Preventing BFD overload

[goatfarm]

I didn't know exactly where to put this thread so I thought I'd start here.

I've always tried to be religious about not letting my 1124's meters get into the red and clipping so I've kept the LFE output on my Sony 930 receiver at -8dB. Been watching Transformers this morning, and had to change the setting in order to keep the BFD from overloading. It is now at -10dB. I hope that's enough because there is no lower setting. And I thought WotW was bass heavy!

Am I right in doing this? I've read that digital distortion can be really bad. Does anyone else monitor their BFD to make sure that it's not clipping and/or distorting? BTW none of filters I'm using add anything. All are used to subtract level from their specified bands. Oh, and I turned up the sub just a smidge to compensate.

Greg

[Otto]

Hi Greg,

That should be OK if you want to keep from clipping. I watched Transformers the other night as well. Definitely some bass there! I must admit, though, I wasn't monitoring the input level, only the output level. What if you change the input level on the back of the BFD to the "pro" setting? Try chaningg the switch on the back of the BFD from -10dBV to +4dBu. That should give you some more input headroom.

[Wayne A. Pflughaupt]

That’s the problem with setting the BFD’s input levels to maximum: It leave no headroom for boosting filters, if they are needed, or unexpectedly challenging program material – as you have seen.

The "set the levels as high as possible” recommendation dates back to the days when 12, 16 and 18-bit AD/DA converters were the norm in professional-grade digital processors - i.e., before the proliferation of 24-bit converters. Even the manual for my Yamaha digital parametric, an older model with 20-bit converters, recommends setting the input level as high as possible. But I’ve spent the last few days poring over the manuals of many digital pro-audio equalizers from Ashly, BSS, Klark Teknik, Rane, Sabine, Presonus, Alesis, Behringer, etc. – i.e., from cheap to ultra-expensive. All have 24-bit or better processing, and I have yet to find a manufacturer that recommends maxing out input levels as we typically do with the BFD.

Indeed, they hardly make any statement about input levels at all, as one would expect if it was as big a deal as we've been led believe. One manual that does mention signal levels, the Rane DEQ 60 1/3-octave EQ, recommends setting the gain level at ~10 dB below max. (Good advice for live sound applications; in our less demanding home environment we could probably go with ~5-6 dB [after applying filters, of course].) The BFD manual, another rarity that mentions input levels (some other Behringer EQ manuals are also silent on the subject), only mentions keeping levels out of the red. It does not state that levels are supposed to be pushed “as close to clipping as possible” in order to maximize performance.

Older 16- and 18-bit processors had a relatively high noise floor, which is why they needed their input levels set as high as possible. But what many people apparently don’t get is that higher bit rates have increased dynamic range downward, enabling an ultra-quiet noise floor and better resolution at low signal levels. Digital processors with 24-bit or better converters function identical to analog gear for all intent and purposes, which is why level setting has become a non-issue.

For instance, a 24-bit converter has a theoretical S/N ratio of ~140 dB. Of course, the BFD’s rather inferior converters can only manage about 95, but that still means an extremely low signal, say 50 dB, is 45 dB above the noise floor!

So – back off the input signal, or change the range setting, to get yourself some more headroom. Contrary to what you may have heard, the BFD’s noise floor is static. It’s not going to increase or decrease with varying signal levels. Anyone with the means to connect the BFD to their main speakers can verify this.

BTW none of filters I'm using add anything. All are used to subtract level from their specified bands.

Remember that cut-only filtering will lower the meter’s level with the EQ engaged. So when you see the meter hit the top, or close to it, that will mean your input levels are already clipping.

Regards,
Wayne

[goatfarm]

Wayne,
Thanks for your usual thoughtful reply. I do have some questions regarding your response.

You said:

"That’s the problem with setting the BFD’s input levels to maximum: It leave no headroom for boosting filters, if they are needed, or unexpectedly challenging program material – as you have seen."

I can't find any method in the manual as to how to change the BFD's input levels. So I assume you're only referring to the manufacturer's settings. If that is the case then most everyone using a BFD is pretty much at the mercy of their pre-pro/receiver output level for sub/LFE to determine the input level for the BFD. If that level can't be lowered enough to eliminate the dreaded "red zone" on one's BFD dial, then distortion is inevitable, yes?

How much do the -10dBV and +4dBu mentioned by Otto actually differ in sensitivity? Should those of us who habitually inhabit that fine line netherland between totally safe and regularly "red-zoned" consider the alternate latter setting?

Unless my receiver has some oddball subwoofer output level, I think most BFD owners live inside the 20-yard-line. And here nobody wants a touchdown!

Greg

[Wayne A. Pflughaupt]

Greg,

You are correct that there is no means for level adjustment on the BFD (one of my pet peeves with the thing), so we are at the mercy of whatever our receivers or pre-pros can provide. I had the same problem as you in my system, with the signal level hitting the BFD’s red LEDs.

Switching to the +4 dBu range should take care of the problem. To put the two settings on equal footing, the consumer –10 dBV setting equals 0.316 Vrms, while the pro-audio +4 dBu setting equals 1.23 Vrms. Most receivers have a maximum output of +2dBV (~1.25 Vrms); the BFD’s maximum output is +16 dBu (~4.9 Vrms). So switching to +4 dBu nearly quadruples your headroom.

I can’t imagine that switching the operating range wouldn’t take care of your problem. Pro audio gear has miles more headroom than most consumer gear, even a low-end piece like the BFD.

Regards,
Wayne

[Blaser]

Dear Wayne,

Since I have built my dual 18s subs, I have powered them with about 4KW RMS. The subs can go very loud and deep. I once changed the setting of the BFD to +4dBu, but I noticed there was a hiss coming out of the subs when the amps are maxed out.
I have discussed this matter with some known names in this industry and we concluded that the +4dBu setting will have greater dynamic range, but when using a strong amplification, the noise will become audible...and therefore I had to apply the -10dBu setting to have the noise down.
But thanks to the equ. flexibility, one can have any desired output level. In my case, I applied +3 db at 25.9 Hz with the widest Q of the BFD 2496, and this raised the output by 3 db everywhere.
So I guess this can be the trick, people who desire to raise of cut the output level could chose to do as I did, take FRs, LFs and it is magically done

[goatfarm]

blaser wrote:

Since I have built my dual 18s subs, I have powered them with about 4KW RMS. The subs can go very loud and deep. I once changed the setting of the BFD to +4dBu, but I noticed there was a hiss coming out of the subs when the amps are maxed out.
I have discussed this matter with some known names in this industry and we concluded that the +4dBu setting will have greater dynamic range, but when using a strong amplification, the noise will become audible...and therefore I had to apply the -10dBu setting to have the noise down.
But thanks to the equ. flexibility, one can have any desired output level. In my case, I applied +3 db at 25.9 Hz with the widest Q of the BFD 2496, and this raised the output by 3 db everywhere.
So I guess this can be the trick, people who desire to raise of cut the output level could chose to do as I did, take FRs, LFs and it is magically done

Are you simply saying that the +4dBu setting is noisier and to avoid it if one plays at higher volume levels? This would pretty much eliminate that setting as an option in most high volume situations. OTOH I can't see that the amplifier power you are using would have anything to do with hiss unless the amps themselves are noisy to begin with. And raising the output by 3dB everywhere just means having to lower the input level by a corresponding 3dB to avoid clipping. I feel reasonably good about that last statement.

[Ilkka]

blaser wrote:

Dear Wayne,

Since I have built my dual 18s subs, I have powered them with about 4KW RMS. The subs can go very loud and deep. I once changed the setting of the BFD to +4dBu, but I noticed there was a hiss coming out of the subs when the amps are maxed out.
I have discussed this matter with some known names in this industry and we concluded that the +4dBu setting will have greater dynamic range, but when using a strong amplification, the noise will become audible...and therefore I had to apply the -10dBu setting to have the noise down.
But thanks to the equ. flexibility, one can have any desired output level. In my case, I applied +3 db at 25.9 Hz with the widest Q of the BFD 2496, and this raised the output by 3 db everywhere.
So I guess this can be the trick, people who desire to raise of cut the output level could chose to do as I did, take FRs, LFs and it is magically done

Hi Ahmed,

I can't be 100 % certain but I think the hiss you were experiencing wasn't caused by the BFD per se. I'm blaming your amps instead. I can say it because I too own one. I too get a hiss unless I lift the ground on the amp. The TA2400 has a switch for this but for some strange reason it doesn't necessarily work. So try it first but if it doesn't work, just tape the ground prongs (yes, I know it CAN be dangerous) on the wall socket or buy a so-called "cheater plug" (it is possible that they aren't available there).

[Blaser]

Ilkka!!!!!! Long Time!

WOW thanks! I think I will keep the -10 db instead Everything is now perfectly calibrated (to my taste), Should I change my mind I will for sure take your custom tweeked amp and give you mine

[Blaser]

goatfarm wrote:

Are you simply saying that the +4dBu setting is noisier and to avoid it if one plays at higher volume levels? This would pretty much eliminate that setting as an option in most high volume situations. OTOH I can't see that the amplifier power you are using would have anything to do with hiss unless the amps themselves are noisy to begin with. And raising the output by 3dB everywhere just means having to lower the input level by a corresponding 3dB to avoid clipping. I feel reasonably good about that last statement.

OK, the problem was propably due to some grounding problem in the amps (I will have to check), but the increasing or decreasing the overall output of the BFD by the method I described above would propably cure the input clipping problems

[Wayne A. Pflughaupt]

Blaser,

Not to discount anything Ilkka said (since he’s used your amp and I haven’t), but typically here when we’ve seen noise problems like you describe, its been caused by going overboard with cutting filters on the equalizer. While the typical home receiver has a maximum output level of only 1.25 Vrms, a good pro amp can handle input signals up to 14 Vrms. Severely reducing an already low signal via equalizing and sending it to a component that wants a hot signal level is a sure recipe for noise.

OK, the problem was due to some grounding problem in the amps, but the increasing or decreasing the overall output of the BFD by the method I described above (applied +3 db at 25.9 Hz with the widest Q of the BFD 2496) would propably cure the input clipping problems.

...except that using the equalizer as a level-matching device is really poor form. Its desigated purpose is for smoothing response.

Regards,
Wayne

[Blaser]

Thanks Wayne,

The amps I have will deliver their max. with only 0.707 V RMS input. And as I do not have grounding in my Flat, normally the amps (and BFD) should not have any hiss problem due to grounding or so.
The hiss I was talking about is not really a hiss. It is even hardly hearable when I am say 1 feet away from the subs. What cought my attention is that the first led of my amps started to light very very slightly (when I used the +4dBu configuration) and then I thought it is (unhearable) noise. Therefore I just decided to use the -10 dBu setting to avoid seeing the first led "trying to light".

My conclusion was that there is a S/N ratio to any equaliser (even the FBQ 2496), and therefore, when I switched to the +4 dBu setting, this noise was amplified.

What is this "unhearable" noise caused by? It might be amplified noise from the FBQ that was revealed by the first Led of the amps (which gain is set all the way up). But again, I should not have called that a hiss or so because it is way too week to be heard, but the 4KW amps began to feel them.

Concerning using a filter as a levelmatching, I took some REW measurments and nothing was badly afftected, neither FRs nor Waterfalls... But the FR was shifted everywhere according to the amount of boost or cut.

[Blaser]

Something just raised to my mind while we are speaking about noise. What would be the noise SPL if the subs can hit 130 db+ uncorrected (equalized) at LP?

[Wayne A. Pflughaupt]

blaser wrote:

Thanks Wayne.

Concerning using a filter as a levelmatching, I took some REW measurments and nothing was badly afftected, neither FRs nor Waterfalls...

An equalizer can do audible things to response that aren’t always readily apparent from a response chart. If you haven’t already, you might want to take a look at my Minimal EQ thread for more details.

Regards,
Wayne

[Wayne A. Pflughaupt]

Came across this picture on Rane’s web site that helps put the –10 dBV/+4 dBu thing into perspective.

Regards,
Wayne

[goatfarm]

Just wanted to post an update. After switching the BFD from -10dBV to the +4dBu setting there has been no sign of overload. I have to admit that in the "consumer audio" position I never really heard any digital distortion when the meters showed momentary overload. That's when I would hurry and back off the volume, and I just hated doing that in the middle of a movie or Mahler symphony. Kind of destroys the intensity and immersion in the work. But I figured it was better to be cautious than sorry. I did have to readjust the LFE output on the receiver and the volume level on the sub amp a bit to compensate for the change, but now all is well. My thanks to all who made suggestions, especially Wayne.

Greg