Searching for problems in fiber, splicing or connector

[mimonici]

I've made fiber laser system. Previously, there's no problem in my set up and achieved good result.
Further, when I want to repeat that experiment with similar set up, spectrum shows quite worse than previous.
I'd try to check,change certain new fiber components,connectors but it still worse.
How can I improve that, to find out what actually the probable problem?

 

[sophiecentaur]

Hi
You don't include any details of your system so I don't think anyone can help you much, yet. There are two approaches to 'something not working'. You can replace one component at a time (why car repairs are so expensive) or examine each piece in turn, testing the other pieces in alternative configurations. If you have any test equipment, then you have to sit down and use it to eliminate one element at a time until you can zero in on the cause of the problem.
A very generic answer to a very generic question.

 

[Claude Bile]

Some things to think about;

Photo-darkening - prolific at short wavelengths.
Damage to the end-facets - prolific at high powers.

As sophiecentaur said, it is difficult to guess with limited details of the problem.

Claude.

 

[mimonici]

Thank u for concern to both of u gentlemen. Alright, now I'll make it clear. I am configuring a set up of fiber laser. In the configuration contains optical components; 90:10 couplers, single mode fiber, erbium-doped fiber(s), optical circulators, laser diode(s), wavelength-division multiplexing(s), 3dB coupler, Polarization Controller and Polarization Maintaining Fiber. Certain of the components i'd spliced by using optical splicer and certain were connected with connector. I'd analyze the output by optical spectrum analyzer(OSA).
This configuration I've made to achieve broad wavelength range of spectrum of the fiber laser as 11+nm wavelength. I've got the desired result previously but unable to repeat day by day. So, after discussed, I think the desired obtained result was not clear enough because of wrong span and resolution set from the OSA. Thats why I repeat the experiment to obtain same result with improved span and resolution of the OSA. Consequently, it failed. I really need the significant data.. Any suggestion to detect what are actually the problems of my set up?