Working on my son's PA cabs. The design is a 15" Eminence KL3015HO, 2× 6" Celestion NTR06-1705B and CDX1-1430 CD on Faital STH100 WG. Mids and HF will be arranged in MTM. Xover is passive at 350 / 2.5 - 3k 2nd / 3rd order. Cab is vented, Fb tuned to 45 hz. I'm using a pair of the Parts Express 15" trap cab flat packs I had left over. They have a good size baffle and decent volume for good low end extension.
My issue here is insufficient baffle height to place the WG between both mids, so I have to compromise by offsetting the WG to the side and place mids right underneath each other. In the past, that's what yielded the best off axis response and performance but I'm wondering if there's a better way to improve off axis behavior.
I know the mid LP is a little high for optimal vertical axis MTM behavior, but I'd rather keep the xover little higher for cleaner CD performance and better HF headroom. The CD/WG combo sounds very good above 2.5k and lowering it ruins the HF clarity. The concern however is creating a problem horizontally off axis by placing the WG to the side of the mids.
Only running one of the mids up to the HF isn't an option due to insufficient upper mid sensitivity. After xover losses and BSC there should be about 97dB sensitivity. I allowed some extra loss for the mid LP inductor series resistance, so running both mids parallel at 4 ohms will end up about 6 ohm in the mids and net a little lower mid acoustic gain.
Any suggestions would be great.
My issue here is insufficient baffle height to place the WG between both mids, so I have to compromise by offsetting the WG to the side and place mids right underneath each other. In the past, that's what yielded the best off axis response and performance but I'm wondering if there's a better way to improve off axis behavior.
I know the mid LP is a little high for optimal vertical axis MTM behavior, but I'd rather keep the xover little higher for cleaner CD performance and better HF headroom. The CD/WG combo sounds very good above 2.5k and lowering it ruins the HF clarity. The concern however is creating a problem horizontally off axis by placing the WG to the side of the mids.
Only running one of the mids up to the HF isn't an option due to insufficient upper mid sensitivity. After xover losses and BSC there should be about 97dB sensitivity. I allowed some extra loss for the mid LP inductor series resistance, so running both mids parallel at 4 ohms will end up about 6 ohm in the mids and net a little lower mid acoustic gain.
Any suggestions would be great.
There's little to be gained from the symmetric arrangement if you use a symmetric crossover (like LR). If you use an asymmetric cross (like Butterworth), you lose some of the synergy if you offset the tweeter.
It comes down to why you are using MTM and what the trade-offs are.
It comes down to why you are using MTM and what the trade-offs are.
I'm using two mids to allow for a lower mid xover and more lower mid sensitivity. I also want to keep as much of the lower mids away from the 15" as possible. I intended to have a more narrow vertical directivity similar from MF to HF via the mid driver spacing. The WG only does about 50 degrees vertical, so by simming the mid driver spacing and xover, there won't be such a big shift in vertical directivity at xover.
Attachments
Ok, so having the mids spaced as far apart as a regular MTM would be a bad idea, and the way you cross them to the tweeter is the real issue. Do you cut to one mid before the cross? Is two stacked mids going to produce an asymmetrical pattern or not? Will it have the same DI? How will that extend with lobing through the cross.
I see that the waveguide pattern has flipped by 2.5k The horizontal is waistbanding and the vertical is losing control. I'm not suggesting it isn't usable. Perhaps it would be a better match to two mids side by side rather than vertical, if they were just the correct size for the cross. Just saying.. there may be more to it than meets the eye.
I see that the waveguide pattern has flipped by 2.5k The horizontal is waistbanding and the vertical is losing control. I'm not suggesting it isn't usable. Perhaps it would be a better match to two mids side by side rather than vertical, if they were just the correct size for the cross. Just saying.. there may be more to it than meets the eye.
DIYSG had a 1099 and a 1299 design that incorporates 2 side by side mids under the horn in a mtm configuration, here’s a link to Erin’s audio with info on the design of the 1099…..maybe you could size it up to a 1599?
https://www.erinsaudiocorner.com/loudspeakers/diysg_1099/
https://www.erinsaudiocorner.com/loudspeakers/diysg_1099/
I'm sorry to say but - your concept makes problems.
You would need a VERY steep crossover to make this work - off axis performance is really important in such a speaker! 2,5kHz is an important range, you will get crazy horizontal dips. (better as peaks but still ...)
The only way to rescue this bunch of chassis - do a TT-M-M-H configuration and blend the lower M earlier out. You should achieve 95dBSpl sensitivity that way - that's definitely enough when you want some bass from the 15".
Or ditch the 2 6" and get a high performance 8".
You would need a VERY steep crossover to make this work - off axis performance is really important in such a speaker! 2,5kHz is an important range, you will get crazy horizontal dips. (better as peaks but still ...)
The only way to rescue this bunch of chassis - do a TT-M-M-H configuration and blend the lower M earlier out. You should achieve 95dBSpl sensitivity that way - that's definitely enough when you want some bass from the 15".
Or ditch the 2 6" and get a high performance 8".
Yeah, I totally expected this would be a compromised situation being stuck on the HF crossed at 2.5 k. Its just that I absolutely love the way the top end sounds with that CD + WG. I do have other WGs to use and perhaps they may be a better choice. The mids have a narrow cutoff around 2.5 k as mentioned before, so I thought they would work crossed there being the WG is also narrow at that frequency.
Other WGs I have are the B&C ME20, 18sound XT1086, Selenium HM25-25. I may have a pair of XT120, but those are also 50 deg vertical. The diffraction of the wider WGs sounds bad at higher levels, that's the other issue for me.
I've toyed with placing the mids side by side. I've also thought about angling the mids out by a few degrees to overcome their narrow pattern at xover, something I had success with previously using multiple mids playing the same range.
I don't really know I have other options if I want to keep the current recipe of parts. I'm ok with boosting low end if I can retain a bit of the mid band efficiency of the LF driver. I just didn't want to run it higher up and kill the tighter low mids of the two midbass drivers. I just hate the sound of common affordable WGs in the 1 - 3k range.
I have other mids like the 8PE21, but they don't have enough xmax to cross under 500 hz or so, plus the mids aren't as good. Being I chose the components on the basis of SQ in their intended range, using other mids and HF would be throwing the existing recipe out the window, even if the pattern was a little wonky where I have to cross them. The other factor is weight. Hence the desire to use all neo drivers so they can be moved around easier, especially with my health being as bad as it is.
I know this sounds silly, but I'd rather compromise the off axis behavior a little and retain more of the on axis purity. Most people would opt for the better pattern control over SQ. I'm just torn here trying to use existing parts forced into a compromised design. I just hate the sound of a 1" CD running under 2.5k, especially if it doesn't have the top end low THD performance of the CDX1-1430.
I have a bunch of other CDs ie NSD1095N. DE250, CDX1-1745 and HF108, but they all don't have that open top end as the 1430. The NSD1095N can come close, but still not as good, especially in the lower mids as it gets really rough under 2.5k. I'm just one of those guys that doesn't like the small compression driver sound in the 1k to 3k mids. 90 percent of them sound harsh to my ears, especially with vocals. Its like fingers down a chalk board.
Other WGs I have are the B&C ME20, 18sound XT1086, Selenium HM25-25. I may have a pair of XT120, but those are also 50 deg vertical. The diffraction of the wider WGs sounds bad at higher levels, that's the other issue for me.
I've toyed with placing the mids side by side. I've also thought about angling the mids out by a few degrees to overcome their narrow pattern at xover, something I had success with previously using multiple mids playing the same range.
I don't really know I have other options if I want to keep the current recipe of parts. I'm ok with boosting low end if I can retain a bit of the mid band efficiency of the LF driver. I just didn't want to run it higher up and kill the tighter low mids of the two midbass drivers. I just hate the sound of common affordable WGs in the 1 - 3k range.
I have other mids like the 8PE21, but they don't have enough xmax to cross under 500 hz or so, plus the mids aren't as good. Being I chose the components on the basis of SQ in their intended range, using other mids and HF would be throwing the existing recipe out the window, even if the pattern was a little wonky where I have to cross them. The other factor is weight. Hence the desire to use all neo drivers so they can be moved around easier, especially with my health being as bad as it is.
I know this sounds silly, but I'd rather compromise the off axis behavior a little and retain more of the on axis purity. Most people would opt for the better pattern control over SQ. I'm just torn here trying to use existing parts forced into a compromised design. I just hate the sound of a 1" CD running under 2.5k, especially if it doesn't have the top end low THD performance of the CDX1-1430.
I have a bunch of other CDs ie NSD1095N. DE250, CDX1-1745 and HF108, but they all don't have that open top end as the 1430. The NSD1095N can come close, but still not as good, especially in the lower mids as it gets really rough under 2.5k. I'm just one of those guys that doesn't like the small compression driver sound in the 1k to 3k mids. 90 percent of them sound harsh to my ears, especially with vocals. Its like fingers down a chalk board.
Last edited:
The Selenium HM25-25 is an expo WG and sounds decent, but that pushes the acoustic center way back factoring in the cone mids at xover. I wanted to use the Celestion H7050 WG, but those aren't currently available here in the US. All the decent small WGs are hard to get here right now.
I also considered a K-tube instead of a traditional WG, but I have no clue what the radiation pattern is on those, plus the typical 3d printed material isn't really trustworthy from a mechanical durability standpoint unless exotic filament is used with a lighter weight CD.
I also considered a K-tube instead of a traditional WG, but I have no clue what the radiation pattern is on those, plus the typical 3d printed material isn't really trustworthy from a mechanical durability standpoint unless exotic filament is used with a lighter weight CD.
Last edited:
@AllenB This is the FR on and off axis of the Celestion NTR06-1705B mids. They get pretty narrow despite what the shallow cone profile may suggest. They just sound very nice, but wide coverage isn't their strong suit.
I really can't afford other drivers at this point so I'm stuck with the handicap of using these mids. I thought their narrow off axis behavior would be a benefit also having the narrow WG performance around the xover.
I really can't afford other drivers at this point so I'm stuck with the handicap of using these mids. I thought their narrow off axis behavior would be a benefit also having the narrow WG performance around the xover.
Attachments
Last edited:
For comparison sake I took my son to Guitar Center and we listened to all their typical big 2 ways they carry. We couldn't deal with the sound any of the speakers had, even with various EQ applied. They all just sound horrible to both of us, especially with vocals and at higher levels. I don’t know how people can settle for this type of sound except for just the convenience of a light weight self powered plastic box. Its just awful. That's what led us to build our own, being there was no alternative in the typical price range.
One of my litmus tests for vocals is the track Jezebel from Sade. The vocals on this song are just exquisite and the nasal sounding sax solo pushes any midrange problems right up front. All the speakers failed this test miserably.
One of my litmus tests for vocals is the track Jezebel from Sade. The vocals on this song are just exquisite and the nasal sounding sax solo pushes any midrange problems right up front. All the speakers failed this test miserably.
Last edited:
I once built a 12"/6"/1" fullrange party speaker. And while sound itself was really good - it sounded different all over the room. Ditched the project.
Not sure of you want it for listening or live - but for professional use that's not really useable.
You should try DE360 - best top end I ever tested. Very relaxed and clear and no resonances.
Best 1" horns are from Limmer Horns, not sure if you can source them. I dislike all the horns you list for your project 😏 XT120 should be very similar to what you already use, XT1086 only makes sense in a 2-way with deep crossover.
The setup I sketched should work, I would sacrify 2dB sensitivity over uneven off axis behaviour for sure! You are limited with the 15" driver anyways and need a powerful amp for it.
Not sure of you want it for listening or live - but for professional use that's not really useable.
You should try DE360 - best top end I ever tested. Very relaxed and clear and no resonances.
Best 1" horns are from Limmer Horns, not sure if you can source them. I dislike all the horns you list for your project 😏 XT120 should be very similar to what you already use, XT1086 only makes sense in a 2-way with deep crossover.
The setup I sketched should work, I would sacrify 2dB sensitivity over uneven off axis behaviour for sure! You are limited with the 15" driver anyways and need a powerful amp for it.
I don't build alot of PA stuff,, but when I do, it has to have some capability of playing intricate material, but it has to deliver good dynamics with the appropriate music..
The speakers here in question are for my son's band he plays guitar in, used for small gigs indoors and . That's why I thought narrow vertical dispersion would help keep reflections off the floor and ceiling. We have plenty of clean power on hand. Weight is an issue for sure.
His band plays alot of acoustic music with delicate percussion parts, airy ambient guitar parts and light vocals. When he heard the monitors I put together for another project, he was hooked. His ears are even more sensitive than mne in the mids, so he specifically wanted a 3 way setup based on the designs I typically built as monitors. I played him this track from an album I helped out on back in the day with digital mastering -
The speakers here in question are for my son's band he plays guitar in, used for small gigs indoors and . That's why I thought narrow vertical dispersion would help keep reflections off the floor and ceiling. We have plenty of clean power on hand. Weight is an issue for sure.
His band plays alot of acoustic music with delicate percussion parts, airy ambient guitar parts and light vocals. When he heard the monitors I put together for another project, he was hooked. His ears are even more sensitive than mne in the mids, so he specifically wanted a 3 way setup based on the designs I typically built as monitors. I played him this track from an album I helped out on back in the day with digital mastering -
Just a thought, PHL have excellent coaxials where you can fit your own 1" and 1.4" HF unit, https://www.phlaudio.com/products/index.html
When I mentioned side by side mids, I was implying the pattern flip in the bigger picture. Other solutions would be to rotate the waveguide, cross it higher or use a different waveguide.
Asymmetrical waveguides have their own problems but I understand you want to stay with this one.
However one of the obvious concerns is the dual mids (as noticed in MTM designs in general). Unless you can justify staying with them, by measurement or simulation (and I figured as long as you're using the waveguide down to 2.5k where it's asymmetrical, there's one possibility...), then I'd also consider cutting out one mid before the cross.
Asymmetrical waveguides have their own problems but I understand you want to stay with this one.
However one of the obvious concerns is the dual mids (as noticed in MTM designs in general). Unless you can justify staying with them, by measurement or simulation (and I figured as long as you're using the waveguide down to 2.5k where it's asymmetrical, there's one possibility...), then I'd also consider cutting out one mid before the cross.
@AllenB I see what you mean with the pattern flip.
I also have these mid WGs from JBL MRX series. They're designed for an Eminence 6" cone, but could be adapted to my Celestions. I just dont know if it would be a waste of time and the bandwidth wouldn't be there to cross higher.
I also have these mid WGs from JBL MRX series. They're designed for an Eminence 6" cone, but could be adapted to my Celestions. I just dont know if it would be a waste of time and the bandwidth wouldn't be there to cross higher.
Attachments
Last edited: