More confusion on house curves and target levels...

**Vader** · **08-06-08, 10:23 AM**

Hi guys,

The more I learn about REW and sub calibration in general the more easily I confuse myself (and that already easy enough!). I started thinking about my application of a room curve (7.8dB between 32Hz and 80Hz, in my case), and how I may be completely misunderstanding several things.
First, by application of a house curve am I not effectively cheating myself of headroom? In other words, the REW target curve is raised by 7.8dB at 20Hz (in my case), so wouldn’t that mean that a 20Hz signal would have its headroom decreased by the same amount? I’m assuming that measurement of the calibration pink noise is a measurement of the highest SPL in the 40-80Hz range.

Second, I originally defined my BFD filters by taking a raw response, manually adjusting the target level to a best fit, and played around with the filters on REW until I got a more-or-less flat response (relative to the house curve). My understanding is that any headroom lost by doing this would be regained when I calibrated the LFE trim from the AVR. I’m assuming, of course, that the AVR’s trim output is the final authority, and that a flat response (higher AVR trim?) and one with a house curve applied (lower AVR trim?) would even out (roughly).

Third, when REW empirically determines the target level, does a house curve have any effect?

Please, have no mercy in poking holes in my (incredibly faulty) understanding!

**spreston** · **08-06-08, 01:42 PM**

Great questions; I look forward to hearing responses from the gurus.

**brucek** · **08-06-08, 02:10 PM**

Quote:

First, by application of a house curve am I not effectively cheating myself of headroom?

Depends. If you simply added gain in the BFD filters to boost the signal over the business area of the house curve, then headroom is lost in the BFD.
If you added cut in the non-business area of the house curve then you haven't lost any headroom in the BFD, but if you subsequently raised the gain of the sub amp to compensate, then you've effectively lost the headroom again.

Quote:

calibrated the LFE trim from the AVR

The trim in the AVR is set before any filters are added to the BFD, and then it's never touched after that. The method of setting the input level of the BFD is covered in the BFD Guide.

Quote:

Third, when REW empirically determines the target level, does a house curve have any effect?

Yeah, if it modifies the bass management curve it will change where the automatic routine places the target level, but I would use the manual method to determine the optimum setting.

brucek

**Vader** · **08-06-08, 02:43 PM**

Thanks, Bruce. Looks like I have a little project this weekend... I think I have it correct, but I wanna go through with your clarification in mind and make sure.

One last question. Am I understanding that even if I manually move the target level up 8 dB in order to get a better fit, and apply filters based on that, and then subsequently reduce the sub gain by 8dB when calibrating to reference, effectively I have not lost or gained any headroom?

**brucek** · **08-06-08, 03:12 PM**

Quote:

effectively I have not lost or gained any headroom?

No, it doesn't work that way. If you lose headroom along a path because of one device, there's no getting it back.

You have to first consider the BFD and ensure you're not adding gain to the point where it limits your voltage swing. Optimally, you'd like to feed it a maximum voltage when you're playing the receiver at its maximum you might ever use it.

So, if you monitored the input level of the BFD (hold down the IN/OUT button until it flashes) and played some bass heavy DVD at its loudest you'd ever use, have the sub trim adjusted to just see RED in the VU meter.

So normally, you won't see the RED LED in the VU, but if you are listening at loud levels some day and you have a bass heavy movie playing you may see the RED flicker. Of course, when the BFD is receiving this maximum voltage level, even a single filter of +1 dB (at that frequency that causes the flicker) will put the BFD into overflow. Avoid gain filters - they reduce headroom in the BFD......

brucek

**Vader** · **08-06-08, 03:23 PM**

ahhhhh now I understand (I think...). That is how I originally set my BFD input (all filters are negative), but I was not sure about the headroom stuff. I'm gonna check it this weekend, just to be sure. Thanx, Bruce!

**salvasol** · **08-06-08, 05:37 PM**

Quote:

Vader wrote:

... I originally set my BFD input (all filters are negative), ...

But this doesn't mean that you're a negative person, Right???

**Vader** · **08-06-08, 05:41 PM**

Quote:

salvasol wrote:

But this doesn't mean that you're a negative person, Right???

que rim tap....

**salvasol** · **08-06-08, 05:49 PM**

...

Quote:

Vader wrote:

que rim tap....

... Is this good or bad???

**Vader** · **08-06-08, 05:52 PM**

Quote:

salvasol wrote:

...

... Is this good or bad???

Good (I think).

**Vader** · **08-06-08, 11:42 PM**

OK, I lied (about that being my last question)....

.... What would happen if the input voltage into the BFD was exceeded besides the pretty red clipping light coming on and the output being distorted? Would the BFD be physically damaged in any way? The reason I ask is that, prior to figuring out that the LFE test tones on HD-DVE are screwed up (5dB too low.... a determination that Roger Dressler of Dolby helped me with), I had calibrated the trim of the AVR 5dB too high (I forgot that the adjustment must be made to the sub's gain, and not the AVR trim). As a result, I bottomed out the subs, and buried the needle on the BFD output (I can only guess what the input looked like). The subs seem to have weathered the storm quite well, but what about the BFD? I see in the docs that the maximum input for the BFD is +2dBV, and my AVR (Denon 2105) has a rated output of 1.2V. I have no idea how these two quantities relate (different units), or even if I am comparing the same thing.

**brucek** · **08-07-08, 02:39 AM**

Quote:

The subs seem to have weathered the storm quite well, but what about the BFD? I see in the docs that the maximum input for the BFD is +2dBV, and my AVR (Denon 2105) has a rated output of 1.2V. I have no idea how these two quantities relate (different units), or even if I am comparing the same thing.

The term dBV is a consumer level term that relates directly to the actual RMS voltage (0dBV reference level is 1 volt). So, +2dBV calculates to about 1.26 volts RMS.

Quote:

What would happen if the input voltage into the BFD was exceeded besides the pretty red clipping light coming on and the output being distorted? Would the BFD be physically damaged in any way?

No, you won't damage the BFD. It is more important though in digital devices to pay attention to clipping levels. They don't respond the same as an analog device. The distortion in a digital device is immediate and quite extreme.

brucek

**Vader** · **08-07-08, 08:27 AM**

Quote:

brucek wrote:

The distortion in a digital device is immediate and quite extreme.

When the subs bottomed, there was simply the extremely loud "clack", but what would the distortion caused by the clipping sound like? My first guess was that the "clack" was due to digital clipping, so I recalibrated the BFD back down (under clipping input voltage), and bumped up the sub's gain to compensate. This time they could not have clipped, but they bottomed again and it sounded the exact same (nasty sound, though!). Since the "clack" was not clipping, and you describe clipping distortion as "immediate and extreme", what would I hear from the sub (that I did not hear either way)... did I dodge a bullet?

**brucek** · **08-07-08, 08:53 AM**

Quote:

Since the "clack" was not clipping

It may be clipping of the sub amp though.

Or you may be simply exceeding Xmech of the sub driver.

Either way, you should stop doing it...

brucek

**Vader** · **08-07-08, 08:58 AM**

Quote:

brucek wrote:

Either way, you should stop doing it...

I re-calibrated back down below clipping, and the AVR trim back using the known correct tones of SD-DVE. That's when I enlisted the help of Roger Dressler in determining that HD-DVE was at fault. Obviously, it has not come close since. I'm just relieved I didn't do more harm than was apparent. Thank you for your explanation (yet again)!

Advertisement