Home Theater Forum and Systems banner

1 - 20 of 33 Posts

·
Premium Member
Joined
·
629 Posts
Discussion Starter #1
Hi all - This is an open thread to discuss subwoofer testing methodology. What is important when testing a subwoofer?

1. Frequency Response?
2. Max output at various frequencies?
3. Distortion at various frequencies and SPL levels?
4. Standard Features?

To get this started ... let's say this was proposed as a method for testing subwoofers, would everyone agree that this is the optimal approach?

Proposed testing method #1.

A. All subwoofers will be tested outdoors at one meter Groundplane.
B. Six different frequencies will be tested on each subwoofer at varied lengths of time, shown below. The purpose of using these test tones will be to find how much SPL, measured in dB, that each subwoofer can deliver at each frequency shown for the duration shown.
C. 20 Hz will have a test tone which lasts 0.325 second.
D. 25 Hz will have a test tone which lasts 0.260 second.
E. 31.5 Hz will have a test tone which lasts 0.206 second.
F. 40 Hz will have a test tone which lasts 0.1625 second.
E. 50 Hz will have a test tone which lasts 0.130 second.
F. 63 Hz will have a test tone which lasts 0.103 second.
G. Each test tone will be delivered at increasing SPL until the subwoofer is exceeding 40% THD at a given frequency.
H. Bass below 20 Hz will not be measured.

This is proposal 1 for a subwoofer test methodology.

Please feel free to chime in with opinions on this.
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #3
Dale - This methodology would use independent test tones, not a sweep.

And for those who hate decimels ... .325 second is less than 1/3rd of a second. These tones range from about 1/3rd of a second for 20 Hz to about 1/10th of a second for 63 Hz.
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #5
For further clarification - There is a real lack of meaningful subwoofer testing being done now, and the home theater world could use some testing done which combines both measurements which reflect real world performance along with (as unbiased as is reasonably possible) listening tests.

I put up the methodology in the first post in order to give people a starting point for testing subs.

The first post is intended to get the members here to say "I think it's a great test, and here is why" ... or ... "I think it needs a lot of improvements, and here is why ... "

Let's see some opinions ...EVERYONE READING THIS is welcome to reply.
 

·
Premium Member
Joined
·
2,539 Posts
I would prefer a down frequency sweep beginning at 200Hz and ending at 5Hz increasing in amplitude until the peak output is compressed 1dB (if 1dB is not adequate for audio testing then 3dB compressed).
I would also like to see distortion at 10Hz or 20Hz intervals along with 2nd and 3rd order harmonics.

I am really not all that concerned about burst mode, if the performance is there in swept mode it will be there in burst mode.
 

·
Registered
Joined
·
4,127 Posts
I agree that we need to establish a better methodology for subwoofer testing that is not only indicative of real world results, but is also repeatable. I think this is a great starting point.
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #8
I would prefer a down frequency sweep beginning at 200Hz and ending at 5Hz increasing in amplitude until the peak output is compressed 1dB (if 1dB is not adequate for audio testing then 3dB compressed).
I would also like to see distortion at 10Hz or 20Hz intervals along with 2nd and 3rd order harmonics.

I am really not all that concerned about burst mode, if the performance is there in swept mode it will be there in burst mode.
This is the type of response that will make for a great thread. :T
 

·
Premium Member
Joined
·
3,777 Posts
Sounds like a great starting point. Have fun. Dennis
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #10
So far ... 78 views and one suggestion. No one has commented on the first criteria as posted.

I will offer up what I think is important ...

1. Max output using sinewaves (either sweep or individual) before audible distortion is detected by the tester, who should be several meters away from the subwoofer.

2. Frequency response curves along with analysis of the posted response curve against maximum output curves. Why this analysis? Let's say we are looking at two sealed subwoofers. Subwoofer "A" is +/- 3 dB from 15 to 100 Hz while subwoofer "B" is down 9 dB at 15 Hz.

The first impression may well be that "A" is superior to "B".

But then we have the max output sweeps ... and we see that "B" delivers, say, 100 dB at 15 Hz while "A" delivers only 91 dB.

We can conclude that "A" likely has more eq built into the response curve, and that "B" was designed with less eq, but more output.

"B" is actually the more powerful, capable subwoofer.

Without analysis, people may miss this.

More later. :)
 

·
Registered
Joined
·
4,127 Posts
If we are going to make this a real world test, there needs to be some sort of way to get an acurate measurement "in room" in order to let people know what they can expect out of the box without an EQ. Maybe a 10Hz-200Hz sweep from the center of the room directly from the sound card to the sub. Of course there would need to be a baseline or some sort of parameters established such as an untreated 15X20 room. It doesn't have to be pretty or incredibly detailed, it just needs to represent what one can expect to have to tame.

I know there are plenty of variables that have to be considered such as insulated walls versus hollow walls, carpet vs tile vs wood floors, 10'ceilings vs. 8' ceilings and so on, but I think it would be beneficial to see. I am not sure on how to go about it, I just think that there would be some benefit to it.
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #13
Dale - Unfortunately, there is no way of doing an "in room response" test that could be meaningful in terms of raw data. However, listening tests are also important to add to the data. Let's say the reviewer in question has a 20x15x8 foot room. After the measurements, a series of listening tests in this room with each subwoofer being properly eq'ed according to its capacity to deliver bass should be done.

A smaller ported sub may only have useful output to 32 Hz, but it may be a terrific budget entry and sound good in this room.

A larger sealed sub may well extend to 10 Hz in this room.

The reviewer needs to be able to properly eq each subwoofer in his room, while staying within the limits of each subwoofer. He/She then needs to report on sound quality, overall feel ... and describe what the subwoofer did with various movie and music scenes.

The reviewer is now setting a good objective and subjective criteria, which is (IMO), the proper combination of criteria.
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #15
I'm not sure 1/3 or 1/4 duration tones are representative of demands made on subs in real world listening such as movies. A longer time frame perhaps? Bass below 20 Hz should be included as some designs do reach the lower frequencies.
Mike - You are onto something here. The opening criteria was posted with a specific reason in mind. I am curious how long it is until someone identifies the criteria as written.
 

·
Registered
Joined
·
6 Posts
I would have chimed in, but I am such a noob I would feel like I just got invited to NASA to help redesign the shuttle.:dumbcrazy:
 

·
Registered
Joined
·
2,431 Posts
Hey guys. Craig. I have recently started up a testing program. I completed the first round back in October. Or data hosting site just launched. We plan to do 2 or 3 rounds a year from here on out on just about anything I can get my hands on. Diy, pro sound, commercial ht, car audio raw drivers, whatever. The next round should be in May. Have a look at the tests I am doing under the "know how" section. We have another test or 2 to be added as well. This battery is way beyond what most tests will entail but hey you asked. :D
 

·
Premium Member
Joined
·
629 Posts
Discussion Starter #18
Ricci - It's nice to see you in this thread ... I clicked on the link, and there is a written "mission statement", but I could not find any data for the tests you already did. Am I missing something, or are you working on getting the results from last March on the site?

This looks intriguing - I also know of another prominent HTS member who is looking to start testing subwoofers.

I am not, in any way, trying to run how these tests should be done. What I hope to accomplish with this thread is helping to get a testing protocol which shows meaningful data, adds some subjective listening tests, and is thorough enough for the experienced diyer to read with confidence while also allowing the newbie, and even moderately experienced subwoofer shopper an understandable test.
 

·
Registered
Joined
·
2,431 Posts
Mike has it down. You can click on almost anything on the site linked in my sig and it will take you where you want to go. Also the graphs are active and will sort by the data column you have selected. We only have 9 systems tested as of now but we should be adding another 10-12 in May. We have been working on this for about 5 months now. Anyway under "know how" is a list of my equipment and the test program here.

http://www.data-bass.com/know-how.

It is rigorous.
 
1 - 20 of 33 Posts
Top