445nm diode info

[krazer]

My understanding is that he is refering to the longitudinal modes, not the latitudinal ones that are causing headaches for those trying to get a decent beam. The physical shape of the beam does not change after the diode has warmed up over threshold.

[Krutz]

my bad! thank you for the hint. will have to re-read all that again..

manuel

[RedlumX]

would love to see a vid of the raw output when slowly ramping up the current..
Manuel

I did this today, see the little movie that shows the mode spectrum when the current is ramped from threshold to approx 260mA; note the beautiful single mode zone at 224mA:
http://pagesperso-orange.fr/redlum.x...reerunning.MOV

Included is also sound that shows the noise in the light output. The popping sounds at the beginning indicate mode jumps. One can clearly hear how the single mode regions are correlated with low noise, which gives hope that a simple noise detector might be sufficient to determine the single mode regions (this is not so for other diodes).

My understanding is that he is refering to the longitudinal modes, not the latitudinal ones that are causing headaches for those trying to get a decent beam. The physical shape of the beam does not change after the diode has warmed up over threshold.

Indeed so, it's about the longitudinal modes. I found this surprising as, I believe, this is a multi-emitter diode and why all emitters should run at the same wavelength escapes me. The diode seems to behave like any other higher power red or 405nm diode in this respect. Perhaps there is some reason behind this. But first of all, more detailed and systematic studies are needed, in particular each diode is different and I know from the red diodes that different samples of the same model can behave very differently.

[mixedgas]

Littman Metcalf cavity anyone? I proposed it last week, but will not have a chance until next saturday to try it.

Steve

[Eidetic]

Indeed so, it's about the longitudinal modes. I found this surprising as, I believe, this is a multi-emitter diode and why all emitters should run at the same wavelength escapes me. The diode seems to behave like any other higher power red or 405nm diode in this respect. Perhaps there is some reason behind this. But first of all, more detailed and systematic studies are needed, in particular each diode is different and I know from the red diodes that different samples of the same model can behave very differently.

Could it be that as the current is decreased that the number of emitters lasing also decreases? I see that with Novalux lasers. Image the emitters out to a screen with a lens and watch them as you decrease the current. At the very lowest currents, maybe only one emitter is working and therefore very coherent.

[Roy G. Biv]

Littman Metcalf cavity anyone?

Would this somehow be better for SLM than a Littrow design?

[RedlumX]

Could it be that as the current is decreased that the number of emitters lasing also decreases? I see that with Novalux lasers. Image the emitters out to a screen with a lens and watch them as you decrease the current. At the very lowest currents, maybe only one emitter is working and therefore very coherent.

Yes I was pondering about that as well, but at least with the naked eye I didn't see any change in the illumination pattern. In fact the region of SLM operation is pretty large, essentially from threshold 180mA to 220mA with some gaps, and I don't think that only a single emitter would lase through this region. Perhaps some LD expert here has some insight in what is going on.

Also, when scanning through the various lobes I didnt see a shift in wavelength. So it really looks like the SLM red diodes I am familiar with. (Well one thing: my spectum analyzer resolves a few pm which translates to a few Ghz. I can't exclude that there are different modes with slighly different frequency; I still need to find suitable mirrors for my scanning interferometer, that would resolve to a few Mhz then.)

Would this somehow be better for SLM than a Littrow design?

Well it would have less power due to an extra output beam, but it may have an even smaller line width.

I am trying variuos Littrow ECDL configurations right now but so far the results are not significantly better than for the free running diode, actually the thing is quite unstable - but as always, the grating, collimator, resonater length need to be carefully optimized which is a lot of time consuming work.

[mixedgas]

RedlumX, He said and I quote:

"Also, when scanning through the various lobes I didnt see a shift in wavelength. So it really looks like the SLM red diodes I am familiar with. (Well one thing: my spectum analyzer resolves a few pm which translates to a few Ghz. I can't exclude that there are different modes with slighly different frequency; I still need to find suitable mirrors for my scanning interferometer, that would resolve to a few Mhz then."

What radius ? What transmission is acceptable? I have about 100 different ion laser mirrors in my personal collection. And access to perhaps 500 more if I really need one. The problem is argon HRs are usually frosted.

Possibly add a 1064 nm HR as a etalon to your cavity? The 1064 is just about a 10% reflection at that wavelength and usually they have no wedge.

Steve

[RedlumX]

What radius ? What transmission is acceptable? I have about 100 different ion laser mirrors in my personal collection. And access to perhaps 500 more if I really need one. The problem is argon HRs are usually frosted.

Possibly add a 1064 nm HR as a etalon to your cavity? The 1064 is just about a 10% reflection at that wavelength and usually they have no wedge.

Well the radius should be a few cm or so, with 0.1-1% transmission or so. These are hard to come by, actually I got mirrors from dye lasers that I had used successfully used for 488nm (Spectra Physics G3845-002, f=50mm, for pumping by argon), so I will try those first and hope the transmission at 445nm is still low enough such as to get a high resolution (finesse). BTW frosted mirrors can be dealt with by gluing a small glass plate (piece of microscope slide) to the back, with some clear glue. Efficiency isn't here important, fortunately.

And playing with etalons... I guess there are zillions of different things one could try ;-)

One thing that bothers me for blue ECDL, among others, is that the collimation of the diode must be very good, I have read an article somewhere that if the waist is at 2m and not infinity, this already gives an 80% loss of backcoupling efficiency. Now given the very different divergences of the two axes I wonder how to collimate optimally...maybe this explains why my ECDL attempts so far gave worse results as for the free running diode. On the other hand, with the grating feedback the threshold current was reduced from 180mA to 145mA, which is a sign of good feedback, in fact more than I am used to for red diodes, perhaps even too much?

I am afraid that this is now going to be OT of the thread ;-(

[kecked]

I've been told that these diodes are +/- 10nm on purpose to combat speckle on the screen. So they could be 440-450nm depending on your luck. Don't forget that there are 50mw single mode 445nm diodes out there. I have one and the beam quality is really great. Nobody needs a watt in their livingroom. I have about 150mw whitelight from diodes and it is plenty bright. It would take 10 times that to double the apparent brightness. Is it worth all that? Plus the beam quality will be very low. Now add to that the danger from reflection and those who try to use this for a beam show and you have a bad situation.