18" build (Arande) - Page 3

I am not understanding how you have it connected. Do you have two AV receivers, one driving main speakers, center and surrounds, with the sub output going to a second receiver that is eventually driving this sub?

A little bit of clarification would be good here.

18 v into 2.8 ohms is around 115 watts... however you would have to see what the actual impedance at 27 hz is... WinISD should give you a value... I am willing to bet the actual impedance around that frequency would be around 20 hz at least, so 18V ^2 / 20 would be about 16 watts.

The only time the impedance will drop near the Re would be right at your tuning frequency, you will have peaks below and above that in frequency.

Bottom line, the receiver is pretty sub par. Test with an actual amplifier that can put out at least 60 volts into the load.

If you are cascading crossovers, you will also be rolling off the high end.

I don't know how you are arriving at a sensitivity figure, are you using a digital multimeter to get voltage?

The 98 hz resonance will be healed with absorption in the box, yes, that is possible with a big, long sonotube, infact I had one very similar.

See here: http://www.avsforum.com/avs-vb/showthread.php?t=1141956

Ok. Measure the box outside at 1 meter from the cone, and then again 2 meters from the cone, outside... you will get two different curves. The 2 meter curve will look a lot smoother without a big hump.

The port and driver will be a lot closer path length to the microphone, when you measure farther away.

Most pro stuff is measured at 10 meters with 28.3 volts input and calculated back to 1 meter / 2.83 V.

I can't say that the Rat Shack or even the Galaxy meter is accurate at low frequencies.... even the Behringer ECM8000 I don't trust down low, unless you are measuring with a lot of voltage into the cabinet. IE: almost full tilt operation.

Gain structure in REW matters, especially when you have multiple gains as some of us do. If your soundcard clips, or in the case of ECM8000 mic / mixer, if you clip that it will affect what REW measures.

Are you placing the SPL meter on the ground when testing outside?

How deep is the dip, and how wide of a bandwidth is it? IE... if it's 2 hz wide, etc...

Sounds good... are you using a loopback on L channel on the soundcard?

I am sure that there will be a lot of people that will disagree, but you won't get meaningful data below about 20 hz outside, unless you are driving the cabinet close to it's limits... and make sure nothing in the signal chain is clipping.. mixer etc.

Trust me, you need more power, as well as some eq in room.

Did you throw some low frequency sines on it to break in the driver suspension when you got it, or just threw it in the box?

What's your noise floor like?

[quote=arande2;17995056]
No, I was doing a diagonal 1 meter to driver/port, but I am guessing on the ground farther away is more accurate (in terms of absolute spl and relative spl)

Since I don't know what you mean, probably not. If you are asking if I am using the left channel to measure, yes.

Is this because of background noise, wind, etc..?

I still think 2 subs will be needed (weird pressure difference between ear) so I don't get distracted by noises (rattles etc..) where the subwoofer is...

I was planning on more power and EQ anyway

Yes, I broke it in for 3 weeks.

I assume you mean what the VU meter says on the input without any signal. -42dB
"

Pro stuff is measured farther away to get away from the 'velocity' issue that will affect the measured FR. Tom talked a bit about this issue, but I don't recall where the post was placed.

Place the mic on the ground when measuring the cabinet outside.

Measure on the right channel, and tick the box that says "use left channel as calibration reference" You will find this under 'settings' in REW.

I use an RCA>RCA cable that physically connects the soundcard output on the left channel to the left channel input. This is what is meant by a 'loopback'.

Yes, I was meaning that outside wind noise, trains / traffic will affect the measured FR down low.

Use the sine wave generator in REW to sweep for rattles and go around, fixing them. Rattles will totally ruin the perception, fix them or you will go loopy...

A little bit of thin foam weatherstripping tape can do wonders for hanging pictures, etc.

Can you post graphs of the measurements with the suggested settings for REW? 15-200 hz, 45-105db. Alternately, you can PM me, and I will give you my email, you can send the .MDAT file ( measurement file ) and I can take a look at it.

Some absorption inside the sonotube will be needed to kill the end to end resonance.

42 db noise floor is pretty decent, you should be able to get relevant measurements.

Did you get your amp yet? You should really hold off on conclusions until you get a proper amp. Its like saying I have a corvette and its not really going fast. It has a go cart engine in it, but it should still go very fast.

What happens when you have a lot of distortion in a speaker, it will measure louder than it actually is. ( it will also sound louder ) An SPL meter reads all frequencies together, so if you have high 2nd and 3rd harmonics, the meter will read artificially high.

I don't think there is a build quality problem, unless you have leaks, even then you should be able to generate half decent levels. Can you give us more details about the room... if it's a main floor with lossy construction, you can be losing output through the floor in the low end.

Pictures are worth a thousand words here.

The only way you can really audition a subwoofer is with decent mains, and having a good blend to the main speakers. It should sound like you have gigantic mains, that dig deep when the need calls for it, not like a 45 hz tuned prosound sub. 'thump thump'

PM me, I will give you my emaill address and look at the .mdat file... you have to save the measurement, and then attatch it to the email.

Arande, I took a look at the picture you posted, and it's really not that bad indoors.... you don't have a lot of gain down low, but if you pull that 40~ hz peak down a bit, you fit within a + 3db, -3db window from about 17 hz to 70 hz which is quite respectable, as long as the measurement is valid.

Don't forget that the shape of the graph will look a lot different when you set the graph limits to the REW defaults, 45-105 db, and 15-200 hz, the peak will 'look' a lot less than what shows up in your settings.

I would do a bit of investigation with a sine wave generator parked at your 22 hz dip... something must be vibrating sympathetically to get that deep of a dip. If you don't end up healing it, not a huge problem because the bandwidth of the dip is pretty narrow... you'd likely never notice it.

Are you using the correction tables for the Radio Shack Spl meter, if that's what you are using?

You will not regret buying the amp, I believe it will solve all your problems. I think the ep2500 will clip much before you blow your driver as they are incredible subs. My 2c's.

1) Radio shack spl meters aren't accurate at low frequencies... even with using the tables to add to make up for the lack of meter sensitivity down low. Individual units can be off 3-4 db easily. I think it was Ricci that posted a picture of two different Galaxy meters and a couple of rat shack meters that had different readings... which one do you trust?

2) Stiff spiders can rob you of some output at low drive levels. Temperature will affect the compliance as well. Cold temps, the surround and spider will be stiffer than room temp.

I dunno what to tell you other than I don't feel you have a problem with sensitivity.. if you add up all the factors that can be accounted for.