[michaelddd]

I'm designing a ported box for a UXL-18 and would appreciate a little help. I'm aiming for a large box (at least 12CF net) with a slot port, tuned to 15Hz. I am modeling with 2,000 watts.

What 1st port resonance am I aiming for? My current design has a port of 2.0" x 22.50" @ 39.2 inches long showing a 1st PR of 172Hz. I have no idea if this is great or it stinks. I have noticed that the higher I go with the port height, the lower the resonance drops. About the only thing I do know is that you don't want a 1st PR within an octave of the highest freq you expect the box to play...at least I think I'm remembering that correctly...

Would appreciate a little direction. Thanks.

 

[kiwijunglist]

Is the port velocity OK?

 

[Scott Simonian]

In general it's best to keep your first port resonance at least one octave above your intended low pass filter frequency.

For example, if you are going to use an 80hz LPF then you want your first port resonance no lower than 160hz.

Higher is always better but usually this is compromise one has to make when designing a resonant system. You can raise the resonance to a higher frequency at the detriment of others. As with all speakers and subwoofers, it's all about finding balance in design.


In your design, a number of 172hz is close enough.

 

[notnyt]

With my large enclosures and huge ports, the first port resonance is at 108hz. Common design practices say this is bad, and while generally it's good to keep it higher up, I haven't been able to hear it. It's visible a little in a spectrogram, but that's about it, and it doesn't really seem any worse than the other room effects. Your resonance seems like it might be high at the expense of port velocity. What is it peaking at at those power levels?

 

[LTD02]

definitely on the right track. suggest you kick it up a hair, 2.5" tall x 44" long to reduce air speed just a little more. also, you get a little effective port extension with a slot port.

 

[michaelddd]

Thank you, everyone. I would've replied earlier but these forums are blocked at work. I appreciate the hints and direction. I will continue to hone my WinISD skills tonight. I will post some graphs too and if you can, please comment.


Mike D

 

[michaelddd]

OK, I've been working with WinISD for about 90 mins now. This can be frustrating! You make the box size bigger to get port size up but the 1stPR goes down and...you know the drill.

So far I have a 15CF box (will be net, after driver and bracing displacement), with a 15Hz tuning. Slot port is 2.50 x 18.0 x 30.25. 1st PR is 223Hz (this is good). And my cone excursion breaks max by just 1mm at 2Kw. Also good. However, port velocity hits 17m/s at just 21Hz and goes way off the chart from there. We need a "bang head on keyboard" icon.

It's like I have to choose b/t dangerous cone excursion and nice port velocity, or safe excursion and lousy PV. There is no free lunch, I know...but I'm sure this is not the best I can get. I'm willing to go to a bigger box if needed...that's not a problem.

The box will be well-braced, will be lined with mattress topper...all the standard good stuff.

Would appreciate some more advice. Thank you.

 

[smcmillan2]

With my large enclosures and huge ports, the first port resonance is at 108hz. Common design practices say this is bad, and while generally it's good to keep it higher up, I haven't been able to hear it. It's visible a little in a spectrogram, but that's about it, and it doesn't really seem any worse than the other room effects. Your resonance seems like it might be high at the expense of port velocity. What is it peaking at at those power levels?

I don't mean to thread-jack, but what in the above spectrogram indicates the first port resonance? Is it the first separation from the yellow mass?

Trying to learn, thanks for any help.

 

[Jay1]

OK, I've been working with WinISD for about 90 mins now. This can be frustrating! You make the box size bigger to get port size up but the 1stPR goes down and...you know the drill.

So far I have a 15CF box (will be net, after driver and bracing displacement), with a 15Hz tuning. Slot port is 2.50 x 18.0 x 30.25. 1st PR is 223Hz (this is good). And my cone excursion breaks max by just 1mm at 2Kw. Also good. However, port velocity hits 17m/s at just 21Hz and goes way off the chart from there. We need a "bang head on keyboard" icon.

It's like I have to choose b/t dangerous cone excursion and nice port velocity, or safe excursion and lousy PV. There is no free lunch, I know...but I'm sure this is not the best I can get. I'm willing to go to a bigger box if needed...that's not a problem.

The box will be well-braced, will be lined with mattress topper...all the standard good stuff.

Would appreciate some more advice. Thank you.

Well since you are running a ported sub the first thing you need to do is add a high pass filter to the model and in real life.

 

[notnyt]

I don't mean to thread-jack, but what in the above spectrogram indicates the first port resonance? Is it the first separation from the yellow mass?

Trying to learn, thanks for any help.

 

[notnyt]

And here's another measurement, with eq, and crossed to the mains. This captures a room mode around 70hz, but you can't even see a hint of the port resonance.

 

[michaelddd]

Highpass filter, fair enough. I wasn't sure what freq to set it at, but I know it's got to be below the tuning frequency, and I am pretty sure it should be steep. So here is a 12Hz, 24db/oct highpass. Port velocity still stinks and cone excursion didn't change (so no pic of excursion).


ps
No worries about thread jacking; I'm learning too. Ask whatever questions you want; we all benefit.

 

[michaelddd]

And here's another measurement, with eq, and crossed to the mains. This captures a room mode around 70hz, but you can't even see a hint of the port resonance.

What made the squiggle representing the port resonance go away?

 

[Jay1]

Highpass filter, fair enough. I wasn't sure what freq to set it at, but I know it's got to be below the tuning frequency, and I am pretty sure it should be steep. So here is a 12Hz, 24db/oct highpass. Port velocity still stinks and cone excursion didn't change (so no pic of excursion).


ps
No worries about thread jacking; I'm learning too. Ask whatever questions you want; we all benefit.

What amp are you using? Do you have a miniDSP? If 12hz isnt cleaning up the cone excursion and port velocity move it up higher in frequency

 

[notnyt]

What made the squiggle representing the port resonance go away?

Likely running the sweep at sane output levels.

 

[LTD02]

2.5 x 18 is a little small for the uxl (port velocity will be too high).


I'd kick it up to about 4x18x44


but not go smaller than about 3.5 tall unless you like "wind effect" ;-)

 

[michaelddd]

What amp are you using? Do you have a miniDSP? If 12hz isnt cleaning up the cone excursion and port velocity move it up higher in frequency

Amp is to be determined. I am modeling with 2Kw because that's the wattage I've read here on AVS to use. I am aware the UXL-18 is rated at 1,200 RMS and have that set in WinISD as the driver's Pe.


I will be running two of these boxes. I am considering a Crown XTi2002 bridged on each driver. That would give me 2,000 watts @ 4 ohms. I'm not into running sine waves at W.F.O., so I'm not worried about blowing the driver. The MiniDSP part is also to be determined. If I can do what is needed with the amp, I won't need the mini-DSP. BUT, I know some of what the MiniDSP can do, such as PEQ and multi-band EQ, etc. I'd buy one if it made that much of a difference. I come from the days of measuring from speaker to ears with a tape measure and sitting there with my gel filters over my eyes to tweak the display...so I'm not new to the tweaking part...just new to REW and DSP, etc.


I have REW 5.1 loaded and bought a UMIK-1 and I'm willing to put the time in to tweak. I know next to nothing about how to use it but am willing to learn. I will state again that I won't actually be in the house for several months. My coworkers think I'm nuts "diving this far in, so far in advance" but I want equipment ready to install and set up within days of moving in.

 

[michaelddd]

2.5 x 18 is a little small for the uxl (port velocity will be too high).


I'd kick it up to about 4x18x44


but not go smaller than about 3.5 tall unless you like "wind effect" ;-)

Thanks very much. How do I force a longer port length in WinISD? The length is auto-calculated and is grayed-out....I can't change it. 4 x 18 x 50.25 (WinISD set the length) helps with port velocity (max is 29...was 40-something!) but now 1st PR is 134Hz...isn't that a little low? I'm planning on a 80Hz HP for the mains, so maybe not. THANK YOU for the help, LTD02.

 

[smcmillan2]

Thanks very much. How do I force a longer port length in WinISD? The length is auto-calculated and is grayed-out....I can't change it. 4 x 18 x 50.25 (WinISD set the length) helps with port velocity (max is 29...was 40-something!) but now 1st PR is 134Hz...isn't that a little low? I'm planning on a 80Hz HP for the mains, so maybe not. THANK YOU for the help, LTD02.

Was wondering how you were getting such different measurements, just noticed in your screenshots that you aren't using the latest version of WinISD. Get the Alpha Pro version here: http://www.linearteam.dk/default.aspx?pageid=winisdpro

 

[smcmillan2]

More to check on my setup...

But thanks notnyt, I do appreciate the info!