I have no idea what you are talking about here. Perhaps the general idea that mathematically, a change inside an integration boundary cannot be known, even though it causes a large change outside? In the case of a foam core speaker enclosure that vibrates, the effect can be estimated based on the difference between total output vs output from driver and vent.
Nice result Phil. I agree that rear facing vent seems best in room boundaries.
An externally hosted image should be here but it was not working when we last tested it.
Which size vent did you end up with here in these sims?
Sorry for mixing or going OT, no ,well, it's still in thread
I meant that everything should be done with the intention of having the maximal intensity perceived by the listener. Talking about the speakers and not the entire system or room, or even about the disc content, the goals are...
wait: what speaker ? We were talking about procedures .
The major issue is the 'covering effect' that happens during reproduction.
It's the sound originated by, somehow, somewhere that overimposes on the
original ( which is not known) one; we could know the amount of it, or verify
how much it can differ ....
So it's not that easy; sorry again to be so pedantic
We're not talking about a ceiling speaker at the supermarket, you know...those diabolic things that inject in your brain those stupid jingles.
I had in mind the answer why it could be nice to hear those things that you made in the cornu thread...but, yes, if referring to a 10cm driver left alone to guvern under avverse conditions, in storm and rain
one thing i don't understand, how did you come to a 40 in line length? My calculation wd be much longer
Practical consideration to get it higher up to ear level and not too long that it will tip over easily. Also, I have built them at 40 in before and it seems to work. It turns out to be about same as what others who designed it with MJK ended up using. If you make it too long, that is close to quarter wave length for fb, the mass loading will drive fb too low, below what driver may be able to do.
bit confused, the line length or cabinet 40 inches in height (or both)? I would always 'fold' and the baffle can be angled by a degree or 2 if needed
I assumed a line frequency length just above the resonant frequency value of said drive unit was correct, same as a standard TL but maybe around 10% extra as its tapered/flared
I assumed a line frequency length just above the resonant frequency value of said drive unit was correct, same as a standard TL but maybe around 10% extra as its tapered/flared
Last edited:
You can fold but if you tune length close to just above fs of driver you do not need to mass load. The mass loading achieves both reduced line length and provides more filtering of output peaks and valleys by low pass filtering output from line. The length as I have said is mostly based on seeing other MLTL designs and their lengths.
It is still quarter wave at tuning freq but the tuning freq is forced to a lower value than line would support if open ended rather than constricted. That is why it is called mass loading - like adding weight to a cantilever to reduce its natural resonant freq. You won't know what this freq is unless you model it in MJK or HornResp or other sim program. It does not correspond to any freq based on physical length alone.
While I would not straight-up commit a RoT design to expensive plywood, I think that at the least, it can be a good starting point assuming the driver has appropriate parameters for the enclosure. Drivers I usually apply the MLTL RoT to will be 0.40>Qts>0.30, with medium Fs relative to size, and will usually not need tweaks so drastic as to completely void the initial RoT geometry, once the numbers are plugged in Hornresp. Where HR falls short is its inability to set Z_vent and stuffing.
IG
IG
I agree that a RoT is to use drivers with moderate Qts, but I have seen several MLTL's using low Qts drivers (TB W8-1772 for example). When I put the parameters for a low Qts 8 in driver in WinISD, the default tuning freq is very high - higher than the driver's fs. So it would seem that the low Qts driver in a MLTL will not have as low bass extension as a higher Qts variant with same fs. The cabinet volume is smaller though. So can the low Qts driver be as good as a higher Qts driver in a MLTL ?
I guess that only looking at Qts is a Rot of its own. Altec/GPA 416 drivers can and have been made into MLTL, but I'd be more suspicious of the same done with a Fostex FE206En for example, which really wants more of a horn.
Take this max flat reflex simulation for the low Qts driver and tune it to ~Fs or a hair higher. You'll likely end-up with something of a bass-shelf or a slight peak at Fb with a valley right above it. IMO, the MLTL should fill-in this depression and allow for flatter response, as the mass-loading will kick up system Q a bit, analog to adding mass to the cone. If any of this is incorrect, feel free to correct, as I don't want to mislead future readers.
IG
Altec/GPA 416 drivers can and have been made into MLTL, but I'd be more suspicious of the same done with a Fostex FE206En for example, which really wants more of a horn.Take this max flat reflex simulation for the low Qts driver and tune it to ~Fs or a hair higher. You'll likely end-up with something of a bass-shelf or a slight peak at Fb with a valley right above it. IMO, the MLTL should fill-in this depression and allow for flatter response, as the mass-loading will kick up system Q a bit, analog to adding mass to the cone. If any of this is incorrect, feel free to correct, as I don't want to mislead future readers.
IG
IG,
I believe you are right about how the MLTL fills in the depression as I have seen very flat shaped looking bass shelf in measurments of the output from the vent even though the BR sim shows a peak and a dip. This is where the MJK sim will clearly show what happens in a MLTL with stuffing.
I want to use AkaBak to simulate a MLTL with stuffing - waiting for Don Hills to develop a MLTL model in AkaBak that has an analog of damping via adjustment of the viscosity of the air in the sections with stuffing. This will take some adjustment and comparison with either measurements or models that have stuffing to get an empirical correlation.
I believe you are right about how the MLTL fills in the depression as I have seen very flat shaped looking bass shelf in measurments of the output from the vent even though the BR sim shows a peak and a dip. This is where the MJK sim will clearly show what happens in a MLTL with stuffing.
I want to use AkaBak to simulate a MLTL with stuffing - waiting for Don Hills to develop a MLTL model in AkaBak that has an analog of damping via adjustment of the viscosity of the air in the sections with stuffing. This will take some adjustment and comparison with either measurements or models that have stuffing to get an empirical correlation.
Quad driver AMLTL with W3-881si
I have been tinkering around with a multi-driver MLTL design for some time.
Originally thinking that the Vifa TC9FD would be used.
I recently received a PM from a member to come up with a design for a quad driver MLTL for the Tang Band W3-881si blow-out sale speaker (which unfortunately, are no longer available from PE for $10 ea). So I am using the AMLTL method to design this.
Here's the AMLTL design:
Volume = 28.3 liters (1 cu ft), tuning freq = 66 Hz
Baffle width = 5.0 in
Depth = 8.5 in
Height = 40.0 in
Vent = 5.0 in wide x 2.0 in tall x 3.7 in long (shelf vent on bottom)
alternatively,
Qnty 3 x SCH 40 1.5 inch PVC pipes cut to 2.64 in long all mounted near the bottom.
Mount 4 drivers with drivers middle between drivers 2 & 3 at 13.3 in from top. Keep driver c-t-c spacing as tight as possible.
Use stuffing in top 2/3rds or 26 in, use net or glue to keep from falling down.
This design achieves some nice efficiency rivaling some much more expensive drivers: wired for 8 ohm, 1 watt should produce 92 dB at 1 m. The MLTL effect should bring bass extension easily down to 50 Hz.
This design can have 4 drivers facing forward or 2 forward 2 back for bipole push-push.
If the cabinet is elevated with a false bottom and feet made by cutouts on the edges, the 3 PVC pipes can be mounted down firing externally to provide even bass output.
I have been tinkering around with a multi-driver MLTL design for some time.
Originally thinking that the Vifa TC9FD would be used.
I recently received a PM from a member to come up with a design for a quad driver MLTL for the Tang Band W3-881si blow-out sale speaker (which unfortunately, are no longer available from PE for $10 ea). So I am using the AMLTL method to design this.
Here's the AMLTL design:
Volume = 28.3 liters (1 cu ft), tuning freq = 66 Hz
Baffle width = 5.0 in
Depth = 8.5 in
Height = 40.0 in
Vent = 5.0 in wide x 2.0 in tall x 3.7 in long (shelf vent on bottom)
alternatively,
Qnty 3 x SCH 40 1.5 inch PVC pipes cut to 2.64 in long all mounted near the bottom.
Mount 4 drivers with drivers middle between drivers 2 & 3 at 13.3 in from top. Keep driver c-t-c spacing as tight as possible.
Use stuffing in top 2/3rds or 26 in, use net or glue to keep from falling down.
This design achieves some nice efficiency rivaling some much more expensive drivers: wired for 8 ohm, 1 watt should produce 92 dB at 1 m. The MLTL effect should bring bass extension easily down to 50 Hz.
This design can have 4 drivers facing forward or 2 forward 2 back for bipole push-push.
If the cabinet is elevated with a false bottom and feet made by cutouts on the edges, the 3 PVC pipes can be mounted down firing externally to provide even bass output.