ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 [2]  All   Go Down

Author Topic: Program & Peak, Amplifier Gain, and other stuff I need to know better...  (Read 5705 times)

Marlow Wilson

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • Ottawa/Montreal CAN
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #10 on: February 04, 2013, 07:24:49 pm »

Formula: (v1 here is amp output voltage, v2 is input or dsp output & dB is known voltage gain)

General formula:
dB = 20 x Log (v1/v2)

working backwards:

voltage multiplier = antilog10 (dB/20)
(NOTE - I fixed an error with the formula, my apologies... I was typing carelessly)
v1 = voltage multiplier x v2

To add the itech back into the equation for a minute, assume you've determined that a peak limiter of 100 volts and an RMS limiter of 25 volts was appropriate.  With the itech hd, you just enter that in System Architect (lets ignore the attack release settings for now).  Whatever happens upstream, the amp 'knows' what the actual resulting output voltage is and limits accordingly.  Going back to your D&B example, the 'sense lines' allow the amp/DSP to adjust for the voltage loss over the cable to make the ACTUAL voltage at the speaker exactly what the system designer intended.  With 'dumb' amps we have to work back a few steps to figure it all out.

So going back to our PL380/GX3 at 32dB gain,

the voltage multiplier =  10(32/20)
=39.81

for a final voltage of 25v RMS:

25 = 39.81 x v2

Therefore,

v2 = 25/39.81

Therefore,

v2 = 0.63.
The complex relationship between the peak and RMS voltage of multiple frequencies in actual music is glossed over here, but I think it demonstrates the point:

As long as the input RMS value doesn't exceed 0.63 volts, the amp won't put out more than 25v RMS.  If this was the actual value desired, you'd see that the GX3 might actually be able to produce more voltage long term than you would have allowed with the I-tech, despite being a pretty 'small' amp, especially when compared with the I-tech.

The same calculation for the peak value is 2.51 volts.  Of course the Gx3 would be clipping, but the PL380 could probably reproduce quite a bit more for a short burst.

Make sense?



Logged

Marlow Wilson

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • Ottawa/Montreal CAN
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #11 on: February 04, 2013, 07:55:31 pm »

But limiting alone cannot let you power a JRX wedge with an I-Tech, can it?

I've used itechs to power surprisingly small speakers.  I've also used surprisingly low RMS limiter values on surprisingly big speakers.

Quote
Limiting is a dynamic reduction of gain you'd need some kind of static reduction in addition to appropriate limiting.

If the amp gain is too high and the limiter threshold very low, you'd essentially have too 'hot' a device and have to run the mixer at correspondingly low levels to keep it out of limiting.  The result is a bad gain structure.  Conversely, if the gain is too low and you have to run your mixer into, or near, clipping to get full output you've then got a bad gain structure on the other end of the spectrum.

The precise balance of what is 'best' is contemplated daily by installers.  Depending on the application the 'best' setting may require a specific gain setting where available and a reduction of a few dB with the attenuators.  You'll find many can get 'close enough' with no attenuation and avoid the potential mischief from the allure for bystanders to 'turn it up'.

Quote
A GX7 does not have a variable gain, which I think is this static variable. An I-Tech does, correct?

I'm not sure what you mean by static variable?

Cheers!
« Last Edit: February 04, 2013, 08:01:48 pm by Marlow Wilson »
Logged

Samuel Rees

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1879
  • Washington, D.C.
Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #12 on: February 05, 2013, 12:16:02 am »

Remove the itech from consideration.  Let's say you want to power your JRX cab, and you've got a QSC GX3 and a PL380.  Two extremes, I know.  Let's say both are at 32db (the GX3 is stuck there, the PL380 can join it) and no attenuation.

Thanks guys. A lot to take in re: those formulas. I get the limiting.

If the two amps had the same gain (even if one was fixed, one set) wouldn't they have the same output? Ok... I think I know that's wrong, but I just don't understand why. It's this gain number I don't get... Maybe it will take more than a forum post to cram it into my head. Is there somewhere I can read about this? I can't seem to find the kinda of description I'm looking for. Tim, any thoughts?

Haha or does anyone and DC want to sit down with me and an amp/DSP and educate me? There's be dinner and beer in it....


Logged

Marlow Wilson

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • Ottawa/Montreal CAN
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #13 on: February 05, 2013, 11:17:38 am »

If the two amps had the same gain (even if one was fixed, one set) wouldn't they have the same output? Ok... I think I know that's wrong, but I just don't understand why. It's this gain number I don't get...

Both would have the same output up until the point where the less powerful amp starts clipping.  For the Gx3, any input beyond the 1.2v RMS range would cause the amp to clip.  Therefore, a 2v RMS sine wave would have the Gx3 blinking red, while the PL380 would happily (at least in theory) reproduce the signal without issue.  The resulting output from input signal greater than 1.2v IS NOT THE SAME.

In terms of the gain number, I picked 32dB because it also happens to be Peavey's preferred gain.  To make it easier to understand, they refer to it as '40x'. 40x is really just the multiplier we calculated: 39.81.  In Peavey terms, with any Peavey 40x amplifier, 1 volt of input signal will result in 40 volts of output.  Does that make more sense?  Most Crest/Peavey amps are 40x, meaning for the same low input they will all make the same low output.  As you move up to more powerful amplifiers, you essentially buying the ability to faithfully reproduce higher input voltage (resulting in higher output voltage).  The gain never changes.

Where it gets confusing is the dB scale, which isn't exactly intuitive for most people.  If you want a resource, you'd really have to start with the Yamaha guide or similar resource to make sure to get your head around the dB scale and its application to pro audio.

The other confusing aspect is that not all amps have a fixed gain of 32dB (40x) or a similar fixed gain within a series.  RATHER than having the amps clip at different input voltages, the amps clip from the SAME input voltage.  A common number would be 1.2 or 1.4 volts.   What that means is that the small amp might produce 25v at 1.2v of input, while the larger amp produces 50v with the same input.  RATHER than a FIXED multiplier, the multiplier grows as you move from small to large amps within a series.  With the Gx series, the Gx3 just happens to have a sensitivity that results in 32dB of gain (it's actually 32.2, which actually result in a 40.74 multiplier). The Gx5 is 34.4 dB, which is a 52.48 multiplier. The gx7 is 36.1 dB, which is 63.83. 

With the fixed, 40x / 32dB, gain across multiple amplifier if you want more power from the higher power amps you have to send them more signal.

With the 1.2v sensitivity, all amps produce full power with the same 1.2v. 

As an example, lets say you had the Gx3 powering a really low power-handling HF driver and were using an advanced limiter to limit to 1.0 v RMS and 1.2 v peak signal voltage.  You could swap in a PL380, set at the same 32dB, and wouldn't need to change anything, right?  The limiter would not let you send any voltage in excess of what the Gx3 was receiving, and the output would be the same (pretty much).  You couldn't get over 40v RMS, despite having completely different amps.

But you couldn't just swap in a Gx5 or gx7 without changing something, because the same signal would result in higher output voltage because of their higher gain.  The same settings would allow over 52.48 and 63.81 volts RMS from the Gx5 and Gx7 respectively.  Despite being smaller amps, you'd risk frying your speaker potentially and all spectral/tonal balance would be thrown off.
Logged

Paul G. OBrien

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1069
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #14 on: February 05, 2013, 11:41:43 am »

How did he end up tearing up my garden en route through my bay window?   ::)

Ha!! He's the driving equivalent to a DJ. ;D
Logged

Samuel Rees

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1879
  • Washington, D.C.
Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #15 on: February 05, 2013, 12:09:16 pm »

So what gain would the PL380 be at to get full power? Set that way, it would have higher output then the GX7 @ 1.2 volts input right?

Thanks I realize this is taking forever, thanks!
Logged

Marlow Wilson

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • Ottawa/Montreal CAN
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #16 on: February 05, 2013, 03:55:55 pm »

So what gain would the PL380 be at to get full power? Set that way, it would have higher output then the GX7 @ 1.2 volts input right?

Thanks I realize this is taking forever, thanks!

Set what way?  I will go through both scenarios below.  Note, however, that while these calculations come up with precise numbers, in practice 'full' power is not easily defined.  Amps can produce much higher voltage for very short bursts, and not all amps produce equal output at every frequency. 

(1) 32 dB Gain setting

(A)   At the 32 dB gain setting, the PL380 would have lower output than the Gx7, which has 36.1 db of gain, for the same input signal BELOW a certain threshold (calculated in (B)). 

See, for example, the output at 1.2v:

1.2 with gx7 = 63.83 x 1.2 = 76.6volts in theory

1.2 with pl380@ 32dB = 39.83 x 1.2 = 47.7 volts in theory

(B)    After 1.2v RMS, the Gx7 would be clipping.  At some point beyond 1.2v the voltage output of the PL380 would exceed the Gx7. To see what input voltage would be required to match the clipping Gx7's max output:

v2 = multiplier x v1

76.6 (which is the voltage of the Gx7 we are trying to match) = 39.83 x v1

v1 = 1.92 volts. 

Therefore at 32 dB, 1.92v of input would be required to 'match' the output of the gx7. Anything beyond 1.92 will allow the pl380 to surpass the max output of the Gx7. Remember, with the Gx7, anything beyond 1.2v causes clipping in theory (in practice it's more complicated depending on the source: pink noise, sine waves and snare drums don't all punish amps in the same way).  Only after 1.92v does the PL380 'catch up' however.

I think the max RMS voltage of the PL380 is 90v, therefore 'full power' at 32 dB requires how much voltage?

v2 = multiplier x v1

90v = 39.83 x v1

v1 = 2.26v

Therefore the PL380 can produce 'full' RMS power with 2.26v of input signal.

(2) 1.2v sensitivity setting.

If set at 1.2v sensitivity input, the PL380 has 39.1 dB of gain, and like the Gx series, would reach that 'full power' with 1.2v.

Takeaway:

"FULL" power can be reached both ways.  2.26v @ 32dB, and 1.2v at 1.2v sensitivity (39.1 dB of gain).

At 32 dB, the pl380 would have less GAIN and therefore less output for any input below 1.92 volts.  The GX7 has more gain (36.1 dB), but gives up after 1.2v.

At 1.2v sensitivity setting, it would have MORE GAIN and therefore would surpass the Gx7 at every input voltage.
« Last Edit: February 05, 2013, 04:41:17 pm by Marlow Wilson »
Logged

Tim McCulloch

  • SR Forums
  • Hero Member
  • *
  • Offline Offline
  • Posts: 21570
  • Wichita, Kansas USA
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #17 on: February 05, 2013, 04:27:18 pm »

In other words, "X" gain is linear, "dB" gain is logarithmic.

Logged
"Practicing an art, no matter how well or badly, is a way to make your soul grow, for heaven's sake. Sing in the shower. Dance to the radio. Tell stories. Write a poem to a friend, even a lousy poem. Do it as well as you possible can. You will get an enormous reward. You will have created something."  - Kurt Vonnegut

Marlow Wilson

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
  • Ottawa/Montreal CAN
Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #18 on: February 05, 2013, 06:34:14 pm »

So what gain would the PL380 be at to get full power? Set that way, it would have higher output then the GX7 @ 1.2 volts input right?

Thanks I realize this is taking forever, thanks!

Maybe easier than the written out explanation: the attached chart shows an approximation of what happens when the the Gx3, Gx7 and PL380 at 1.2v & 32dB is driven with a given amount of voltage.

See how 'full' power is reached both ways?
« Last Edit: February 05, 2013, 07:24:51 pm by Marlow Wilson »
Logged

ProSoundWeb Community

Re: Program & Peak, Amplifier Gain, and other stuff I need to know better...
« Reply #18 on: February 05, 2013, 06:34:14 pm »


Pages: 1 [2]  All   Go Up
 



Page created in 0.062 seconds with 24 queries.