Gain saturation

[Steve]

Hi

I have a few questions about gain saturation and pump power depletion:

How is the "gain saturation due to limited pump power" defined? Is
this the amount of energy deposited within the upper laser level along
the entire length of the amplifying medium, within a certain
timescale? If so, which timescale? 1 second, upper state lifetime,
time between signal pulses?

Next, I'm interested in modelling pulse propagation through a fibre
amplifier. I would like to be able to model the varying gain along the
fibre due to pump depletion. Is it possible to do this within the
standard functions of the programme, or do I have to look at defining
multiple sections of fibre with decreasing gain values to tend towards
a solution?

Cheers,

Steve

[webm...@fiberdesk.com]

Hi Steve,

I'm sorry, the description of the value [gain saturation due to
limited pump power]=E_s was missing in the manual. You can now
download it from www.fiberdesk.com.
It is modeled by a saturable gain g=g0/(1+E/E_s) were g0 is the small
signal gain given in the propagation parameter window and E is the
pulse energy. It account for the effect that an amplifier does not
have the same gain for an arbitrary input pulse energy but saturates,
e.g. to half of its value for E=E_s. (or similar to this, the gain is
limited for a given pump power). The effect is only working if the
option [gain sat] is switched on.

You might be able to simulate pump depletion with this equation as it
saturates the gain during propagation through the fiber, if the pulse
energy increases and becomes comparable to the value of E_s.

I have to note the following: the right value for E_s should be found
by "try and error" in most cases, because the output energy is also
determined by additional effects like losses, spectral broadening and
gain profiles occurring during propagation. Also, don't mix E_s and
E_sat used for [temp gain sat]. Please see the new manual for details.

I hope that helps,

BR

Thomas