Need help with choosing a budget subwoofer - some used options included

[yonyz]

For your sub needing a subsonic filter, I guess it depends on how loud you turn up your sub, and how much content it attempts to play below 20Hz. I doubt you'll need one.

Re: receiver vs. miniDSP vs. DAC, it really depends on whether you want to replace your active monitors with passive bookshelf speakers. Unless you spend a lot of money on a receiver that has pre-amp outputs (like a mid-level Marantz receiver), a receiver wouldn't connect to your current speakers. Re: miniDSP vs. DAC, a miniDSP would certainly give you more flexibility to flatten room response peaks and dial in a house curve, but you would also need a calibrated measurement microphone. Learning to measure your room's response and correct it takes a lot of reading, trial and error, and patience. But, if you enjoy gadgets, the miniDSP can be a fun gadget. Here is a tutorial for Room EQ Wizard, a freeware measurement program whose measurements can be imported into the miniDSP. My advice is for you to try the DAC first. You can change your mind later if you're unable to dial in a smooth blend between your monitors and your sub. But spending $0 is usually better than spending $200 if you can still achieve a satisfactory result with $0.

Rojo, do you have the simulation file for this design? Did you use winISD?
I wanna have a little fun with the dimensions and port design and I got lost in winISD, my resulting graph looks like a mess.

 

[rojo]

Rojo, do you have the simulation file for this design? Did you use winISD?
I wanna have a little fun with the dimensions and port design and I got lost in winISD, my resulting graph looks like a mess.

Yeah, I'll attach a zip file containing both the project file and the driver file. I'm using WinISD version 0.7.0.900. In this model I included a 100Hz low pass filter, and the SPL graph shows predicted response at 200W at 1 meter. F3 is ~24Hz, and still 100dB at 18.5Hz.

Whoa, interesting! Looks like Linearteam is still alive, and they're working on an updated version of WinISD. source

 

[yonyz]

Yeah, I'll attach a zip file containing both the project file and the driver file. I'm using WinISD version 0.7.0.900. In this model I included a 100Hz low pass filter, and the SPL graph shows predicted response at 200W at 1 meter. F3 is ~24Hz, and still 100dB at 18.5Hz.

Whoa, interesting! Looks like Linearteam is still alive, and they're working on an updated version of WinISD. source

Thanks, I finally understood winISD a few hours ago and got the same results as you (by inputting the vent dimensions you mentioned previously and creating the driver file myself). I noticed I could get the F3 point to 22hz by adding 11cm of height to the cabinet, thus increasing the effective volume from 71 to 87 litres.

I also noticed that if we lower the diameter of the port to 1.5 cm or so, we can make the port significantly shorter and remove the need for the elbow.

I wonder what's the magic behind the PB-1000 that allows it to get an F3 point of 19hz with a 10 inch driver. Is the magic sauce hidden in its DSP, or the driver?

 

[rojo]

Thanks, I finally understood winISD a few hours ago and got the same results as you (by inputting the vent dimensions you mentioned previously and creating the driver file myself). I noticed I could get the F3 point to 22hz by adding 11cm of height to the cabinet, thus increasing the "effective" from 71 to 87 litres.

I also noticed that if we lower the diameter of the port to 1.5 cm or so, we can make the port significantly shorter and remove the need for the elbow.

I wonder what's the magic behind the PB-1000 that allows it to get an F3 point of 19hz with a 10 inch driver. Is the magic sauce hidden in its DSP, or the driver?

Don't shrink the port. You should pay attention to the rear vent air velocity. It should stay at or less than 20m/s at 125W. Otherwise you risk getting audible turbulence. If anything, going up to a 3cm port would be better if there's enough space below the amp for another 11cm of vent and still leaving 3cm clearance between the port and vent.

You gain about 1 to 1.3dB for a half octave or so by increasing the box size from 71 to 87 liters. 1.3dB isn't significant enough to justify the added cabinet size and weight in my opinion.

Regarding the PB-1000's number juggling, I'm sure it's a little of both. Its 10" driver probably has a cone with higher mass giving it a lower resonant frequency; its DSP probably decreases output at higher ranges to flatten the overall response; and the driver will have a larger voice coil and longer excursion to handle more power, bringing it back up to respectable output levels.

 

[yonyz]

Don't shrink the port. You should pay attention to the rear vent air velocity. It should stay at or less than 20m/s at 125W. Otherwise you risk getting audible turbulence.

You gain about 1 to 1.3dB for a half octave or so by increasing the box size from 71 to 87 liters. 1.3dB isn't significant enough to justify the added cabinet size and weight in my opinion.

Regarding the PB-1000's number juggling, I'm sure it's a little of both. Its 10" driver probably has a cone with higher mass giving it a lower resonant frequency; its DSP probably decreases output at higher ranges to flatten the overall response; and the driver will have a larger voice coil and longer excursion to handle more power, bringing it back up to respectable output levels.

Where can I see the rear vent air velocity in winISD?

So what kind of driver would be able to deliver the kind of performance that the PB-1000 delivers, and will the miniDSP be able to accomplish what the PB-1000's DSP does?

 

[rojo]

Where can I see the rear vent air velocity in winISD?

So what kind of driver would be able to deliver the kind of performance that the PB-1000 delivers, and will the miniDSP be able to accomplish what the PB-1000's DSP does?

To see the vent air velocity, click "SPL" or "Transfer Function Magnitude" or however the button is labeled in the WinISD toolbar, and select "Rear Port - Air Velocity" from the drop-down menu.

You want to know a comparable ten-inch driver? A Peerless 830452 is one driver I can think of that would give you a lower F3. But the overall output of the Infinity is significantly higher. Put simply, a 12" driver will move more air than a 10". And the much less expensive Infinity driver will still be stronger at 18Hz than the Peerless, despite its higher F3.

The miniDSP can accomplish pretty much whatever curve you imagine. Just remember that changing the shape of the curve requires cutting something. If you want stronger 20Hz, you don't actually boost 20Hz. The response 25Hz and up will have to be cut to meet your 20Hz demand. And regardless of how you shape your sub's response, you'll be limited by the xmax and power handling of the driver. It can only move so far, and can only handle so much wattage without burning itself up.

If you want to see an example before and after of a miniDSP shaped curve, click the CSS SDX12 link in my signature. I close-mic measured the response of the sub without DSP, then measured in-room response after corner loading and reshaping. You can see that without DSP, the F3 of my sealed sub is around 32 - 34Hz; whereas with lots of filtering, my F3 is closer to 14Hz. I have a 500-watt amp and my driver has a 30mm xmax, so I've got enough overhead to make extreme changes to the response like that. A 10-inch sub with a 12.5mm xmax and a 150-watt RMS power limit wouldn't be quite that flexible.

You should also understand that SVS doesn't use off-the-shelf drivers. They custom design the drivers for their subs to their specifications. So you wouldn't be able to buy the same driver they use and recreate their results exactly.

 

[yonyz]

To see the vent air velocity, click "SPL" or "Transfer Function Magnitude" or however the button is labeled in the WinISD toolbar, and select "Rear Port - Air Velocity" from the drop-down menu.

You want to know a comparable ten-inch driver? A Peerless 830452 is one driver I can think of that would give you a lower F3. But the overall output of the Infinity is significantly higher. Put simply, a 12" driver will move more air than a 10". And the much less expensive Infinity driver will still be stronger at 18Hz than the Peerless, despite its higher F3.

The miniDSP can accomplish pretty much whatever curve you imagine. Just remember that changing the shape of the curve requires cutting something. If you want stronger 20Hz, you don't actually boost 20Hz. The response 25Hz and up will have to be cut to meet your 20Hz demand. And regardless of how you shape your sub's response, you'll be limited by the xmax and power handling of the driver. It can only move so far, and can only handle so much wattage without burning itself up.

If you want to see an example before and after of a miniDSP shaped curve, click the CSS SDX12 link in my signature. I close-mic measured the response of the sub without DSP, then measured in-room response after corner loading and reshaping. You can see that without DSP, the F3 of my sealed sub is around 32 - 34Hz; whereas with lots of filtering, my F3 is closer to 14Hz. I have a 500-watt amp and my driver has a 30mm xmax, so I've got enough overhead to make extreme changes to the response like that. A 10-inch sub with a 12.5mm xmax and a 150-watt RMS power limit wouldn't be quite that flexible.

You should also understand that SVS doesn't use off-the-shelf drivers. They custom design the drivers for their subs to their specifications. So you wouldn't be able to buy the same driver they use and recreate their results exactly.

It looks like we're reaching 24.5 m/s at 20hz with the current 2.5 cm diameter port. Increasing it to 3cm seems to solve this, but the port would need to be longer. Will it fit inside the enclosure with the added length ? I guess the volume of the box would also be slightly less now.

That's a massive change with the miniDSP curve. 14hz from a sealed box sounds crazy to me.
So would it be possible to move the F3 lower with this subwoofer design? How can I know if the limits of the driver were reached?

 

[rojo]

It looks like we're reaching 24.5 m/s at 20hz with the current 2.5 cm diameter port. Increasing it to 3cm seems to solve this, but the port would need to be longer. Will it fit inside the enclosure with the added length ? I guess the volume of the box would also be slightly less now.

That's a massive change with the miniDSP curve. 14hz from a sealed box sounds crazy to me.
So would it be possible to move the F3 lower with this subwoofer design? How can I know if the limits of the driver were reached?

Don't forget, the project file I attached in my post above was at 200 watts. That'll be a peak value. For this sub, RMS power probably won't get any higher than 125 watts; and most of your listening will probably happen at less than 50 watts anyway. The 2.5cm slot is still reasonable. I re-downloaded the Sketch-Up file from post #90 and made the vent 5mm larger and 110mm longer. It still came in under the brace. You could do a 3cm slot if you wanted. But yeah, as you say, that will add displacement and decrease the internal volume of the box. I think there's no outstanding reason to modify the design again. The 2.5cm (H) x 34.4cm (W) x 58cm (L) vent will be fine I think.

You can move the F3 lower with DSP, of course. You'll know you've reached the limits of the driver when you hear it bottoming out.

 

[yonyz]

Don't forget, the project file I attached in my post above was at 200 watts. That'll be a peak value. For this sub, RMS power probably won't get any higher than 125 watts; and most of your listening will probably happen at less than 50 watts anyway. The 2.5cm slot is still reasonable. I re-downloaded the Sketch-Up file from post #90 and made the vent 5mm larger and 110mm longer. It still came in under the brace. You could do a 3cm slot if you wanted. But yeah, as you say, that will add displacement and decrease the internal volume of the box. I think there's no outstanding reason to modify the design again. The 2.5cm (H) x 34.4cm (W) x 58cm (L) vent will be fine I think.

You can move the F3 lower with DSP, of course. You'll know you've reached the limits of the driver when you hear it bottoming out.

Alrighty then. Now comes the boring part of waiting for the amp to arrive.
Thanks for the adjustment in the design. You forgot to attach the file, though.

Do you have a surround setup, btw? If so, how do you implement the miniDSP into that?

Edit: Oh, forgot to ask. I read somewhere that it's a good idea to have a space of no more than 25cm between braces. In this case, there's exactly 25cm from the brace to the top plate, and 30cm to the bottom plate. Do you think it would matter?

 

[rojo]

Alrighty then. Now comes the boring part of waiting for the amp to arrive.
Thanks for the adjustment in the design. You forgot to attach the file, though.

I didn't forget. I just did the push / pull thing in SketchUp to check whether a 3cm vent would fit; but it was going to be more trouble than it was worth to resize the vent hole on the second front baffle. I didn't bother. I hope you don't mind.

Do you have a surround setup, btw? If so, how do you implement the miniDSP into that?

I do have a surround setup. The miniDSP only manages my subwoofers. I let my AVR manage my other 5 speakers.

Edit: Oh, forgot to ask. I read somewhere that it's a good idea to have a space of no more than 25cm between braces. In this case, there's exactly 25cm from the brace to the top plate, and 30cm to the bottom plate. Do you think it would matter?

I don't think it will. It might matter for an 18" driver with higher output driving resonance to stronger levels, but I think this cabinet is damp enough for this driver. You can move the brace down and divide the distance if you wish. Its placement in the SketchUp project was completely arbitrary.

 

[yonyz]

I didn't forget. I just did the push / pull thing in SketchUp to check whether a 3cm vent would fit; but it was going to be more trouble than it was worth to resize the vent hole on the second front baffle. I didn't bother. I hope you don't mind.



I do have a surround setup. The miniDSP only manages my subwoofers. I let my AVR manage my other 5 speakers.



I don't think it will. It might matter for an 18" driver with higher output driving resonance to stronger levels, but I think this cabinet is damp enough for this driver. You can move the brace down and divide the distance if you wish. Its placement in the SketchUp project was completely arbitrary.

OK, thanks again.

 

[yonyz]

More contemplations while I wait for the amp to arrive:

1. You went with a sealed design and reached a frequency that I did not think was possible with a sealed design, certainly not with your budget ($700, which would probably translate to about $1100 in my country).
You mentioned in the thread about your sub that it's musical - something I saw mentioned often about sealed subs. I do listen to a lot of music (a lot more than I watch movies), but I just feel that my speakers have adequate bass extension for music.

So my question is, why shouldn't I go for a sealed design too? Your box doesn't look too big.

2. I've added a parametric EQ with these values to the ported design of 71 litres: center freq: 18hz, Q=3, 6db gain. It looks nice on paper (screen, actually). The -3db point is not at 18.5hz, which is a nice improvement, but you said we don't boost, so what I'd do instead is to apply a -6db gain to everything above 18hz? I couldn't find how to do that on winISD, anyhow. Also, the rear port air velocity is rather insane now at over 40m/s at 20hz. I guess the port needs to be gigantic now?

 

[rojo]

More contemplations while I wait for the amp to arrive:

1. You went with a sealed design and reached a frequency that I did not think was possible with a sealed design, certainly not with your budget ($700, which would probably translate to about $1100 in my country).
You mentioned in the thread about your sub that it's musical - something I saw mentioned often about sealed subs. I do listen to a lot of music (a lot more than I watch movies), but I just feel that my speakers have adequate bass extension for music.

So my question is, why shouldn't I go for a sealed design too? Your box doesn't look too big.

Sealed is generally less efficient than ported below 50Hz or so. If I had known when I built my sub what I know now, I might've built it ported. But I do enjoy the extended infrasonic response, so I'm not inclined to change it. Why shouldn't you go for a sealed design? A big part of it is availability of drivers and amps in Israel. My sub has a 30mm xmax and a 500 watt amp. Every bit of both is needed to EQ the sub flat down so deep while still maintaining satisfactory output. Besides, from your first post you seemed to crave deep bass. A ported design will more easily deliver that -- again, because ported is more efficient than sealed for deep bass.

2. I've added a parametric EQ with these values to the ported design of 71 litres: center freq: 18hz, Q=3, 6db gain. It looks nice on paper (screen, actually). The -3db point is not at 18.5hz, which is a nice improvement, but you said we don't boost, so what I'd do instead is to apply a -6db gain to everything above 18hz? I couldn't find how to do that on winISD, anyhow.

Your driver reaches its excursion limit at 104dB at 20Hz, 300W. If you add the parametric EQ filter you describe, your sub will still max out at 104dB at 20Hz; but you'll limit yourself to 100W above the boost. Instead of your sub reaching 111dB from 30Hz+, it'll be limited to about 106dB. Still, if you're getting a miniDSP anyway, it's worth a try. Add it and see whether you like it.

Also, the rear port air velocity is rather insane now at over 40m/s at 20hz. I guess the port needs to be gigantic now?

Remember this?

Don't forget, the project file I attached in my post above was at 200 watts. That'll be a peak value. For this sub, RMS power probably won't get any higher than 125 watts; and most of your listening will probably happen at less than 50 watts anyway. The 2.5cm slot is still reasonable.

I should also add that flaring the vent exit as is in the design will reduce turbulence some. WinISD only shows the air velocity. It doesn't indicate whether that velocity will result in audible port noise.

And it's the filter you added that has increased the simulated vent air velocity anyway. Once you get the sub built and in your room, room gain will probably take care of some of the low end roll off. You'll most likely decide you don't need the boost after all.

 

[yonyz]

For me it's not about overall volume but rather how deep it goes while remaining flat. I'm not gonna use it full power anyway, it's for a 2.5 x 3.5 meter room with no openings to any other areas.
That's why I'm still confused about how deep I can (or should) build this sub to go, without going overboard with the size of the cabinet.

What if I just use the parametric EQ to lower the volume of everything above 18hz, and then raise the gain on the subwoofer's amplifier? Will this allow me to get a flat response all the way to 18hz at a decent (loudenough) volume?

I don't know if it helps to estimate the volume I use, but my speakers are 100W each and their gain is set 1/3 while the volume on Windows is set to 30% during movies, and never more than 12% for music. It gets loud enough rather easily in this room with 8 inch speakers.

 

[rojo]

@yonyz If flatness is what you want, go ported. Here's a comparison with a couple of other decent candidates for sealed subs:

• Green -- Infinity 1260w, 71L vented box, port tuned to 23.5Hz, 300W
• Teal -- Infinity 1260w, 85L sealed box (Q of 0.52), 300W
  
• White -- Diamond D312D2, 85L sealed box (Q of 0.658), 300W
• Peach -- Alpine SWG-1244, 85L sealed box (Q of 0.54), 250W

For sealed, a good target Q is 0.5. That is considered critically damped, where the sub is well balanced between good control with no ringing, and too dry sounding. Higher values indicate less control. Higher than 0.7 I think is unacceptable. Increasing the enclosure size tends to move the Q value down, as well as decreasing the downward slope angle of the predicted frequency response at power levels below the drivers' xmax.

The Diamond will handle 400W of power; but since you're getting the BASH 300 amp, I modeled it at 300W. The Alpine driver is rated at 250W RMS.

As far as EQ goes, that's one of the fun things about owning a miniDSP. You can try filters, and if you don't like them try something else. They're excellent for people who like to tweak and poke things.

 

[yonyz]

@yonyz If flatness is what you want, go ported. Here's a comparison with a couple of other decent candidates for sealed subs:

• Green -- Infinity 1260w, 71L vented box, port tuned to 23.5Hz, 300W
• Teal -- Infinity 1260w, 85L sealed box (Q of 0.52), 300W
  
• White -- Diamond D312D2, 85L sealed box (Q of 0.658), 300W
• Peach -- Alpine SWG-1244, 85L sealed box (Q of 0.54), 250W

For sealed, a good target Q is 0.5. That is considered critically damped, where the sub is well balanced between good control with no ringing, and too dry sounding. Higher values indicate less control. Higher than 0.7 I think is unacceptable. Increasing the enclosure size tends to move the Q value down, as well as decreasing the downward slope angle of the predicted frequency response at power levels below the drivers' xmax.

The Diamond will handle 400W of power; but since you're getting the BASH 300 amp, I modeled it at 300W. The Alpine driver is rated at 250W RMS.

As far as EQ goes, that's one of the fun things about owning a miniDSP. You can try filters, and if you don't like them try something else. They're excellent for people who like to tweak and poke things.

Someone finally bought my old sound card a few days ago for $110 on eBay so I can afford getting a miniDSP or a better driver, but I don't really know where to look for better drivers.
Regarding the amplification, what happens if you match a 300W amp with a 500W driver? Obviously you don't maximize the driver's potential in terms of sheer volume, but what about its extension?

 

[rojo]

Someone finally bought my old sound card a few days ago for $110 on eBay so I can afford getting a miniDSP or a better driver, but I don't really know where to look for better drivers.
Regarding the amplification, what happens if you match a 300W amp with a 500W driver? Obviously you don't maximize the driver's potential in terms of sheer volume, but what about its extension?

Nothing much, really. You just have about 1-2dB lower max output within acceptable harmonic distortion thresholds than you would with a 500W amp. You know, you'd have to pay quite a bit more to get a better driver than that Infinity -- and by better, I mean able to achieve more than a marginal increase in performance. It really is a very well respected driver, and it's really a bit of luck that that's one of the drivers available to you locally.

 

[yonyz]

Nothing much, really. You just have about 1-2dB lower max output within acceptable harmonic distortion thresholds than you would with a 500W amp. You know, you'd have to pay quite a bit more to get a better driver than that Infinity -- and by better, I mean able to achieve more than a marginal increase in performance. It really is a very well respected driver, and it's really a bit of luck that that's one of the drivers available to you locally.

What about this 11.5kg beast? Fs of 21.5hz.

http://www.parts-express.com/dayton-audio-rss315ho-44-12-reference-ho-dvc-subwoofer--295-467

 

[rojo]

What about this 11.5kg beast? Fs of 21.5hz.

http://www.parts-express.com/dayton-audio-rss315ho-44-12-reference-ho-dvc-subwoofer--295-467

That's certainly a nice driver, and if you lived in the US and hadn't already bought the BASH 300 amp I might recommend that as a worthwhile option. But it's important to note that that Dayton driver has a resistance of 6.5 ohms, as opposed to the Infinity's 3.5 ohm load. Your 300-watt BASH amp will put out something closer to 200 watts with that driver I think. The Infinity at 300W is about 4dB stronger than the Dayton at 200W from 30Hz up. The curves meet about 20Hz, and pretty much ride together within infrasonics. I think you'd get more benefit from your money with the miniDSP. Try modelling both in a 71L box, the Infinity tuned to 23.5Hz and the Dayton tuned to 21.5Hz, the Infinity with 300W of signal and the Dayton with 200W, and you'll see what I'm talking about.

 

[yonyz]

That's certainly a nice driver, and if you lived in the US and hadn't already bought the BASH 300 amp I might recommend that as a worthwhile option. But it's important to note that that Dayton driver has a resistance of 6.5 ohms, as opposed to the Infinity's 3.5 ohm load. Your 300-watt BASH amp will put out something closer to 200 watts with that driver I think. The Infinity at 300W is about 4dB stronger than the Dayton at 200W from 30Hz up. The curves meet about 20Hz, and pretty much ride together within infrasonics. I think you'd get more benefit from your money with the miniDSP. Try modelling both in a 71L box, the Infinity tuned to 23.5Hz and the Dayton tuned to 21.5Hz, the Infinity with 300W of signal and the Dayton with 200W, and you'll see what I'm talking about.

Well I trust you did that yourself already. If you say the miniDSP is a better purchase, then I'll get it instead.

I'm just curious, though. Why not go for the 87 litre box instead? Don't you think it would make things a bit easier for the amp, once I apply some EQing?
Or, at the very least, I'll "gain back" some (1.5db, you said) of the higher frequencies that I'll lower in volume with the parametric EQ, to flatten the freq response.

Another 11cm in height, and I think now would be a good time to use two braces, but the amp would have to go in the middle of the rear baffle so that the braces would be above it and below it, one on each side.