despite apprehensive attitudes I come across criticizing its design. I'm here to prove those people wrong and show how good it really is when used correctly within its capable parameters.
I'm still not convinced the M74A and B are straightforward to cross under 500 hz, even at a much steeper slope. Thats why i decided to run a pair of them per side to get the distortion down as low as a larger cone midbass. Despite how much linear excursion these things are capable of, I'm not a fan of pushing them that hard. Small drivers shouldn't be played that way mechanically.
.... not unless you want to beat them up, wearing them out sooner and reducing their mechanical life expectancy.
Not to speak about a lot more distortion by pushing these small membranes to it's limit. I noticed that when i went from a 6 dB cross over of 600 Hz to 1.2 KHz 6dB with my 54 mm dome mid range. My ears got tired when i pushed the volume up, now with the higher cross over point, i can listen at much higher volume without fatigue.
Performed a few tests setting up the first layer for printing the D7608 chamber. The compiled file is estimated at 16 hrs to print with 70% infill and support for the center inside tip of the toroidal shape. I'm using Sunlu PLA pro filament to keep things simple for the first print. Once the dimensions have been tweaked for an accurate fit and I'm confident the first layer looks just right, I'll let it rip for a full print.
@Henk Haring Cutting the excursion to 25% of what it was definitely makes a difference in distortion. These 2" domes don't do that well at higher volume levels at a 600 hz 1st order HP. I usually run those D54s at 900 hz 2nd order, which appears to be about where they work all around best. It does sound very good at lower SPLs listening to a D54 crossed lower, but it all just falls apart fast once you bump the volume just a little bit.
A test I use to help determine the right crossover point and slope on a mid dome is a well recorded snare drum. If the mid can keep up with this type of transient, it will do well with that filter setting. What's interesting is the stated xmax on most mid domes isn't accurate acoustically speaking. I've heard domes with 1mm xmax fall apart at lower volume levels than domes with 0.5mm xmax. It may be something to do with the magnetic field symmetry or possibly manufacturing tolerances.
A test I use to help determine the right crossover point and slope on a mid dome is a well recorded snare drum. If the mid can keep up with this type of transient, it will do well with that filter setting. What's interesting is the stated xmax on most mid domes isn't accurate acoustically speaking. I've heard domes with 1mm xmax fall apart at lower volume levels than domes with 0.5mm xmax. It may be something to do with the magnetic field symmetry or possibly manufacturing tolerances.
There’s a recent video on YouTube showing that the infill ratio barely matters in terms of frequency response or resonance. You can go as low as 30% without penalties in performance. It could help with faster iteration of the design.
@SunRa I can appreciate that thought, but I'd rather start with a more dense part, just to make sure the wall density itself won't change or affect anything outside of the chamber's physical shape and/or geometry. I don't feel its worth the reduction of print time to end up with a less dense part and potentially dealing with other issues this may present. People will be welcome to change the infill ration on their own, which is easy to do with the help of most slicing programs.