Hi all, a while ago I purchased a very lightly used set of, all Morel, ET 338, CAW 538 and CAW 938. My intention was and still hopefully is to pack these drivers into a nice 3 way speaker.
I have been learning VituixCAD but am a bit stumped with designing an enclosure for the CAW 938. I have a few questions about this regarding simulated SPL, Xmax and overall rules around these topics.
Please correct me if I'm wrong with anything above. Please see some enclosure simulations below and also some nearfield measurements of the CAW 938 and 538 test enclosures, these measurements are also gated from memory.
Hopefully this data is enough to go on, let me know if not or you need any more info. All measurements in REW have 1/24th smoothing applied.
EDIT: Sorry I also forgot to add the nearfeild accuracy cut off's.
Thank you!
I have been learning VituixCAD but am a bit stumped with designing an enclosure for the CAW 938. I have a few questions about this regarding simulated SPL, Xmax and overall rules around these topics.
- What SPL should these speakers be able to play up to, as in max db SPL for loud listening in a medium to large lounge at between 3-5m listening distance? Or if there is a standard for general high end home audio speaker I would be happy to aim for that.
- In terms of woofer roll off at what point do you not need to worry about cone excursion, weather that's referencing the SPL roll off, the power dip at the impedance spike, cone force or velocity. I have looked around but cannot find any solid information on what is safe in terms of Xmax, everything from sealed enclosure to bass reflex and band pass all pass the rated Xmax of driver even at fairly low wattage, why is this not a big concern. What are some good rules or parameters to follow in terms of this while getting the desired SPL output.
- I really want to do a nice job with these speaker as it will be the first I have designed myself and will be my daily drivers. Keeping this in mind they will be mainly used for music and watching movies. I would ideally like the qtc to be reasonably balance around 0.707 to avoid sloppiness or compression.
- I have built and done some measurement on some test enclosures. The volumes and dimensions are listed below:
- CAW 938: 320mm x 730mm x 436mm, built in 18mm OBS, the volume is 76.5L with a qtc around 0.71.
- CAW 538: 260mm x 320mm x 210mm built with the same 18mm OBS, the volume is 10.5L giving a qtc of 0.5 to hopefully keep the mids nice and dynamic.
- Finally, is the CAW 938 not suited for a low range woofer, my initial goal was anywhere from 500 to 250Hz down to the driver roll off with no HPF. If this does not seem viable I could definitely use it for a 2 way using it with a beyma compression driver.
Please correct me if I'm wrong with anything above. Please see some enclosure simulations below and also some nearfield measurements of the CAW 938 and 538 test enclosures, these measurements are also gated from memory.
Hopefully this data is enough to go on, let me know if not or you need any more info. All measurements in REW have 1/24th smoothing applied.
EDIT: Sorry I also forgot to add the nearfeild accuracy cut off's.
- CAW 938: 655Hz
- CAW 538: 1023Hz
Thank you!
Attachments
Last edited:
Hello
It's more like can they meet your SPL requirements in your room at a given distance. For example what are your Peak SPL requirements at the given listening distance? You can use the calculator to help work it out. Just drop in sensitivity, distance, placement and power available to get the max SPL. The other question is for example do you plan on using them to watch movies at THX levels? In that case SPL requirements are clearly defined. So you plug it in and see if you can meet them.
This should help on this:
WRT X-Max just design not to exceed it for your chosen peak SPL. You can for brief periods as there is an X-Mec which you avoid exceeding at all cost. Just use common sense. You should have displacement vs frequency and power in your sims as a reference. The obvious is you need to know again what the target peak SPL requirement is.
Rob 🙂
https://mehlau.net/audio/spl/
Last edited:
There's no obvious answer to this question. It comes down to how you intend to use the driver. If you model with excessive power levels at low frequencies, you're definitely going to exceed Xmax in lots of conditions. A 9 inch woofer is only going to do so much. At low frequencies, it's mostly about volume displacement in the end. Each octave you go down requires 4x the volume displacement to maintain the same SPL level. So at Xmax that driver might be able to run you out of the room at 200 Hz, but it's going to be much more polite at 20 Hz.
For normal home audio, there's typically a handoff in the midbass or bass region where you switch from excursion limiting output to thermal considerations limiting output. So at higher frequencies the wattage limit of the voicecoil comes into play more directly, but at lower frequencies it's more to do with the enclosure and excursion.
To me, that woofer looks best for what you originally described: a 3-way. The impedance and frequency response discontinuities starting at 800 Hz would give me some concern from a general design standpoint. Maybe you can stretch the usable range to 1500 Hz, but it's something I'd listen to and measure before committing to. It comes down to what you find important from a design standpoint.
https://www.parts-express.com/pedocs/specs/297-086--morel-caw-938-spec-sheet.pdf
A 9-inch woofer also isn't that large in the grand scheme of things, so it's unlikely to be "loud" at 20 Hz. My idea of loud and your idea of loud may be two completely different things though. Do you have experience with any speakers you think were loud enough for your listening style? That might help set a reasonable target for what SPL you want. Erin's Audio Corner has quite a few speaker tests up that include distortion (86 and 96 dB at 1 meter) and dynamic range testing (76, 86, 96, 102 dB at 1 meter).
Last edited:
My main concern is around the woofer being safe from mechanical damage from over excursion and also have low compression as to keep dynamic. I will be using them for both music and home theatre as I said previously. Which leads me to the concern of not being able to play them loud enough before distortion starts to affect the low from mechanical limits. I don't mind if it only plays to 60Hz as I will definitely be building some subs in the future to good bass extension.
Yes, that is exactly what I'm going for but don't have a calibrated SPL meter so can't tell what is "loud enough" for myself. So I do think following what reference levels for movies, such as Dolby and THX, including the headroom, is a good baseline as this is likely the loudest it would be comfortable to play would be.
From what it looks like is an average level of 85dB at the listening position with 20dB headroom, taking it to 105dB. Below is a chart of SPL loss due to distance to the listening position.
| Listening Distance | SPL Loss | Required Speaker Output at 1m |
|---|---|---|
| 2m | -6 dB | 91 dB |
| 3m | -9.5 dB | 94.5 dB |
| 4m | -12 dB | 97 dB |
| 5m | -14 dB | 99 dB |
| 6m | -15.5 dB | 100.5 dB |
| 7m | -17 dB | 102 dB |
| 8m | -18 dB | 103 dB |
| 9m | -19 dB | 104 dB |
| 10m | -20 dB | 105 dB |
So I'm wondering if this is the main this I should be simulating to. Or am I needing to simulate everything to the peak/headroom of 105dB at the listening position? Here is the chart for 105dB at listening position.
| Distance (m) | SPL Loss (dB) | Required Speaker SPL @ 1m (dB) |
|---|---|---|
| 2m | -6.02 | 111.0 |
| 3m | -9.54 | 114.5 |
| 4m | -12.04 | 117.0 |
| 5m | -13.98 | 118.98 |
| 6m | -15.56 | 120.56 |
| 7m | -16.90 | 121.9 |
| 8m | -18.06 | 123.1 |
| 9m | -19.08 | 124.1 |
| 10m | -20.00 | 125.0 |
This however looks unrealistic to achieve even at 2m for even the caw 938 which does not hit Xmax at its RMS of 150W, with a 1st order HPF at 250Hz, and only achieves about 107dB. In the headroom aspect does it only need to be under peak power ratings and X_mech?
Cheers!
Attachments
Most of the time, model for the worst case output level, not the power input rating. Since you are targeting home theater, shoot for 105dB, and adjust the wattage input to reflect that. Then fiddle with the enclosure and see where it models safely at those benchmarks. I do not feel you'll need 3 cubic feet, likely more like 1.5 cubes to make it happy. You could buy another pair of woofers and likely use both per cabinet in that 3 cubes. This will also help meet your output requirements as I'm uncertain you can meet them without a total quad of woofers.
938 seems to have a Qts of .52
So the box would be on the larger size.
Whatever volume it takes to get .7 Qtc
Not a very strong magnet, so it wants a big box
The Fs of the speaker seems pretty decent around 29 Hz
Assuming without doing sim, QB4 is what it wants, which will tune slightly below 29 Hz 3 or 4 Hz
That is about it. Speaker is what it is. Not much to overthink
All vented enclosures will start to unload below the port tuning.
Xmax is 10% distortion, so most would listen around 3 to 6% distortion.
With movies there is big transients for sound effects. Is what it is.
Assume you listen around 86 to 88 dB and 92 to 96 dB if you like loud.
Or if the movie sound jumps all over the place from low dialog to loud sound effects.
So around 5 to 8 volts RMS with 10 to 16 volts for loud parts in the movies.
So as with most 85 dB speakers your listening at 3 to 4% distortion in low frequency and with big transients
It shoots close to 7% to 10%. Why there is plenty of towers out there with 2x woofers to tame that down.
Assuming the front plate is typical 6mm to 7mm the magnet is only strong enough to use 4.2 mm of that
before it reaches 10% distortion. Or 70% of available magnet Bl
Something with a stronger magnet will get lower into .3 Qts showing more magnet control.
So the box would be on the larger size.
Whatever volume it takes to get .7 Qtc
Not a very strong magnet, so it wants a big box
The Fs of the speaker seems pretty decent around 29 Hz
Assuming without doing sim, QB4 is what it wants, which will tune slightly below 29 Hz 3 or 4 Hz
That is about it. Speaker is what it is. Not much to overthink
All vented enclosures will start to unload below the port tuning.
Xmax is 10% distortion, so most would listen around 3 to 6% distortion.
With movies there is big transients for sound effects. Is what it is.
Assume you listen around 86 to 88 dB and 92 to 96 dB if you like loud.
Or if the movie sound jumps all over the place from low dialog to loud sound effects.
So around 5 to 8 volts RMS with 10 to 16 volts for loud parts in the movies.
So as with most 85 dB speakers your listening at 3 to 4% distortion in low frequency and with big transients
It shoots close to 7% to 10%. Why there is plenty of towers out there with 2x woofers to tame that down.
Assuming the front plate is typical 6mm to 7mm the magnet is only strong enough to use 4.2 mm of that
before it reaches 10% distortion. Or 70% of available magnet Bl
Something with a stronger magnet will get lower into .3 Qts showing more magnet control.
In terms of excursion required, it depends hugely on the content you like to play and the levels you want to play it at. Most music has little energy below 40Hz, and high level signals in the bottom octave are extremely rare. But home theatre is a different game altogether, with high level low bass (and subsonic) content very common.
Hence inefficient compact ported speakers, with woofers which will exceed xmax at 20 Hz with just a couple of watts, often play loud and survive perfectly well for years when used for music only. Feed them HT content at the same levels, though, and you'll destroy them quickly.
So the question is whether your woofers, in a box big enough to give your desired Q, will cope with HT content at your preferred levels without an HPF. The answer really depends on your preferred levels. Vituixcad should tell you what level can you get out of them (across the passband) when they are reaching linear xmax at 20 Hz.
Hence inefficient compact ported speakers, with woofers which will exceed xmax at 20 Hz with just a couple of watts, often play loud and survive perfectly well for years when used for music only. Feed them HT content at the same levels, though, and you'll destroy them quickly.
So the question is whether your woofers, in a box big enough to give your desired Q, will cope with HT content at your preferred levels without an HPF. The answer really depends on your preferred levels. Vituixcad should tell you what level can you get out of them (across the passband) when they are reaching linear xmax at 20 Hz.
Hmm. Just tried a quick model of the CAW938 in Basta. In a 60 to 70 litre box, hitting xmax at 40 Hz and below with 8 V, passband SPL in the low 90s. Not great (for HT).
(Edit: they model quite nicely in a much smaller box - even 10 litres - with a series 500 uF cap. The -6dB is about 60 Hz, and 10dB more is possible before they run out of excursion.)
(Edit: they model quite nicely in a much smaller box - even 10 litres - with a series 500 uF cap. The -6dB is about 60 Hz, and 10dB more is possible before they run out of excursion.)
Last edited:
10 liters for a .5 Qts speaker.
I dont think so, be rather boomy and slow. It is a weak magnet.
For midbass on a high pass Qtc be acceaptable no higher than .8 Qtc
Extremely generic speaker, why they bothered with a cast frame beats me.
Same old same old.
Probably wants 2.5 cubic feet roughly 75 liters to be .7 Qtc
Smallest tolerable probably be 1.5 cubic feet or 45 liters.
likely have the typical high Qts peak.
Tradeoff between the 2 likely be whatever the heck a BB4 alignment is.
Whatever volume it needs for .8 Qtc, then tune at Fs. All you get
Magnet has a hard time doing its job, the volume will be large regardless.
Excursion will be what it will be. Making boxes too small doesnt fix much.
CB4 or BB4 push auto calculate, your done. No magic science
I dont think so, be rather boomy and slow. It is a weak magnet.
For midbass on a high pass Qtc be acceaptable no higher than .8 Qtc
Extremely generic speaker, why they bothered with a cast frame beats me.
Same old same old.
Probably wants 2.5 cubic feet roughly 75 liters to be .7 Qtc
Smallest tolerable probably be 1.5 cubic feet or 45 liters.
likely have the typical high Qts peak.
Tradeoff between the 2 likely be whatever the heck a BB4 alignment is.
Whatever volume it needs for .8 Qtc, then tune at Fs. All you get
Magnet has a hard time doing its job, the volume will be large regardless.
Excursion will be what it will be. Making boxes too small doesnt fix much.
CB4 or BB4 push auto calculate, your done. No magic science
Yes. Without the high-pass cap, certainly. Q would be 1.2.
With the cap, I don't know. The model looks promising. (I've not tried a setup like that, though.)
Well, that's problem the OP asked about. Efficiency is modest, and in a large sealed box you're going to go over xmax quite easily with HT.