cylindrical lens pair vs anamorphic prisms?

[catalanjo]

@ " I actually made this hit record lol " ..........interesting ........ which record are you referring to here ?
Cheers

[planters]

Only if you are trying to cool the very high power blue diodes will it be worth it to use water cooling. Here, the many 10's of watts of heat production can raise the base plate temperature enough that the water cooling will help a little. Otherwise, where water cooling really provides a significant benefit is when you introduce hundreds of watts of heating with the hot side of a TEC. Otherwise,the base plate can become so warm that the net cooling of the TEC can be much smaller than the differential you are producing.

[CDBEAM]

Well....ALL I have seen is that " Two-headed"

POS for the 1.2W red...useless !!.....nothing more for now.... Beam out !!

Not necessarily .. see http://www.photonlexicon.com/forums/...D-638nm-1200mW

Maybe a RED herring, (pun intended), with double emitter width as Phillip suspects, ...but until someone tests it properly to demonstrate this detail, since Hitachi often makes good stuff, there is still a glimmer of hope.

Cheers

[planters]

King...,

If you use water cooling then you do not need any heat sinks. They will actually interfere with the process. All TEC's are very inefficient. On the other hand, they are solid state, convenient and relatively inexpensive. Their ratings are for heat flow vs current at a given temperature differential. Their differential drops quickly as you ask them to transfer more heat from the cold to the hot side.

The way you can achieve a large temperature differential is by limiting heat flow. In an open room environment, the heat that flows into the cold side quickly becoms dominated by heat flow from the room air rather than the operating diode. And, if the temperature drops sufficiently to allow condensation of room humidity on the cold surfaces then you are completely screwed as this process releases tremendous amounts of heat.

To get a large NET temperature drop you need a large differential as I just described, but you then need to ensure that the hot side rises as little as possible above room temperature. Water cooling shines here. If you take your larger TEC and pump a couple of hundred watts through it, don't just measure the temperature drop, but also measure the temperature rise of the hot side as well. You might be surprised at how hot this gets. Take this measurement at the hot surface of the TEC itself and not a couple cm away on the support plate.

Water coolers can remove hundreds of watts from CPUs with a surprisingly small temperature rise.

I can actually get to below -50C at the diode within an air tight enclosure by stacking two TECs on top of each other with the bottom and much larger device powered with a couple of hundred watts and the upper one powered with around 50 watts. The bottom dumps its heat into a water cooling block. As soon as the diode is turned on the temperature rises to -40C from just this small additional heat load.

[planters]

You don't have to guess. Go to this website and look up their standard TEC devices.
http://www.customthermoelectric.com/


Each device will have several graphs in their data sheets. You start with a TEC similar to the one you have and determine how much total heat it will produce including your load and its operation. You might double the load for the additional heat that will come from the environment. Then, choose one of the much larger TECs and determine how much power you will need to pump the combined load from the first system to an acceptable temperature differential.

Finding the best mix can take some time. "do I achieve most of the differential with the first stage or is it better to balance the temp drop across the stages" An optimum mix can make a big difference.

You can even buy pre stacked modules up to 6 stages where the calculations have been done for you and the interfaces are pre built. This company sells up to two stage modules.

[catalanjo]

@ " .I reckon much less than 100mw loss " .......is this part of .......... "the rest is guesswork lol "


Cheers

[planters]

King,

This may be obvious, but I just want to assure that you are using two prisms. This is because you show an image of "the prism I am using". Also, typically the type of prisms that are used are not right angle prisms. They are called anamorphic prisms and have a significantly shallower angle. Although a right angle prism will work, the shallower angle does two things. The adjustment sensitivity is less demanding and more importantly the prism pair can be coated to have much less reflectance loss at the wavelength they are intended to be operating with. That prism looks uncoated. At best, the losses you will have when passing through four, air/glass interfaces, perpendicular to the glass surface is 20% and this loss increases as you incline the intercept in order to achieve the beam expansion you need.

The lens pair that I have been using with my reds is:

http://www.edmundoptics.com/optics/o...-lenses/69798/

http://www.edmundoptics.com/optics/o...-lenses/69729/

This will fit the beams easily. The second lens acts to recollimate the beam after the first lens expands it and is round rather than rectangular to allow very fine adjustment in the alignment of the two lens so that theri axis of expansion exactly corresponds. I show how this is done in my spatial filtering video.

You could substitute a 75 mm lens for the second lens if you wanted three rather than four power expansion. The lenses would be positioned a little closer together if you choose this option.

You might find that you are having slow or little help on this forum. A couple of years ago, when most of the participants on this forum were building their own projectors there was a lot of activity and discussion about how to deal with these issues. Now, a lot of members are buying their projectors and this activity has slowed down but, not stopped.

[planters]

Aim for "perfect" and have the wisdom to accept "good enough".

I get what you saying about the anamorphic prism pair....because they have a shallower angle to them..more of the beam goes through...is this right?

No. I am assuming all of the beam goes through in both cases. The advantages of the anamorphic pair are, as I have said above, adjustment ease and lower reflection losses.

So what you are saying is I could get tighter beam specs and more round with the cylindrical and prism pairs?

Yes, regarding the tighter beam specs The roundness depends on how much you choose to expand the beam. This has been discussed before, but bears repeating. The prism(s) have no focal power. They neither converge nor diverge a beam. The beam exiting your collimator lens has some astigmatic properties. This lens cannot fully correct the divergence of the laser diode which is highly asymmetric. It is either a little too close for one axis or too far for the other. The cylinder pair allows you to correct for this compromise along with the beam expansion that lowers divergence. This allows you to produce an even better beam when you use the lens pair.

[catalanjo]

Yes indeed LOL
If I put the prism on a 300mW red the losses are very high in comparison to the beam going through the prism....with my 2 diodes the losses are minimal(maybe up to 150mw) but I get a nice powerful red beam from it...it seems to work better at higher power :P

Does the "300mw red" have polarised output?
If it doesn't, this might explain the massive difference in comparative losses.

It occurs to me, that your rather "unconventional" use of a right angle prism, might (by total fluke) have arranged for the incident angle, to be close to Brewsters angle (for red) on the way into the prism, thereby neatly obviating the need for an AR coating on the incident face of the prism.

If you rotate this prism so the beam exits in an upward direction........do the losses (reflected beam power) increase drastically.

Cheers
PS. Obviously this experiment won't be at all useful for conforming your beam, it would just confirm a vague suspicion I have, with regard to your rather odd (albeit subjective,) results, as prisms IMHO, most certainly do NOT work better, at higher powers !

[catalanjo]

Ok I need to ask you a question before giving an answer....because my lasers are combined by a pbs cube...does that not make it polarized?
I wish i had a power meter to confirm the exact results

No, exactly the opposite, it means that the beam is NOT polarized ... or more exactly it contains both orthogonal polarizations.

@ "and also does it make a difference if one of the beams goes through a waveplate...is that not polarization? "
This means a rotation of polarization, so that despite conserving the beam shape, the polarization of this beam is rotated allowing the mix later on in the PBS cube.

Oh well .... another great theory up the spout !

Cheers