Help with very strange subwoofer curve

[Matt1980]

Yes, sorry. It's there now

 

[John Mulcahy]

Looks more extended with the direct connection. How do you have everything connected for your measurements when going through the AVR? Is the AVR in a mode that will redirect bass to the LFE output?

 

[Matt1980]

I send the hdmi from the PC to the avr.

 

[Matt1980]

I set all speakers to small, LFE at max and crossover at max (250 hz)

 

[John Mulcahy]

Try using different AVR surround modes to see if that changes the signal sent to the sub. Otherwise you could change to using ASIO drivers and send a signal to the LFE channel directly, using the process set out in AustinJerry's guide.

 

[Matt1980]

Thanks, I'm just thinking about changing the subwoofer at this point... I think these are simply not very capable units

 

[John Mulcahy]

Probably true, but your measurements so far don't look to show what the subs can do but are suffering from some configuration problem, it could be worth knowing the actual performance first.

 

[Matt1980]

Wouldn't the curve from the PC directly connected be the actual performance ?

 

[John Mulcahy]

It should be, assuming there is nothing strange in the PC loudspeaker configuration, but it is far from the specification for the sub. Was that a near field measurement of a single sub? If both were playing and they had opposite phase settings or different delays there would be cancellation between them, greatly reducing the combined output.

 

[Matt1980]

It was near measure of a single sub.

 

[Matt1980]

And I checked the CPL at 75 db

 

[John Mulcahy]

Probably worth also measuring at the port output to see what contribution that is making, and then measure about 1m away from the sub to get an idea of the combined response, all with the direct PC connection.

 

[Matt1980]

I hooked the headphone out of the PC to the AVR, its the curve in green (which is very similar to the PC direct to sub) and compared with the hdmi PC out to the avril in violet

 

[Matt1980]

These are the settings for the noise generation in REW

 

[John Mulcahy]

If you want to use pink noise for RTA measurement it has to be full range! The speaker cal signal rolls off below 500 Hz and above 2 kHz, it is used for speaker level checks, not measurement. It is better to use Pink Periodic Noise (again, full range) or to make sweep measurements.

 

[Matt1980]

This is what I have for the measurement. It seems I cannot change to noise.

 

[John Mulcahy]

OK, so you are using sweep measurements, that's good!

 

[Matt1980]

I'm totally puzzled, I've now tried with a new sub connected directly to the out of the PC. It's a B&W PV1D. And I still get a similar measure...
The green curve is the B&W sub

 

[Matt1980]

Could it be that the level in the measurement is wrong ?

 

[John Mulcahy]

Please post an mdat file, but you aren't comparing like with like. Align the measurements so they have the same level at around 105 Hz (use the offset tool in the graph controls) and you will see the difference in extension more easily.

 

[Matt1980]

Yes, but the curves are very similar below 100hz

 

[John Mulcahy]

Please post an mdat file. Where are you placing the subs in the room when you measure them?

 

[Matt1980]

The sub is on a side wall

 

[Matt1980]

Here is the file

 

[John Mulcahy]

The measurement looks odd, more like a measurement of the main speakers than the sub. How do you have the sub connected? How does the measurement look if the sub is turned off? Can you post one of your Focal measurements?