Can someone model a sub for me? ported 18 inch "Johnny sub" (willing to tip)

Hello

I'm building two "Johnny sub" size sub boxes

Outside dimensions are 32x24x24, using a double front baffle, all panels are 3/4.
Port is 2.5 inches tall x 22.5 inches wide with 2 3/4 beams.

This is the driver I have, Eminence NSW4018-8
TS and specs are here: https://eminence.com/products/nsw4018_8#specifications

I also have this driver which is very similar, I'm hoping to match gain and use them both in the same room
LaVoce SAF184.03 18"
TS and specs are here: https://www.parts-express.com/LaVoce-SAF184.03-18-Subwoofer-8-Ohm-293-730?quantity=1

How long should the port be for each sub for the best response?
The generic plans I found online specify back port panel and the port bottom panel to both be 20.75" x 22.5

I will be running both into a MiniDSP with Dirac after, if that changes anything

I ran a couple of simulations using WinISD for a sealed sub, but ported is just out of my league

Thanks guys
 

Attachments

  • sub.png
    sub.png
    27.8 KB · Views: 130
  • sub2.png
    sub2.png
    19.5 KB · Views: 93
WinIsd wont calculate a slot port as shown.

It uses the usual math but assumes
non shared walls.
So end correction in Winisd only ranges .732 to .835
which they label as no flare, single flare, dual flare
blah blah. Works fine for normal everyday ports
round or square.

Slot port with 3 shared walls gets up above the decimal
port for end correction. Which makes the port shorter
with 2.227 end correction .
Which is why people use them. Makes the port shorter
easier to fit in a box.

Speaker is what a speaker is.
The compliance wants a certain volume.
And you tune few Hz above or below
Fs of the speaker. Nothing more magical
for more bass.
Run a QB3 or BB4 alignment and that is what it does.

To calculate a slot port, use a program that can
actually adjust end factor correction to 2.227
and that is the length needed.
Tuning is 3 to 7 Hz above or below Fs
Depends on alignment.
BB4 will tune to Fs
far as I remember normal QB3 usually
few Hz below Fs. and no it wont be a
" Flat" line or response in transfer function.
Just a real QB3 alignment with live sound driver
will be LARGE
 
How long should the port be for each sub for the best response?
Depends on what you decide "best response" is.
A longer port will tune the box lower (Fb) with reduced output before reaching the linear output limits of the driver, a shorter port will result in a higher Fb, more upper output but not usable as low.

The port you depicted is around 42" long, which puts the Fb somewhere in 20 Hz range, well below the Fs of the drivers you are considering using, so the response will have considerably less output than if Fb was near Fs.
 
For convential music you should be fine with F3 of 35 Hz. You can build the cabinet with the port dividers and only cover a part of it from the inside. If the tuning is too high, you can still make the port longer by covering the divider through the opening for the driver. Quite easy to work inside such a large box with glue and a few screws.
If you measure your actual drivers TSP you may get quite near to a simulation, but measuring the drivers impedance inside the box show's the real tune.
 
@weltersys Thanks for the reply. How long should I make the ports? These subs will serve 90% for music, I don't need to be tuned to 20hz
A good portion of the music I listen to has response down into the 20 Hz range, so my preference for home use would be an Fb that low.
If I were using the cabinets for PA use, I'd trade the LF response for output and use an Fb around 37Hz.
In either case, a BW24HP filter would be used a few Hz below the Fb to avoid cone flap and wasted power below Fb.

This port calculator will get you in the length ballpark for whatever Fb you may decide on:
https://www.the12volt.com/caraudio/boxcalcs.asp#porsq

Remember to deduct the
Driver Volume Displaced0.205 cu.ft. / 5.81 liters
and volume occupied by the port from the cabinet interior volume.
 
WinIsd wont calculate a slot port as shown.

It uses the usual math but assumes
non shared walls.
So end correction in Winisd only ranges .732 to .835
which they label as no flare, single flare, dual flare
blah blah. Works fine for normal everyday ports
round or square.

Slot port with 3 shared walls gets up above the decimal
port for end correction. Which makes the port shorter
with 2.227 end correction .
Which is why people use them. Makes the port shorter
easier to fit in a box.

Speaker is what a speaker is.
The compliance wants a certain volume.
And you tune few Hz above or below
Fs of the speaker. Nothing more magical
for more bass.
Run a QB3 or BB4 alignment and that is what it does.

To calculate a slot port, use a program that can
actually adjust end factor correction to 2.227
and that is the length needed.
Tuning is 3 to 7 Hz above or below Fs
Depends on alignment.
BB4 will tune to Fs
far as I remember normal QB3 usually
few Hz below Fs. and no it wont be a
" Flat" line or response in transfer function.
Just a real QB3 alignment with live sound driver
will be LARGE
What software allows me to model end factor correction to 2.227?
 
Don't worry about custom end correction factors. I haven't found them to be entirely accurate. These custom factors are about dimwits chasing decimals that don't exist. T/S theory isn't accurate at high signal levels anyway.

Just to make it round numbers, call your box 260 liters (without the port 280)
For the Eminence:
This is a PA driver, Ideal QB3 box would be 107liters tuned to 38Hz with 3x6"dia ports about a 34" long each,
For your 260 liter box, tuning to 24Hz, ideal port would be about 3x6" ports about 36" long..

For the LaVoce:
Ideal QB3 is 100 Liters tined to 36Hz with 2x 6" ports, each 29" long
For the 260 liter box, tune it to 24 Hz with 3x 6 " ports about 36" long.

For your 21"X2.5" shelf port to tune to 24 Hz in either box, make it about 20" long, but you will have significant chuffing at high power levels. Better to make it 5" x 21" and 40" long.
 
Last edited:
No, if you double the area, you double the length. Basically think of the port as a solid piston of air compressing the air in the box. A larger piston will move more air for each inch it moves in or out. The port actually compresses the air in the box just like the driver does. Because the doubled displacement will make the air in the box look twice as stiff, the mass of air in the vent has to double to maintain the same port resonance frequency.
 
Last edited:
I see. Thanks for taking the time.

I don't think making the port 5 inches is in the cards, the design is pretty final I already have all my wood and mdf planned out and ready to go.

I currently have a "Full marty" sub with the eminence driver in it, which is the exact same design as what I posted above except it's 48" deep instead of 32". I don't even know what the length of the port is, but it's really long and it's also 21x2.5. Tuned to 17hz or something stupid like that. I haven't heard any audible chuffing at all. Also, there's only one, this time I will have two subs, so each sub won't be moving as much air as the one big sub (I would think)

I will try with only the bottom port panel installed (20ish inches long ) and run some tests. If I'm not satisfied I can add some length to the port with a brad nailer and test some more. If I like it, I'll glue the panel in place, if I don't I'll remove it and put it back the way it was.

Thanks guys for the answers!
 
Make sure the inner exit area of the port is the same as the outer. It will probably be a good idea to put a 45 degree piece of wood ( or even better a scoop) in the corner behind the port to ease the transition. The port chuffing is likely at very high power 500-1000W or so. Your port should handle about 250W no problem, and there isn't much content at 24Hz in most music to excite the port.

Here is your slot port with Eminence design at ~1000W:
Eminence 24.png


And below is your huge Eminence design tuned to ~17 with the same port at ~700 W (it handles less power due to larger enclosure and lower tuning)
Eminence Big 17.png

The port velocity should be under 17m/s for best results, and yours is over 30m/s (red curve in lower right).

FWIW, here is the Eminence the way it could be used in a PA application 150L, tuned to 45 This is at 1600W in: It would have a sharp highpass below tuning (not shown).
1600W PA.png
 
@Ron E Oh wow. So if I make the port 5 inches tall instead of 2.5, I'm going to sacrifice enclosure area, the tradeoff would be worth it? 250w seems low, I have a lab Gruppen clone connected on 220v, it has like 6000w on tap.

This would mean all the marty subs suffer from this, as they all have the 2.5 inch ports?
 
The difference in volume isn't that much. It's only a box, doesn't cost that much. Don't forget to brace those big panels! The thing with music is that it masks port noises, so you may be fine with the smaller ports. If you play test tones you will probably hear it. Run with what you got if you haven't had problems previously.

I think a speaker with a 30m/s port velocity at full power should have big port flares inside and outside unless you know you won't be using it at that power.
 
My maple veneered Baltic birch costs a fortune I can assure you. I'll go text my current sub with test tones see if I can reproduce any chuffing with test tones. In the 3 years I've had it I never noticed anything audible.

Thanks again