In the context of optical amplifiers, what's the difference between gain bandwidth and amplification bandwidth?
December 28, 2004 2:43 PM Subscribe
In the context of optical amplifiers, what's the difference between gain bandwidth and amplification bandwidth?
I've found a bunch of technical material that talks about one, the other, or both, in ways that make it fairly clear they are not the same thing, but I haven't yet found a pair of concise definitions or a description of the difference.
I've found a bunch of technical material that talks about one, the other, or both, in ways that make it fairly clear they are not the same thing, but I haven't yet found a pair of concise definitions or a description of the difference.
Gain bandwidth is a very specific thing. Most amplifiers have their greatest gain in low frequencies. As you go out in frequency eventually you'll reach a point where the amplification is only 1. This is the unity gain frequency or the gain bandwidth. This is from voltage or current amplifiers but it will be the same definition for optical amplifiers.
I'm drawing a blank on amplification bandwidth. I have some optical electronics texts but they're 650 miles away at the moment. I've seen a couple references via google but they refer to a couple of different things!
posted by substrate at 6:23 PM on December 28, 2004
I'm drawing a blank on amplification bandwidth. I have some optical electronics texts but they're 650 miles away at the moment. I've seen a couple references via google but they refer to a couple of different things!
posted by substrate at 6:23 PM on December 28, 2004
I'd say that onalark is correct and substrate is close. Even though "gain bandwidth and "amplication bandwidth" sound like two different measures in the same units, the former is indeed a product (units = db Hz) and the latter is a bandwidth (units = Hz).
You might want to post this in a more techie forum like one of the sci.* newsgroups. Remember newsgroups? Ahhhh ....
Note: I got mostly A's in pursuit of my EE degree, but if someone makes a good post that conflicts with mine, I'll defer.
posted by intermod at 7:04 PM on December 28, 2004
You might want to post this in a more techie forum like one of the sci.* newsgroups. Remember newsgroups? Ahhhh ....
Note: I got mostly A's in pursuit of my EE degree, but if someone makes a good post that conflicts with mine, I'll defer.
posted by intermod at 7:04 PM on December 28, 2004
Response by poster: Thanks, all. I did post in sci.optics and sci.electronics.misc, but no results yet. If the units are different (as intermod says), then I may be looking at a patent that is not internally consistent. No surprise there: many are written by lawyers who never actually practiced engineering (let alone got good at it! Rimshot!)
posted by spacewrench at 7:22 PM on December 28, 2004
posted by spacewrench at 7:22 PM on December 28, 2004
Even though "gain bandwidth and "amplication bandwidth" sound like two different measures in the same units, the former is indeed a product (units = db Hz) and the latter is a bandwidth (units = Hz).
Not quite. The gain-bandwidth product is the frequency range over which the gain is above unity. Since unity gain is zero dB, the units can't be in dB-Hz, as that would make the gain-bandwidth product zero. The unit is Hz since the gain unit is dimensionless.
The amplification bandwidth is the frequency range over which the gain is within 3 dB of the maximum gain. 3 dB has been chosen because -3 dB (i.e., 3 dB below the maximum) corresponds to half-power. (dB = 10 log (p_out/p_in); the common log of 2 is 0.3010.)
The gain-bandwidth product is generally much larger than than the amplifier bandwidth. The amplifier bandwidth can be increased to nearly the gain-bandwidth product by using dominant-pole compensation, at the expense of maximum low frequency gain.
That should clear everything up!
posted by Wet Spot at 8:20 PM on December 28, 2004
Not quite. The gain-bandwidth product is the frequency range over which the gain is above unity. Since unity gain is zero dB, the units can't be in dB-Hz, as that would make the gain-bandwidth product zero. The unit is Hz since the gain unit is dimensionless.
The amplification bandwidth is the frequency range over which the gain is within 3 dB of the maximum gain. 3 dB has been chosen because -3 dB (i.e., 3 dB below the maximum) corresponds to half-power. (dB = 10 log (p_out/p_in); the common log of 2 is 0.3010.)
The gain-bandwidth product is generally much larger than than the amplifier bandwidth. The amplifier bandwidth can be increased to nearly the gain-bandwidth product by using dominant-pole compensation, at the expense of maximum low frequency gain.
That should clear everything up!
posted by Wet Spot at 8:20 PM on December 28, 2004
This thread is closed to new comments.
I'm guessing the gain bandwidth term you're referring to is actually the 'gain-bandwidth product'. The gain times the bandwidth for an amplifier in all configurations equals this number.
Just a forewarning, I got a C in the Electrical Engineering class relevant to this, but maybe this will get you pointed in the right direction.
posted by onalark at 4:43 PM on December 28, 2004