Question

[DBfan187 0]

Dave Navone says that you should use a -10dB tone to set your amps gains for a 3:1 ratio overlap.

Should I do that?

I've been using the 0dB tones. Switch?

[ANeonRider 3]

3:1 ratio overlap, I don't follow....

[Supa_c 1]

you lost me at dave navone

[DBfan187 0]

3:1 ratio overlap, I don't follow....

<{POST_SNAPBACK}>

please explain why?

[ANeonRider 3]

I don't follow why that would be necessary at all.

I would like to hear a bit more reasoning. -10db is alot of power difference with 0dB, and if you set your gains at -10dB, when you turn up the volume, it makes it much easier to clip the chit out of everything, seeing as most music is recorded at a level of at least -6dB.

I know Dave doesn't make things up... so I'd like to hear the reasoning

[DBfan187 0]

grrrrrrrr you making dig up this thread.

hold on.

[DBfan187 0]

oh there's too many threads to link to.

So just search his forum.

http://www.carsound.com/UBB/ultimatebb.php?ubb=forum;f=2

[DBfan187 0]

little bit of it.

http://www.carsound.com/cgi-bin/UBB_CGI/ul...t=003671#000000

[Nikuk 1]

he generally knows his chit, so I have no place to negate him.

I however will continue to use an impedance chart & 0dB tones.

I don't understand his reasoning yet (even with the link), i do understand mine.

<shrug>

[Supa_c 1]

I dont get it....

No wonder i set my gain by eye...

[///M5 2,833]

Uhh, other than his statement of

"we now must switch the 0dB tones for -10dB ones"

There are no reasons, nor does it make any sense. Bottom line when you set up your HU, EQ, or amps you don't want any of the above to clip. The post was logical in separating each of the signals independently to make sure that you eliminate the possibility of clipping at each possible source.

[Impious 1,476]

Yeah...I don't understand the 3:1 ratio myself. Would be nice if you (I...someone in general) could find more info on the rational behind it.

Dave Navone is a trustworthy source, I'd generally listen to what he has to say.....but this one doesn't make sense in my simpleton mind.

[///M5 2,833]

I won't knock Dave either; however, since the quote was so short there is some real important missing info...

[DBfan187 0]

half-assed answered - http://www.carsound.com/cgi-bin/UBB_CGI/ul...t=002743#000000

The man Mr. Clark chimed in on this one - http://www.carsound.com/cgi-bin/UBB_CGI/ul...t=002373#000000

[Nikuk 1]

much better link..

http://www.carsound.com/cgi-bin/UBB_CGI/ul...t=002373#000000

This makes much more sense now.

It's early for me right now, and I'm a little hung over. I'll try to put together a long pucking post about how I understand it. You guys tell me if it'sright.

[Nikuk 1]

WARNING: This post contains generalizations!

To start:

0dB 1khz tone amplified @ 25 watts = 100dB SPL

-3dB 1khz tone amplified @ 25 watts = 97db SPL

-6dB 1khz tone amplified @ 25 watts = 94 dB SPL

To make a -3dB 1khz tone "as loud as" the 0dB tone, You'd need to double power. So:

0dB 1khz tone amplified @ 25 watts = 100dB

-3dB 1khz tone amplified @ 50 watts = 100dB

-6 dB 1khz tone amplified @ 100 watts = 100dB

Right?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Let's say we have an amplifier as follows:

6.0vac input signal will result in 100watts per ch. @ 4ohm.

Regardless of input, to make 100 watts @ 4 ohm You'll need 20vac out of the amp's voltage rails (wattage = voltage^2 / resistance). So, we know that with a 1khz 6.0 vac input signal, the voltage rails will be at full swing (no clipped wave) of 20vac. This is the limit of what this amp can do (AC wave).

Now, amplifiers don't see a dB reference, they only see input voltage & amperage. Now, a -3dB tone out of the HU will pass less V & A into an amp then a 0dB tone (given same HU Vol. Pot. setting). The impedance of the input stage of an amplifier won't change.

Assume that a HU will output 6.0 vac (through the patch cables) @ the amplifier's impedance, with a 1khz tone recorded @ 0dB.. playing vol. 30.

Now, if all other things remain constant, but You decrease the media info by 3dB reference, then the HU will no longer be outputting 6.0vac @ vol. 30.

Let's assume that with a -3dB 1khz tone @ vol. 30 from the HU means an input voltage of 5.0 vac into the amp's input impedence. The resulting SPL is 3dB down from a 6.0 vac signal.

But the rule of thumb is that to gain 3dB You have to double power. So, to lose 3dB You'd have to cut power by half. Now, the rails of the amp putting out 14vac accroding to that generally accepted rule.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This is where I get truly lost:

It's generally accepted that most commercial media is not recorded at 0dB reference, but actually somewhere well below that level.

However, there is ALOT of information in the media for any given moment. So there are transients & events in the media that could result in a higher output signal by the HU (at vol. 30).

--==BIG VOID IN KNOWLEDG BASE==--

This "stronger" signal into the amplifier will result in more voltage thrown across the rails of the amp, and therefore higher output from the amp.

If You set the sensitivity of the amp with a 0dB reference (or 6.0Vac input), then You are now clipping the waveform - albeit momentarily.

If you set the sensitivty of the amp with a -3dB reference (or 5.0Vac input), then the voltage of the rails is still of the AC variety... no clipping / distortion.

However, Your average SPL is also 3dB down. Lets say 97dB instead of 100dB.

This is where having larger amplifiers with "headroom" comes into play. You can set things up for a higher wattage to compensate for the lower reference level of the signal... resulting in the same SPL as that of a of a "less powerful" amp which is being further "stressed".

God I hope that makes sense...

-Nick

[Impious 1,476]

Well, I haven't read the links yet.

But, going by your post Nick, he's basically saying that due to the dynamics in music you want a more powerful amp with a lower gain setting (-10db rather than 0db, resulting in less output from the amp compared to it's maximum available output) so that the amp won't clip when it encounters these very dynamic transients ??

Close?? But that doesn't seem right....because then wouldn't the preamp equipment still be clipping since they aren't set using the -10db test tone?

Oh well...off to read those other links before I get totally confused.....

Wait....now I feel like an idiot. If you used a -10db setting, the amp gain would be higher, not lower. So now my entire post doesn't makes sense.......

*goes to read links*

[Nikuk 1]

read the one link that I posted... that helped me the most.

I'm not saying that -10dB reference is the end all - be all right way to doit, just that I see his point... i think... lol

[Nikuk 1]

wait, i just re-read my post and I think that i ended up saying You have to set the gain higher... i'm so confused right now. lol. but my hangover's almost gone.

[Impious 1,476]

Okay...I think I get it.

In short; Since music is recorded at something less than 0db (generally in the vicinity of -6db), by using the standard 0db test tone to set the amplifier gains we end up using much less of the amplifiers available power for a given volume setting. So if we set everything prior to the amplifier for maximum unclipped output with a 0db (simple terms; it will never clip with music) and then *increase* the amplifier gains to compensate for the decreased recording level we have the best mix of unclipped signals and maximum power output from the amplifier for a given volume setting. The amplifier *may* clip with this setting....but due to the short time period of the event, we will hardly ever hear it and the effects won't be damaging to the equipment (i.e. no blown speakers).

My interpretation sound right to anyone else??

However, this works contrary to standard SNR theory, where you want the gain in your system as early as possible rather than later since each device in the chain not only adds it's own noise to the system, but also amplifies the noise from the previous components. According to what I've read previously, the method RC and DN recommend would result in the worst SNR, I believe, since all of the system gain is in the power amplifier.

Here is a quote:

HIGHER gain setting on the power amp means MORE noise or worse SNR ... providing that the comparison is done at equivalent output levels.

http://forum.elitecaraudio.com/showthread....05&pagenumber=7

[Nikuk 1]

My interpretation sound right to anyone else??

Yes...

However, this works contrary to standard noise theory, where you want the gain in your system a early as possible rather than later since each device in the chain not only adds it's own noise to the system, but also amplifies the noise from the previous components.  According to what I've read previously, the method RC and DN recommend would result in the worst SNR, I believe, since all of the system gain is in the amplifier.

<{POST_SNAPBACK}>

See, I'm seeing it the other way 'round.

If You set allof the preamp equip @ full sensitivity to the HU, then You are increasing chance of noise. By running EQ's and amps at a higher sensitivity for a given input voltage, You are increasing the likelyhood of noise amplification.

Whereas, with this method, You are trying to minimize all the preamp levels and merely amplifying induced noise at the end of the processing chain.

i.e.- cranking the amp gain at a HU vol. setting of 2 will lead to much more audible noise then a 'reasonable' gain setting at HU vol. 30.

<shrug> x2.

[Impious 1,476]

<<----- 666

See, I'm seeing it the other way 'round.

If You set allof the preamp equip @ full sensitivity to the HU, then You are increasing chance of noise. By running EQ's and amps at a higher sensitivity for a given input voltage, You are increasing the likelyhood of noise amplification.

Whereas, with this method, You are trying to minimize all the preamp levels and merely amplifying induced noise at the end of the processing chain.

i.e.- cranking the amp gain at a HU vol. setting of 2 will lead to much more audible noise then a 'reasonable' gain setting at HU vol. 30.

<shrug> x2.

You got me before my edit

http://forum.elitecaraudio.com/showthread....05&pagenumber=6

start reading at werewolf's posts. It explains better and has some nifty math involved.

[Supa_c 1]

That doesnt seem right.....

When i go to turn the volume dial i dont want it so loud at such a low volume number.

More gain= more volume at a lower volume setting on the H/U

you want a nice gradual increase in volume when you turn the voume dial, not a big hump going from lets say 10 to 12.

[Nikuk 1]

^ ?

There is no big hump...

[DBfan187 0]

sO, WHAT IS EVERYONE'S OPINION ON THIS SUBJUCT?