I'm thinking about reusing 4 of these speakers that I currently have in some Keystone subwoofers since I'm doing shows in smaller clubs where front loaded could be a better fit. could someone please tell me if these figures look right? not much of a port according to the subbox.pro designer.
https://subbox.pro/en/b/aPuzX0mJS
https://www.bcspeakers.com/en/products/lf-driver/18-0/4/18tbw100-4
https://subbox.pro/en/b/aPuzX0mJS
https://www.bcspeakers.com/en/products/lf-driver/18-0/4/18tbw100-4
That program must be calculating the port dimensions based on the small signal specs, it's not even close to being big enough to keep port air velocity within reason at higher power levels. WinISD suggests a pair of slot ports with dimensions 4"x11"x27.4" for 17m/s air velocity at 1000w.
You can go with 6,357 ft3 (180 Liters) and a port of 400 cm2 (62sq.in.) and a lenght of 13,74 inch (35cm), all taken with the specs from B&C...if you got them and be able to measure them, it is possible to get an more accurate simulation!
It should be noted, that for PA aplications you should go with at least 4 or even 3 times the SD of the driver, for the port area, so in this case 300cm2 respective 400 cm2 or in your case 62 sq. in. Also, keep in mind that, when working with square ports the lenght of the port, gets "virtually" longer at the same lenght, like a round one. What this means is, that with a square port you need an other lenght of port, which you would need when using a round one. This has to be measured, best with a prototype housing where you could adjust the port until it has reached the wanted tuning frequency. (measured with an impedance measurement device light DATS...etc) . Or you go with round ones, but still measuring should be made!
Greets Swany
It should be noted, that for PA aplications you should go with at least 4 or even 3 times the SD of the driver, for the port area, so in this case 300cm2 respective 400 cm2 or in your case 62 sq. in. Also, keep in mind that, when working with square ports the lenght of the port, gets "virtually" longer at the same lenght, like a round one. What this means is, that with a square port you need an other lenght of port, which you would need when using a round one. This has to be measured, best with a prototype housing where you could adjust the port until it has reached the wanted tuning frequency. (measured with an impedance measurement device light DATS...etc) . Or you go with round ones, but still measuring should be made!
Greets Swany
I built some reflex boxes with the 18TBW100-4, 6.5 c.f. @ 33hz.
Single slot port at 4"x21" gives 18m/s port velocity at 2000W (8% over Xmax)
First port resonance at 279Hz, XO at 80-100hz.
They sound very nice.
Your port could be larger in area, but also watch the aspect ratio and try to keep it under 1:8 to avoid turbulence/compression/etc.
WinISD port length is much too long for a slot port that shares cabinet walls, I cut the partition down and measured impedance until I got the right tuning frequency.
Single slot port at 4"x21" gives 18m/s port velocity at 2000W (8% over Xmax)
First port resonance at 279Hz, XO at 80-100hz.
They sound very nice.
Your port could be larger in area, but also watch the aspect ratio and try to keep it under 1:8 to avoid turbulence/compression/etc.
WinISD port length is much too long for a slot port that shares cabinet walls, I cut the partition down and measured impedance until I got the right tuning frequency.
My worry about the port is that if I put a 3000 watt peak thru the speaker I don't want it to fry. In the keystone cabinets I currently have them in I have cranked them up to 3000 watts and they just sound great. I have heard of people with the wrong port size and the speaker doesn't hold up too well.
As I understand it, the port size doesn't relate directly to the power handling of the system. That's determined by the cabinet volume and tuning frequency (volume of air inside the port, not cross sectional area of port). So I find a box volume/tuning frequency that gives me the output I need while hitting Xmax at a peak burst. Then I look at the port resonance and air velocity and adjust to get the resonance far enough outside of the woofers passband while keeping the velocity under 20m/s (some say 17m/s). Iterate all these factors until you get an optimized system for your usage. A smaller port will need to be longer for a given Fb, as it gets longer the port resonance gets lower and the air velocity get higher and that's not good, it will create audible distortions (even if the Fb and power handling doesn't change)
This is just how I do it using WinISD, which will show you what parameters are directly linked to each other and which aren't. Playing around with WinISD was way more educational for me than reading about how all these parameters interact. The only thing that seems to be broken is the port length correction, so I leave some wiggle room in the cabinet design and test/adjust with a prototype and impedance sweeps.
As an aside, if you are using DSP on your sub channel you can "force flat response" and add HP/LP filters in WinISD, then you can design for an output target curve instead of a raw box output curve. For example, shaving off output with a HP filter below the tuning frequency can work wonders for keeping excursion below Xmax at peak output.
TL;DR use a good sim application to learn the theory, it's baked in.
This is just how I do it using WinISD, which will show you what parameters are directly linked to each other and which aren't. Playing around with WinISD was way more educational for me than reading about how all these parameters interact. The only thing that seems to be broken is the port length correction, so I leave some wiggle room in the cabinet design and test/adjust with a prototype and impedance sweeps.
As an aside, if you are using DSP on your sub channel you can "force flat response" and add HP/LP filters in WinISD, then you can design for an output target curve instead of a raw box output curve. For example, shaving off output with a HP filter below the tuning frequency can work wonders for keeping excursion below Xmax at peak output.
TL;DR use a good sim application to learn the theory, it's baked in.