How can I use a fiber EOM to lock a laser with a large input power?

I am trying to buy an EOM for a PDH lock with a large laser input power and I was told I should go for a free space EOM. I see that they come as resonant and non-resonant, but I can't seem to understand the advantage of a non-resonant one. The resonant one seems to do pretty much everything the non-resonant one does, but with a lot less voltage needed for modulation so it seems like an obvious choice. What are the advantages of a non-resonant one? Thank you!

I think the resonant one will apply the required modulation voltage to the EOM with expenditure of less power. Because it uses resonance to do this, it is a narrow band device, which will restrict the modulation bandwidth. This might or might not be important to your usage.

I haven't used one myself, but all of my coworkers who have ended up switching to fiber EOMs just due to the sheer hassle (getting the requisite voltage, and alignment issues resulting in residual amplitude modulation). I think the only good reason to use a free space EOM is if you absolutely need more optical power than what you can get with a fiber EOM.

I know this isn't what you asked about, but I encourage you to make sure that's what you need before investing a lot of time.

Twigg said:

I haven't used one myself, but all of my coworkers who have ended up switching to fiber EOMs just due to the sheer hassle (getting the requisite voltage, and alignment issues resulting in residual amplitude modulation). I think the only good reason to use a free space EOM is if you absolutely need more optical power than what you can get with a fiber EOM.

I know this isn't what you asked about, but I encourage you to make sure that's what you need before investing a lot of time.

Yeah that was the main reason for me. The power input is ~10W/mm##^2##, I don't think there is a fiber able to do that, no?

But what I need this for is to lock my laser to a bow-tie cavity using PDH. Is there a way to somehow send most of the power inside the cavity without the EOM in the way, and take a very tiny amount of it, and somehow send it inside the cavity through some other mirror, such that you use this small amount for locking, while the main laser beam will just follow the cavity as it is in phase with the other part of the beam? However it sounds challenging to align both path to the same cavity. You can align the mirrors for one path, but for the other you'd need a perfect input to the cavity, as you don't have the freedom to adjust the mirrors anymore, right?

BillKet said:

The power input is ~10W/mm2, I don't think there is a fiber able to do that, no?

Yeah, no. That'll make some nice fiber flambé. You had the right idea going free space.

But you can actually still make it work with a fiber EOM.

BillKet said:

Is there a way to somehow send most of the power inside the cavity without the EOM in the way, and take a very tiny amount of it, and somehow send it inside the cavity through some other mirror, such that you use this small amount for locking, while the main laser beam will just follow the cavity as it is in phase with the other part of the beam?

Sure you can. You basically already said it. Split a milliwatt or two off your main beam and send it into a fiber EOM (or free space EOM, whichever ends up being cheaper and less hassle). Do phase modulation on the low power beam for PDH and send it into the cavity. Now you can use that PDH signal to lock your laser. Then take the main beam (which is now resonant, thanks to your PDH lock) and send it into the reverse-propagating cavity mode. Use an optical isolator to protect the fiber EOM from the bulk of the main beam. I put a sketch at the end of the post. The blue beam is the PDH beam, and the red beam is the high power beam.

BillKet said:

However it sounds challenging to align both path to the same cavity.

With a little strategery, aligning the second beam is much easier than aligning the first. Take a look at the transmitted PDH beam in my sketch (dotted blue line). This beam (if there's enough light for it to be visible) can be used as an alignment reference to couple the main beam (red line) to the cavity. Just adjust mirrors M3 and M4 to overlap the main beam with the transmitted part of the PDH beam. (If the main beam is too bright and you can't see where it overlaps the PDH beam on a card, then just use an ND filter temporarily to bring it to lower power without dropping the PDH beam power.)

BillKet said:

You can align the mirrors for one path, but for the other you'd need a perfect input to the cavity, as you don't have the freedom to adjust the mirrors anymore, right?

In the setup I have in mind, you would have independent degrees of freedom for both beams (see sketch below). The mirrors labeled M1 and M2 allow you to couple in the PDH beam, and the mirrors M3 and M4 allow you to couple in the big beam.


Edit: Oh, you might want a lambda-by-2 waveplate on the high power beam to keep it on the same polarization mode as the PDH beam. Otherwise, your beam won't be exactly on resonance due to the splitting of resonances for different polarizations. Annoying quirk of bowtie cavities.