16Hz from an 8' Sub - What is going on?

After realising my link to the sub readings didn't work I thought I'd post a working one (I have redone the tests at -10dB)

Pro 50 readings

Thing is I must be doing something wrong, why am I getting output at 16Hz from an 8' 50 watt sub?? During the 16Hz tone I couldn't hear much but the door rattled like crazy!

Are my RS anolog spl meter correction values right?

 

Something wrong somewhere. I'd check to see what you're doing wrong. As you say, its impossible for an 8 inch sub with a 50 watt amp to reach this low with any real loudness.
The firstime I heard REAL bass was when I had the pleasure of visiting Uncle Eric last month. Until then, I thought my M&K was special. That Velodyne of his is like the vice of death. If I was a surfer and I got hit by one of those 50 foot waves, I think thats what I'd feel like. Oh my god!

 

Place SPL meter at listening position, start tone cd, write down raw spl values, insert into spreadsheet which corrects each value depending on the rs correction values (which I've double checked)
I've ran the test half a dozen times all with similar results.

???

 

Look like a blinding responce to me, a bit of a null at 63 hz but I would be chuffed with a such a curve.

Just one question what size room do you have and what is your subwoofer placement??

 

I'm not getting 16htz from my 2 grand MK. Better chuck it in the skip and pick up a 200 quid jobbie
Do me a favour please.

 

If the room is small(I thin it is 13/14'?)...with no large openings to other areas...room gain will start in the mid 40s. Adding (in theory)12dB/octave. In the real world...figure 9dB/octave max. But that is still 9dB by 22hz and about 15dB by 16hz.

If you can measure the subwoofer outside(50-60ft from the nearest reflective boundary) this will give you an easy way to find out what the room is adding to the low freqs.

Remember, just because the SPL meter is reading *X* SPL output using say a 16hz test tone...that no way implies the output of the subwoofer is only 16hz. Chances seem good the subwoofer in question could be producing a lot of the 2nd harmonic(32hz) and maybe even a lot of the 3rd and 4ths.(48,64hz). The harmonics (harmonic distortion)will show up on the SPL meter as output too.

Even order harmonics tend to sound *harmonius* to the human ear...so you could have a subwoofer producing 50% THD and if it is based mostly in the even orders(2nd,4th,6th,ect)...it won't sound too bad. Odd orders tend to sound quite harsh though. Odds are usually produced when the amp clips or the driver exceeds linear excursion capabilities. Harmonics are often to blame when folks contend they can *localize* bass <100hz. It isn't the real bass they are locating...is it the harmonics which are much higher freqs(and not attenuated by any crossovers).

If you have the time and a PC nearby...goto www.soundtechnology.com and DL one of the spectra analyzers...you can then see the THD as you perform your tests. I have a bit of experience with PC based RTAs so if you have any questions...fire away.

TV

 

Tom,
Great post. Welcome to the forums.
Eric

 

Tom,

I agree that with many subs, when you get down to the sub-20Hz region, much of what you (and your SPL meter for that matter) hear are harmonics. I have done some spectrum analysis on my own REL Stentor II and can confirm this fact. If you go low enough, the fundamental frequency all but disappears and all you have left are harmonics. This is the main reason I have recently changed my REL for one of Uncle Eric's Velodyne HGS18s. Servo-controlled subs like the HGS18 suffer this problem much less than the alternatives.

One thing puzzles me in your post though. Assuming that the test tone being played is a pure 16Hz tone, how is putting a 1/3 octave equalizer in the circuit going to affect the signal that the sub sees and therefore the output it produces? The harmonics we are discussing here are produced 'mechanically' within the sub itself and arise because of the way the driver and port interact with the air they are trying to drive.

Of course, if you're talking about placing the EQ in the measurement chain then I understand what you are suggesting but the drop in sub level would not be audible, just measurable.

As Uncle Eric said, welcome to *the* forums.

 

Hi Tom, lots to consider there. If the 16Hz tone produced harmonics at 31Hz I should've heard the tone shouldn't I? I couldn't hear the 16Hz tone when it played, just felt it, all I heard was the door rattling.

Room is about 12' x 13' (actually maybe less than that)
Sub is about a foot and a half from the corner behind my front left.

 

Hey Tom, welcome, I hope we see much more of you, it's a friendly place

Steve Bate

PS For those who dont know Tom is half of SVS

 

The sound meter wouldn´t be picking up the door rattles by any chance?? Just a thought!

 

I knew

 

Hi Tom good to see you at AVforums.com, I see you post to Hometheater forum quite a lot.

 

Figures,

My first post here and I make a rookie mistake

Charlie made a great catch on my error!

Since the harmonics are produced downstream from most circuitry that might attenuate it (as I alluded too when I said the crossover won't lower the harmonics)...adding an EQ isn't going to do anything. I'll Edit my post...thats what I get for typing faster than I am thinking.

The servo feedback circuits are possibly the one circuit that you can point too that can lower harmonics. But whether or not they are actually downstream,upstream...or mid-stream is probably debateable. They work very well in what they are designed to do regardless.

TV

 

Thanks for the welcome everyone,been lurking for quite a bit now.


>>>Hi Tom, lots to consider there. If the 16Hz tone produced harmonics at 31Hz I should've heard the tone shouldn't I? I couldn't hear the 16Hz tone when it played, just felt it, all I heard was the door rattling.<<<


Well, that is a tough call. 32hz would be audible to most folks...but only once a certain output level is reached. And the door may need very little 16hz output to *rattle*...there are a few subjective variables here that we can only guess at.

This is an example of harmonics produced using a 20hz input signal. I think this will clearly show some things I might be struggling a bit to explain



This was taken using the RTA program anyone can DL for free(works for 30 days,then you have to purchase the *key* to keep using it) from Soundtechnology. So if you have a PC and a RS SPL meter...you can take some VERY detailed measurements of this subwoofer in very little time for no cost. Spectra/soundtechnology can be a little tricky to setup the first time once you DL it though...so if you have any questions...just fire away.

TV

 

Another important consideration in all this is the human hearing rolloff down low. You can search on Fletcher/Munson to find the hearing curves for humans.

If memory serves a 20hz tone needs to be about 45-50dB louder then a 120hz tone to have the same audibility to us. In other words...

If a 120hz tone needed to be 50dB to be audible to us in a given room...a 20hz tone would need to be about 95-100dB for us to hear it.

So this works against us when a subwoofer is producing harmonics. Since the harmonics are much higher in frequency, our ears will be more sensitive to them compared to the actual source material. On the example graph with subwoofer *X*...you can see the 3rd harmonic is almost as loud as the 20hz input signal. And when we factor in the FM curves...that 60hz tone will sound MUCH louder than the original 20hz tone.


TV

 

Tom,

Forgive me pointing out your mistake. I am prone to them myself.

Thanks for publishing that spectrum analysis. For anyone doubting what Nic has been saying all these months about the distortion levels that some subs produce, take a look at that THD figure - nearly 47% !!

I also published some similar spectra a few months ago from my own REL Stentor II in this thread. I didn't get the spectrum analysis software to display a THD figure on that occasion but did subsequently, and it didn't make good reading.

As for the original question, I guess that at these test levels, an 8" 50W sub *could* produce a reasonable output at 16Hz capable of rattling doors. But crank the sound up to anything near reference level and it would struggle.

 

The readings were taken at -10dB (0 calibrated to 105dB reference) which is aroundabout my usual listening level.

If I did increase the volume wouldn't the very low frequencies still stay in the same formation in relation to the higher bass tones? i.e upping 10dB to reference, both 16Hz and say 40Hz would both raise by aprox 10dB? (ignoring room gain)

When subwoofer specs are quoted they never give a overall volume level, just a comparsion figue i.e -3dB @ 20Hz. Wouldn't that remain -3dB through the volume range.

Anyway, 16Hz at the level I watch movies at for 300 beans, I'm happy!

 

Tom

Welocme to AVForums, I think your presence might save me loads of time