Just had a look at some Lavoce spec sheets, and I'm glad I did.
A pair of WSF122.02 's should be arriving in the mail.
I like how they had tight Xmax specs without B.S.(TM) formulae to stretch them by 1/3 of the magnetic gap height for no reason (you know who you are...), and they always provided both the coil height and magnetic gap height.
It was funny how I ended up being guided by the lowest Xmech numbers. We'll see if my theorising about a combo of low Xmax + non-conductive VC former results in a nice midrange or not.
I was dithering between the 12" and the 15" version, which seemed to have exactly the same motor, but with more cone area and different suspension tuning. The 12" simulated well in 70L, while the 15" required upwards of 120L, and preferably more to avoid a big 80-100Hz peak coinciding with an impedance peak.
Incidentally, they had some guitar and bass speakers as well. I see no reason for avoiding speakers designed for bass guitar at least -- the specs look good, and most likely have to perform well up to 2-3kHz. One of the guitar speakers had an 8mm magnetic gap, and ~8.5mm for the coil! Seeing those numbers, it looks like they would actually sound really clear, until a bass note kicks in and introduces 'overdrive'. So I guess that that sort of speaker could be used as a large format midrange, say, 300Hz and up.
A pair of WSF122.02 's should be arriving in the mail.
I like how they had tight Xmax specs without B.S.(TM) formulae to stretch them by 1/3 of the magnetic gap height for no reason (you know who you are...), and they always provided both the coil height and magnetic gap height.
It was funny how I ended up being guided by the lowest Xmech numbers. We'll see if my theorising about a combo of low Xmax + non-conductive VC former results in a nice midrange or not.
I was dithering between the 12" and the 15" version, which seemed to have exactly the same motor, but with more cone area and different suspension tuning. The 12" simulated well in 70L, while the 15" required upwards of 120L, and preferably more to avoid a big 80-100Hz peak coinciding with an impedance peak.
Incidentally, they had some guitar and bass speakers as well. I see no reason for avoiding speakers designed for bass guitar at least -- the specs look good, and most likely have to perform well up to 2-3kHz. One of the guitar speakers had an 8mm magnetic gap, and ~8.5mm for the coil! Seeing those numbers, it looks like they would actually sound really clear, until a bass note kicks in and introduces 'overdrive'. So I guess that that sort of speaker could be used as a large format midrange, say, 300Hz and up.
As I only spotted a singleton at the local Idlefish Market (China ebay-like), wondering if it made any sense as a high-cross Sub supporting a pair of small/sealed fullrange (like FE108ez). Very much appreciate any tips.
The GA-SC64 is certainly not a subwoofer, it is more of a very efficient mid-bass / midrange. If you just wanted to fill in the the mid bass through the lower midrange to give some extra body to the full range drivers, it is not a bad idea. Sometimes low bass is not necessary, especially if you have nearby neighbors.
Last edited:
Thanks for the reply! Right, response doesn't go much lower than the 4" in theory but sometimes synergy happens.... I'll ponder it some more during Springfest (month-long pause in online shopping).
It seems like something that I would try also. Even though the 12" doesn't go much lower than the 4" drivers, I think it would sound cleaner and more powerful in the lower registers. I can feel bass drum beats even at low volumes from the 12" driver.
Xmax not withstanding, Linkwitz ( https://www.linkwitzlab.com/filters.htm#9 ) had some formulas that could be useful here. At least for me, I'm counting that with ambient / household listening levels (guesstimating around 75-80dB max?), the 95+dB sensitivity would have plenty of overhead to be flattened down.
The room is not very big, so it may not need much adjustment.
Looking at the absolute sensitivity:
My old 10" subs VS new 12" mid-woofers
20Hz: 73.5dB | 74dB
30Hz: 80dB | 81dB
40Hz: 83.5dB | 85.5dB
~~
100Hz: 90.5dB | 95.5dB
~~
300Hz: 92dB | 98dB
The 12" is actually a touch more sensitive in the deep bass. However, without a solid 6-7dB of EQ to counter the mid-range sensitivity, it would probably seem to have a lot less bass if comparing apple-to-apple vocal levels.
Will have to measure soon!
The room is not very big, so it may not need much adjustment.
Looking at the absolute sensitivity:
My old 10" subs VS new 12" mid-woofers
20Hz: 73.5dB | 74dB
30Hz: 80dB | 81dB
40Hz: 83.5dB | 85.5dB
~~
100Hz: 90.5dB | 95.5dB
~~
300Hz: 92dB | 98dB
The 12" is actually a touch more sensitive in the deep bass. However, without a solid 6-7dB of EQ to counter the mid-range sensitivity, it would probably seem to have a lot less bass if comparing apple-to-apple vocal levels.
Will have to measure soon!
Attachments
djk liked 6th order for low qts (.3) drivers.
You tune at its natural F9, then apply 6db of boost there.
Technically, you would run a high pass filter with a q of 2 at tuning frequency that acts like a subsonic filter below that.
Worked well on klipsch lascala's (you can search).
I looked at linkwitz boost for subs (nice tiny boxes), but distortion seemed really high for a woofer needing a lot of power to go flat to 30hz.
In the end, boosted 6th seems a better option.
To be fair, most big subs you buy have a ton of eq then subsonic protection, otherwise boxes would be too big to sell and move.
So linkwitz transform (used by bag end for a while) is an "elegant" solution.
You tune at its natural F9, then apply 6db of boost there.
Technically, you would run a high pass filter with a q of 2 at tuning frequency that acts like a subsonic filter below that.
Worked well on klipsch lascala's (you can search).
I looked at linkwitz boost for subs (nice tiny boxes), but distortion seemed really high for a woofer needing a lot of power to go flat to 30hz.
In the end, boosted 6th seems a better option.
To be fair, most big subs you buy have a ton of eq then subsonic protection, otherwise boxes would be too big to sell and move.
So linkwitz transform (used by bag end for a while) is an "elegant" solution.
Guitar speakers have a lot of color. They howl, cry, scream, and ring. What sound you get is often dependent on how much power is being used. If I want to play really clean I use a couple of hifi 8” midranges and only 5 watts or so. If I want a good old fashioned power amp distortion I use triode amp at 15 watts into one or two 12” blues toned woofers and turn it up to 11. If I want a tighter modern sound I use a pentode amp with a solid state rectifier at 30 or more watts into two or more 10’s or 12’s that scream. I don’t play in the low register but those players often use a mix of bass and guitar speakers. Also, some players who are into computer modeling simply plug into the mixing board and use the house PA.
Using different types of loudspeakers is like using different mutes on your trumpet
Using different types of loudspeakers is like using different mutes on your trumpet
I am pretty sure my guitar drivers don't howl, cry, scream, or ring when I am listening to a few milliwatts or less, sometimes much less...
Agree 100%. With my main rig the speakers are rather pedestrian until the volume on my amp is at 6-7, then they come alive. It’s a cool effect for a guitar rig but horrible for hifi!
So I guess the question is, are hifi speakers being 'sabotaged' at low SPL because of an over-emphasis on staying clean until they blow up? This "coming alive" seems like Xmax could be the driving force behind it. I thought there was some special sauce for guitar speakers, like offsetting the coil or detuning the motor to make it distort all the way down, but then I don't see how guitarists could choose to play cleanly just limiting the output level.
Correct! Les Paul's original amp that became the standard used a W.E./Lansing cinema (mid) bass horn driver with basically zero Xmax due to it having just a half roll flex out at the otherwise very rigid hemp paper cone and of course over time all sorts of electronic special effects were originally added to mimic the original's with differing driver designs over the decades required to offset the advancements in power system design that when I came to the forums was causing all manner of technical 'conflicts' between the hugely disparate tube Vs SS driver/speaker design routines/requirements.
I wouldn't say 'sabotaged', but I do think that some hi-fi speakers trade away low level details for flat response and possibly low distortion. I find that guitar drivers have excellent low level details, but they maybe sound a bit rough sometimes depending on the quality of the source material.
Have you tried Lavoces yet? Looks like they could be nice midbass for cheap two way. What horn-driver combo are you going to use?
Just a visual inspection and quick no-baffle test. Ignoring the 'squawking' tone due to lack of box, they seem to have really good sensitivity, and a kind-of 'depth' of clarity to the sound. The class-D amp hisses and crackles a lot through them. More incentive to hurry up and make a low-wattage class-a contraption for them.
I've finally started building the boxes. They're going to be big 70~75L made with solid larch wood, which may or may not be way too ambitious for my actual skill level. The larch is a conifer and I don't know what the gluing and shrinkage is going to be like. The grain looks quite resinous. The sides are going to be a cylindrical heart thingy with the grain aligned vertically all around, and a flat top and bottom. So, ideally, the flat parts should be made up of crazy pizza slices so the shrinkage matches the sides, but we'll see...
I've finally started building the boxes. They're going to be big 70~75L made with solid larch wood, which may or may not be way too ambitious for my actual skill level. The larch is a conifer and I don't know what the gluing and shrinkage is going to be like. The grain looks quite resinous. The sides are going to be a cylindrical heart thingy with the grain aligned vertically all around, and a flat top and bottom. So, ideally, the flat parts should be made up of crazy pizza slices so the shrinkage matches the sides, but we'll see...
EUROPEAN LARCH
Shrinkage:
Alt Text: "This denotes how much the wood will shrink dimensionally when going from green to ovendry MC. This is a good overall indicator of the wood's stability when encountering changes in humidity."
Radial: 4.2%
Tangential: 8.2%
Volumetric: 12.5%
T/R Ratio: 2.0
https://www.wood-database.com/european-larch/
https://www.wood-database.com/wood-articles/dimensional-shrinkage/
Shrinkage:
Alt Text: "This denotes how much the wood will shrink dimensionally when going from green to ovendry MC. This is a good overall indicator of the wood's stability when encountering changes in humidity."
Radial: 4.2%
Tangential: 8.2%
Volumetric: 12.5%
T/R Ratio: 2.0
https://www.wood-database.com/european-larch/
https://www.wood-database.com/wood-articles/dimensional-shrinkage/
Following up, it seems close to Noble Fir in movement, although that's just a guess. So using that (and Berlin for weather/temp), "Expected maximum movement: 0.12cm - 0.24 cm over the course of a year." for a 10 cm piece.
https://www.wood-database.com/noble-fir/
https://kmtools.com/pages/wood-movement-calculator
Not sure if you are an expert and just dissembling, you really want to make a complicated pie grain patterned top, or what.
I am far from an expert, but just in case:
https://www.woodmagazine.com/woodworking-how-to/wood-preparation/dealing-with-wood-movement
Most solutions I have seen have been along the lines of aligning the grain on the top, bottom, and sides, and screwing the front and back on in a manner that allows for the (presumably traverse) movement.
https://www.wood-database.com/noble-fir/
https://kmtools.com/pages/wood-movement-calculator
Not sure if you are an expert and just dissembling, you really want to make a complicated pie grain patterned top, or what.
I am far from an expert, but just in case:
https://www.woodmagazine.com/woodworking-how-to/wood-preparation/dealing-with-wood-movement
Most solutions I have seen have been along the lines of aligning the grain on the top, bottom, and sides, and screwing the front and back on in a manner that allows for the (presumably traverse) movement.
Attachments
Not seeing this in here in response to the originating post, but these laser interferometric scans of the cones of two drivers built for the different purposes are a great visual representation of the difference between guitar speakers and hifi speakers:
"The tone is in the cone" as the musician's saying goes. Guitar speakers are intentionally designed to break-up early and in ways that sound musical to the ears ie euphonic distortion. Hifi speakers try to remain pistonic and as well-behaved as possible so they minimize what they add to the music. It's production vs reproduction. As for the use of alnico magnets, the grades chosen for guitar speakers have a much lower coercivity than your typical barium ferrite magnets. Going to simplify this alot...what this basically means is that the magnet in the motor is more susceptible to being pushed/overpowered by that of the voice coil when a signal goes through it. To an extreme degree, it can result in permanent loss of flux, but before that point is reduces the magnetic force in the gap. This means that in this particular application the use of alnico allows for what is in effect compression of the dynamic range of transients. Subjectively, this compression makes things sound louder for the same output due to how the psychoacoustics of it work. Compression is used for stylistic reasons all through the music production chain (from tubes to tape to actual compressors) and this is just one more stage of it.
"The tone is in the cone" as the musician's saying goes. Guitar speakers are intentionally designed to break-up early and in ways that sound musical to the ears ie euphonic distortion. Hifi speakers try to remain pistonic and as well-behaved as possible so they minimize what they add to the music. It's production vs reproduction. As for the use of alnico magnets, the grades chosen for guitar speakers have a much lower coercivity than your typical barium ferrite magnets. Going to simplify this alot...what this basically means is that the magnet in the motor is more susceptible to being pushed/overpowered by that of the voice coil when a signal goes through it. To an extreme degree, it can result in permanent loss of flux, but before that point is reduces the magnetic force in the gap. This means that in this particular application the use of alnico allows for what is in effect compression of the dynamic range of transients. Subjectively, this compression makes things sound louder for the same output due to how the psychoacoustics of it work. Compression is used for stylistic reasons all through the music production chain (from tubes to tape to actual compressors) and this is just one more stage of it.
Attachments
Cutting and gluing the boxes is taking a while, and I'm not sure I want to rush that in the winter. In the meantime, I cut out a pair of "flower" shaped open baffles for the Alpair 5.3 for the high frequencies. 5 smooth slots in the circle, so it resembles a flower, are meant to eliminate peaks and dips, but honestly, my first reaction was that I kind-of hate the sound.
A light-bulb in series was an instant life-saver, with a significant reduction in 'grit' and harshness. So that gives me more motivation to finish a custom amplifier for it. But I'm not sure if I can tolerate the 4th-order roll-off of the open baffle around 1kHz. I can retrofit a closed box behind the baffle (or even a vented baffle to bring the resonance down), so it's not totally reliant on a woofer to neutralize the squawk-ish frequency response.
I was naive in thinking I could mount the magnet assembly and baffle on a separate stands to reduce vibrations -- the differential air pressure on its own is enough to rattle the dipole baffle even if it's not touching the speaker. Oh well.
I'm looking at the Lavoce DN07.10LM as a compression driver option..But I'm not sure if the datasheet has enough info to simulate different horns. The response in the datasheet goes way below the double impedance peak, which is confusing. And I'm hoping to run it down to 1kHz, assuming 10-100mW or something like that. For the price, I might just have to wing it and assume the 1.6T field strength means it's good enough.