cylindrical lens pair vs anamorphic prisms?

[planters]

Here's a question I think very relevant to the topic title: isn't one fundamental difference between cylindrical lenses and prism pairs for doing knife edging the fact that with prisms you can stack the "tall" beams horizontally and have them expand equally while with cylindricals since the beam has to be in the exact center of the lens you have to expand these tall beams stacked vertically? With the latter I don't see how you could knife edge any of the multimode 500mW+ diodes without hitting the scanner mirror size limit with more than two diodes.

You are correct that the cylinder needs to be limited to expanding a single, on axis stripe. However, assuming these beams have already been PBS combined and the expansion will widen the stripes then you can very effectively expand the 2 vertical stripes (4 beams) into a square that will fit a 6 mm aperture scanner. The problem with the prism is that the horizontal location of the beams COMBINED with horizontal expansion will quickly produce a wide rectangular beam at even modest expansion and that beam will be only 3mm high.

I got another question for you Planters....how do you get your TEC pelters to drop below zero degrees?

Firstly, never use the TEC's to cool the heat sink. This should be limited to high heat capacity extraction with fine fins and a powerful fan or better, water cooling like the CPU water cooling modules marketed by computer companies. The TEC's are so inefficient that the heat they add to the small heat from the diode is what limits is your ability to remove all this combined heat. This heat extraction needs to occur with as little warming as possible to the side of the TEC where the heat is being extracted. This surface can become hot and the differential may be large, but the diode sees little net drop from ambient.

Condensation of humidity from the surrounding air, on the cold surfaces, produces a lot of energy and this needs to be prevented in order to reach lower temperatures. Cover the TEC stack with an air tight box.

[catalanjo]

@ "I don't see how you could go higher." ..only two ways ....bigger mirrors (taller beam) OR compress the result (more optics & losses) and accept a bigger divergence.

From the way you raise the question, it is quite clear that you now fully understand the stacking concepts and limitations!

Cheers

[Maltes]

I hope I do.
But I do not get one thing, from Kvant to the cheaper chinese modules the upper power limit for pure diode green modules is 10 Watts and for 445nm blue diodes it is 30 Watts. I don't notice any crazy cooling technology that would fit in those modules and one of the chinese module producers (wouldn't want to tell which one) explained that their 10W greens and 30W blues don't have overdriven diodes inside them. So knowing the diodes that are available today, each green diode in the geen modules would give about 1.2W power and each blue would give about 3.5W. To keep things simple and not counting optical losses you'd need about 8 green diodes and 8 blue diodes to achieve these power levels. Knowing they are polarized that leaves 4 beams that need to be corrected and fit into a 6-6.5mm aperture scanners, as the spec sheets of these modules mention the beam size to be.
I just don't see how this would be doable with cylidrnicals (too tall, 12mm, won't even fit stock 10mm Scannermax). With prisms, maybe? I don't know how wide it would need to get for the claimed <1 mRad divergence.

[catalanjo]

Maltes.
Are you sure these are multimode diodes not a shitload of single mode ones, .... I recently purchased a used, 30x single modeLD red stack, giving almost 5w in a 4.5x4.5mm beam, so as to be able to compare it to my own quad multimode builds,..... and it has very low divergence indeed, but was a tad pricey.

@ "(wouldn't want to tell which one)" .... why not ?

@ <1 mRad divergence. = careful with 1/2 angle divergence figures.

Until I get my mits on one, I can't think of a good answer to this question !

Maybe as Churchill once said ...... "Terminological Inexactitudes".

Cheers

[planters]

I do not like Kvant, primarily because of one very misleading spec. Their divergences have historically ( in conflict with industry standards) been listed as 1/2 angle. That means their real divergence is twice as high as they list. With twice the divergence you can pack four times as much power on to a given scanner.

Do not trust the Chinese. This is a bit of a cultural thing. They themselves do not like this characteristic and admit they prefer working with Westerners, but they cheat.

Prisms are not superior to cylindrical lenses in their ability to improve divergence at the price of increased beam size. They are easier to work with, but they have higher losses and do not produce a superior result. There is a reason for both of these properties.

The loses are in part do to the high angle of incidence that they operate at and they need to pass both S and P polarizations. This is very difficult to coat for.

They do not work as well because they have no optical power. Part of the problem with diode lasers, as was mentioned above, is that they are astigmatic and the radically symmetrical collimnating lens is at best a compromise. By adding power in one axis, the lens pair can compensate for this asymmetry. Remember, the lens pair does not have to exactly cancel, it can introduce a focusing or de-focuing effect that you can adjust by optimizing the far field spot.

[Maltes]

Thank you

Remember, the lens pair does not have to exactly cancel, it can introduce a focusing or de-focuing effect that you can adjust by optimizing the far field spot.

I don't understand this part.

Maltes.
Are you sure these are multimode diodes not a shitload of single mode ones

I think for reds you are right, they use 32 or more singlemode diodes.
But I think for blues and greens they use multimode. I wish I had some pictures of the insides of those bad boys saved somewhere to double check.
If I recall these up to 30W pure diode modules appeared around the time the higher power multimode greens and blues appeared, that might also hint to this.

Thinking about this some more they might be using a cylinder or prism pair on each consecutive two vertically stacked beams in the knife edge system. For 4 diodes that would mean two more cylindrical pairs (one more for each polarization) which is not much for a $1000+ build. THen these two (four) corrected beams could be stacked (knife edged) horizontally.

@ "(wouldn't want to tell which one)" .... why not ?

I think that would be an anti advertisement for them. I'm sure all the rest have nothing fancy besides one or two TECs but if they dont have that info publicly lited on their website I'd rather not share info from a private conversation.

[catalanjo]

@ "I don't understand this part." ....only because you are looking for some kind of hidden abstruse meaning behind it,.... it's really just another nail in the prism coffin since prisms can't do this!

@" I'd rather not share info from a private conversation." ....Oh! .... I thought this was a "spec sheet".

@ "they might be using a cylinder (or prism pair = ??) on each consecutive two vertically stacked beams in the knife edge system." ...this is very clear confirmation that you DO understand the Cyl-lens setups. If you push this idea a little further forward, you will possibly want to add a few more cylinders (if you can find them cheap), and explore the possibilities of slit spatial filtering ....... if you add in a spot of individual diode cooling, you will by then, be able to guess what I am mucking about with lately!

@ "Do not trust the Chinese. This is a bit of a cultural thing." ....I disagree here very strongly ... BUT... everyone is entitled to their own opinion, however much of an absurd over generalisation that may seem to me.

Cheers

[Maltes]

Oh! .... I thought this was a "spec sheet".

I wish they'd include info like these in the specs sheets as well so I wouldn't waste time with emails. They only mention things like beam diameter in the spec sheets.

If you push this idea a little further forward, you will possibly want to add a few more cylinders (if you can find them cheap), and explore the possibilities of slit spatial filtering

you mean like planters does it? I don't think he uses few cylindricals for that.

[catalanjo]

@ king4quarter

No ... you will be better off using peltier cooling to run at 1 amp. (unless you live in a very cold climate).

@ Maltes
"you mean like planters does it? I don't think he uses few cylindricals for that. " ...no a bit similar but NOT like planters does it, so far, he uses spherical lenses and "square" spatial filters (almost classical style), but I do often copy his magnets and blades idea which is brilliant!

Cheers

[catalanjo]

I don't go below about 20C (to avoid condensation) if not in a sealed unit.
There are cheap 12vdc thermostat units (with LED display and sensor) on fleabay which I use to cut mod signal if temp goes too high and D4Chill (or just NTC with a FET) to stop temp going too low.

Cheers
PS. Seller is called "Heavends"