Sub Placement with Examples

[Flint]

There have been many discussions about subwoofer placement lately, which is intriguing since it seems this topic has been so thoroughly covered in previous threads over the years - and none of the previous advice has changed.

But, recently I built a small 12" sealed subwoofer which is about a 16" cube and easy to move around, so I measured it in various places with all the test parameters remaining the same. Here's a discussion of the results.

The mic was placed at the listening position I use to demonstrate my Rocketman custom loudspeakers. I did not move the mic or change any of the electrical or measurement settings during these tests. I also included a reference curve of the sub with the mic placed mere inches from the cone to compare to how the room affects the output.

Here we go.

SUB TO RIGHT OF SEATING NEXT TO PILLAR:




This is a large space, so the sub is in one of the worst places it could go.

Notice the step roll-off below 40Hz and lack of room gain in the main operating range.

 

[Flint]

SUB FAR TO RIGHT BY STAIRCASE:



Here you can see the output is reduced by more than 6dB SPL across most of the operating range. Also, there are deep nulls at 51Hz, 92Hz, and so on.

 

[Flint]

SUB FRONT LEFT CORNER:





You can see we get some room gain at 50Hz of about 7dB SPL, but the huge null at 67Hz reduces most of that benefit. However, from about 80Hz up to 180Hz it is smooth and natural. Also, there is a 9dB gain at 20Hz.

 

[Flint]

SUB NEAR MIDDLE OF FRONT WALL





Here there is some room gain, but mostly nulls and dips in the primary operating range. This is a pretty terrible placement.

 

[Flint]

SUB HALFWAY UP WALL IN CORNER (ON TOP OF OTHER SUBS)





This gets the most room gain with 12dB of gain at 20Hz and 11dB of gain at 50Hz. The depression centered around 105Hz is a problem, but otherwise this isn't a terrible placement.

 

[Flint]

SUB IN MIDDLE OF FLOOR





This looks very good from about 40Hz to about 150Hz - very good indeed. However, the deep bass is cut significantly - as much as 18dB at 28Hz. But in the main operating range it is pretty darned good.

 

[Flint]

SUB ABOUT 20 INCHES FROM CORNER ON FLOOR





This is a pretty good response in the deep bass with little to no null issues. There is a 10dB null at 68Hz and a huge 15dB dip at 100Hz.

 

[Flint]

SUB ON SIDE WALL FLOOR DIRECTLY TO LEFT OF SEATING





As expected, this is not a good location in most aspects other than the narrow band of about 45Hz to 80Hz. There is also a surprising gain at 20Hz. Otherwise, the null are very prominent.

 

[Flint]

SUB ON CABINET ON SIDE WALL





This is an odd placement, but you can see the results. The room gain is minimal and the nulls are extreme. Not recommended.

 

[Flint]

SUB DIRECTLY BEHIND LISTENING SEAT (no photo)



This is in fact the most ideal placement in this room for this subwoofer. There is a 7dB gain at 50Hz and some decent gain right up to 100Hz. In a common config with a 80Hz crossover, this is the flattest response of the lot, though there is little output below 38Hz. For music and more serious listening, this is the position I'd choose if I were using this sub in this room with this seating position.

 

[Flint]

And that's it.

If you act quickly, I could try other locations in this room with the same settings. Just post and let me know you ideas.

 

[Flint]

One last chart...

I took all of the response curves above and put them on one chart and normalized them to the close mic measurement of the subwoofer to show the range of deviation caused by the room.



The impact of the room is significant. Most of the nulls are not that narrow and drop at least 15dB blow the reference level.



Looking more closely at the operating range of a subwoofer in a typical system, 30 - 200Hz, and expanding the resolution so each vertical division is a 3dB difference, you get this chart:



I mean, those are some VERY dramatic response differences from reference. All from room placement and room impact.

This is why I, and many others, rant on and on and on about bass trapping, getting the most ideal placement, and tuning things correctly.

 

[heeman]

This certainly tells the story.......... Thanks for all that hard work!!

Placing my 2 subs in my Texas HT and my 2 subs in the WI HT properly made drastic improvements in performance. If you have the ability to put your subs in Non-attractive locations, the performance, once place properly can be very dramatic!

 

[Towen7]

With the sub outputting well above 80Hz (the point at which we’re told that most people can begin to locate the source of bass frequencies) would a listener not find that distracting?

 

[Flint]

With the sub outputting well above 80Hz (the point at which we’re told that most people can begin to locate the source of bass frequencies) would a listener not find that distracting?

Yes, of course. But in this case I set the crossover on the amp at the maximum frequency, which is about 400Hz.

However, even if the crossover were set to 80Hz, it will still be contributing to the sound in the room up to 2 octaves above that, or 320Hz. This is because the nature of crossovers is that the filter rolls off the output as the frequency increases, it isn't a brick wall. Different Pre-Pros have different crossover slopes. I've seen some which are 6dB per octave and others which are 18dB per octave. Most are affordable receivers have 12dB per octave low pass filters for the subwoofer and most high end pre-pros are 18dB per octave.

That said the LFE channel is typically filtered off above 120Hz, but sometimes directors and engineers put higher frequency stuff in the LFE signal. So, the crossover setting in a HT is more about how bass gets redirected, not how high the sub can potentially make sound. If you set all your speakers to large and only use the subwoofer for LFE content, then none of this crossover discussion stuff should matter.

But even that can get complicated because I've seen pre-pros which redirect the higher frequencies from the LFE channel to all the speakers if the subwoofer crossover setting is at a frequency lower than the content in the LFE channel. I discovered this by accident in the early days of discreet digital surround channels where I discovered a signal in the main and center speakers when playing a test tone off a DVD which as supposed to be only an LFE signal. So, the pre-pro in that instance was redirecting the higher frequencies in the LFE channel to the front speakers. Crazy!!!


This is a messy business.

But, in this thread I am showing how the room impacts the frequency response of a subwoofer and how placement has a HUGE impact on the frequency response at the listening position. This principle also applies to all of the other speakers in the room. Given the impact you see above 100Hz, this stuff clearly impacts one's main speakers and surround speakers.

Fun!

 

[Xgm3]

Thanks for putting all this together! It looks like it takes a lot of work to setup a subwoofer correctly.

 

[Flint]

I decided to go a step farther with this because it is pretty rare that I have a small enough subwoofer that I can move it around to make measurements in different locations.

This time I put the subwoofer in a small bedroom which is about 12L x 11W x 9H feet.

I placed the microphone where a listening position would be if I used this room for a listening room, 2/5th from the rear wall and dead center between the sides.

I placed the subwoofer is the common placements most people use - in the corner, halfway along the side wall, a third of the way from the front wall, and two fifths of the way from the front wall.

Here's a drawing of the room and the sub & mic placement(s):


I took the subwoofer outside and made a ground plane (pseudo-anechoic) measurement from about 3 meters, which is the average distance from the sub in the listening room and adjusted for the gain from the ground plane (-6dB SPL). Then I took the sub inside and measured the response using the same electronics settings and the same calibrated SPL levels to show the room gain.



There are several things we can take from this:

• There is a gain of about 18dB at 20Hz for all of the placements. It would be greater if the mic was against one or more boundaries.
• The room gain at 50Hz ranges from 6dB (middle of side wall) to 12dB (in corner).
• The room null between 80 and 90Hz is greatest for the 2/5ths placement while it is the smallest with the corner placement.
• The worst performance comes from middle of wall placement and the best is corner placement.
• EDIT: Also note that the gain at 20Hz is greater than the gain at 40Hz by a significant amount regardless of the placement.