How to create super bright laser beams

[kecked]

what if you slightly work the surface of the prism to incorporate a concave or convex surface. Or one of each on either side of the prism. Would that not amplify the angle of separation into the prism so you can spread the diodes out more and have a less sensitive system

[mixedgas]

That could be a pain. is that caused by stray light from one diode "seeding" another?

Yes, and I've seen diodes cross talk like two 16 year old Girls discussing Brad Pitt's boxers.
(sorry, worse, sadistic, cliché image I could come up with!)

Only need a few % to seed for injection locking or mode beating.

Probably not happening in this case, but is a consideration if the gain curves overlap.



Steve

[planters]

The lateral position of the knifed beams is the "fine tuning" adjustment of where the far field spots will overlap. Actually its the angle it enters the prism.

The beams will exit the box as close to parallel as I can get them and about 2.5cm apart. Each will step back the same distance as the diameter of the beam that came from the prism before it. The beams would "normally" be adjusted by these prisms to be exactly parallel and hence the usual composite beam. But, in our case they will be adjusted to converge in the prism. The angle is so small that no other adjustment should be required beside a small turn of the "X" screw.

Any idea how long it will take to reach equilibrium of the thermal gradient in the bar?

Good question. The SS bar weighs 0.37Kg and requires 185J/ degree C. The brass weighs 140gm and requires 53J/ degree C. 238J x the average delta is 1/2 the gradient or 30 C, so 7140J. At 5A the Q-max of the TEC begins at 45J/second and drops to 5J/sec at equilibrium. The curve of temperature with time will not be linear. The last few degrees will take a lot longer than the first few degrees, so the spots should converge quickly at first and the final best spot will be slower in coming and theoretically never arrive. My guess would be something like 3-4 minutes. I specified a really thick SS bar (1CM) and 1/2 this would likely be enough. When the spots became close enough that the beam was acceptable will have to wait for the experiment. Also, the TEC is now a dumb resistor for simplicity. Reducing the power at operational temperature might be a nice feature.

[planters]

Steve,
Where do you envision the seeding originating from, I mean in the layout? I don't doubt you and so I am running the beam through my mind as it were. The windows are flat, but with AR coatings and the beam striking around the originating diode's collimator, the fraction of this scattered light entering the neighboring diode's exit facet should be way, way less than 1%. The knife edge prism's back side is a risk. It is flat and close to the passing beam which is colimated. Need to watch out for this. A wave plate (if used) can be tipped slightly. None of the combining prism's faces will be anywhere close to normal to the entering or exit beams. A PBS should like the wave plate, be tipped slightly. That's my rough analysis. Do you think I am missing anything else?

[planters]

what if you slightly work the surface of the prism to incorporate a concave or convex surface. Or one of each on either side of the prism. Would that not amplify the angle of separation into the prism so you can spread the diodes out more and have a less sensitive system

I'm not able to visualize how this works. I don't disagree, I just am not able to understand it. The prisms can bring in beams that are roughly 1mm apart. This is similar to what we do now. A curved surface in these colimated beams runs the risk of decollimating them and this would have to be compensated, although it could be, at the original collimators.

[logsquared]

The beams will exit the box as close to parallel as I can get them and about 2.5cm apart. Each will step back the same distance as the diameter of the beam that came from the prism before it. The beams would "normally" be adjusted by these prisms to be exactly parallel and hence the usual composite beam. But, in our case they will be adjusted to converge in the prism. The angle is so small that no other adjustment should be required beside a small turn of the "X" screw.

I am pretty sure I understand your logic here.

In my "mental" design there was a space left between each knife edged beam. I would reason that the nm shift is not going to be perfectly spaced. My thinking is that if the "x" screw is turned alone to change the angle it also changes the lateral position it hits the prism (or the virtual apex of the beams inside the prism)and therefore the overlap at the prism exit. Although small, it may (in my brain) be a large percent off due to the small width of the exit beam due to anamorphic effect. Even though not noticeable in the far field it may make the near field too large when re-expanded. I am probably missing something simple here? Or over complicating things?

My OCD brain has been thinking way too much about this for the last 2 days.

Another small issue we may have overlooked is the effect of nm shift due to injection current. I know in one of your videos you touched on this. What effect do you foresee this having on the stability of the final beam?

To throw more fuel on the fire, I had thought it may be possible in theory to use a long glass parallelogram to accomplish the combining. The difference would be the entering beams would be exactly parallel and "stacked" next to each other when entering the glass.

[logsquared]

what if you slightly work the surface of the prism to incorporate a concave or convex surface. Or one of each on either side of the prism. Would that not amplify the angle of separation into the prism so you can spread the diodes out more and have a less sensitive system

I had thought about something similar but with a PCX lens in front of the prism and PC lens after. It is to complex for my brain to wrap around if it would work.

Thinking about it did bring up another question.... When we are thinking of light entering the prism we are generally (me anyway) thinking of the text book drawing of a white line going in and the colors coming out. In fact the diode beam is not perfectly collimated. Different Fl collimators may have any effect on the way the beam is refracted due to the spread of the beam when it interacts with the prism faces???

[planters]

The fine tuning adjustment of the knife edges is really small. Let's take the example I gave above of a 500mm baseline and a 1mm beam. If there is no prism at all and the convergence is 2mrad for each incremental diode on each side of center then this 2mrad is the same for the divergence and even before we turn on the cooler and with the prism gone the red beam will start at the same size as a single diode and the divergence of the (say 5) diodes will be a 8mrad stripe. Now cool the diodes and for the sake of argument in this example nothing happens. Next, insert the prism and rotate it to minimum spread. This is an unknown, but reducing the spread to 2mrad (25%) is a pretty conservative guess. That is the same as 0.5mrad/step or diode. Now, adjusting a diode to eliminate this in the far field requires you to change the angle at the prism by the same deviation and this would be 0.25mm of displacement. This is independent of the anamorphic effect. If the beam is narrowed by that effect the displacement is reduced proportionally and recovered when you re-expand the beam.

Thinking about it did bring up another question.... When we are thinking of light entering the prism we are generally (me anyway) thinking of the text book drawing of a white line going in and the colors coming out. In fact the diode beam is not perfectly collimated. Different Fl collimators may have any effect on the way the beam is refracted due to the spread of the beam when it interacts with the prism faces???

You do want the beams collimated because of the differential angle as you suggest. I think this will, like the finite line width, cause a smear in the far field. The effect will be proportional to the dispersion which is in turn is controlled by the angle of incidence on the prism. The effect will similarly be minimized by a long baseline to the prism. I have a prism analysis program and I'll take a look at the magnitude of this effect.

[jors]

About controlling temp gradients on diodes, I think could be more precise & stable just isolating each diode mount inside a mini-thermal box (a little isolating fiber-box) with TEC/each diode.
Of course, Each diode housing with NTC inside and mini TEC driver.

[tribble]

No, the thermal gradient from one end of the bar to the other is related to the conductivity vs the heat flow and given a finite amount of watts that any cooler can produce, then the maximum thermal gradient occurs when the conductivity is lowest. I looked at the relative conductivity of practical metals allowing enough materiel to be screwed into and stiff enough for the mounts not to move much while say focusing. The gradient for the cooler I looked at needs to be allowing somewhere around 10 W or hopefully less to be conducted to get about 60C from one end to the other. Copper, for example, would require 150W for the same cross section and length.

The diodes each add about 1/2 W at each of their contact points along the bar and so end to end heat flow will dominate.

How do you determine where on the conductivity spectrum you want to be? Somewhere between copper and, say, Delrin, there must be a sweet spot. You could probably find a lower conductivity solution, but at some point the diodes are effectively isolated.