Cylindrical lens setup in combination with beam expander

[gozert]

Thanks CDBEAM. Always good to get advice from an experienced member. If you're willing to share the project you're working on, that'd be great. I'm always up for new ideas and concepts.

 

[CDBEAM777]

Thanks CDBEAM. Always good to get advice from an experienced member. If you're willing to share the project you're working on, that'd be great. I'm always up for new ideas and concepts.

I am checking now !

There are many things afoot concerning the use of Cylindrical lenses. I have an two (2) offering's for a machined holder for the lenses...See BST section !

What is that you are looking to do ???? I will post in a separate thread all the choices we have....many combinations !! Think about it !!

Single diode or Multiple diodes
Corrected with Cylindrical lenses or Anamorphic prism's
Beam Combining done with Knife edge or PBS cube
Final beam output put thru a Beam Expander or not

SO...again....what do u want to achieve !!! I am unclear on this !! ??

Please advise.

Thanx.

 

[gozert]

I've been thinking about putting it in a Jetlasers PL-E Pro host instead of a Maglite. Likely multiple diodes like the concept I posted on the first page of this thread.

 

[CDBEAM777]

I've been thinking about putting it in a Jetlasers PL-E Pro host instead of a Maglite. Likely multiple diodes like the concept I posted on the first page of this thread.

OK...
Same color diodes...or two(2) different color diodes ??

Beam expander add on....or no beam expander ?

Corrective optics..?? I assume yes...likely
Cylindricals.

 

[gozert]

It would be 2x 520nm >1W each. both corrective optics and a beam expander.

 

[Glutton]

Yup your drawing looks good. As long as you use two sets of the Cylindricals before the cube then things should go fine, but if you combine the beams before cylindrical correction then you have to take special care to make sure each diode is the exact same distance from the pair of Cylindricals in order to get the same single axis expansion ratio with both beams.

This is because these new multimode diodes have such bad divergence that the separation distance between the module and the Cylindricals drastically changes the expansion ratio and hence beam diameter and divergence.

 

[CDBEAM777]

Well....let me start by noting....this build can be done. The closest build...to date is the Jander 6442 Dual 405nm....and my Cyan Cannon ll build ( Going on now).

You do realize that the output from two (2) LD NDG7475 @ 520nm....collimated ,combined, corrected...and shot thru a Beam Expander....in a HH format....will be the most luminous build HH ....done to date !!! That is my guess....mainly based on the 520 nm line of relative brightness !!

Let me give you the outline of the beampaths involved...

First....we start with two separate 520 NDG 7475 diodes in Copper modules...driven with Lasereer drives...set at about 2A each.

Both modules should use the G2 collimation lens.

I would use a PBS to first combine the beams...then shoot the beam thu a single set of 6X Cylindrical lenses.

The beams....now combined need to pass thru the Cylindrical lens set for Aspect Correction ( Go from a bar....to almost a square )

After the single set of Cylindrical lenses....then the beam would go thru the Beam Expander.

I must note that for the PBS to work....one of the beams...B4 being passed thur the PBS....needs to go thru a 1/4 wave plate.
This is just a specialized filter....which rotates the beam polarization....90 degree.

For the PBS to combine the beams....each beam must have its electric field at a 90 relationship to the other beam. In other words....the polarization must be 90 degree to each other.

Let us not worry about the theory....on which I only understand the basics....suffice to say....that one of the beams needs to go thru a small glass filter.....and, that filter rotates the beam polarization 90 degree.

It is really not hard at all.....many on LPF have done this. You may say...hey...." just rotate the diode ".....well...yes....one could....but....That would also rotate the fast/slow axis orientation....and then....one would combine the beams.....but....that would not work....because at the next step...both beams need to have the fast/slow axis....at the same orientation.....for the Cylindrical lenses to work correctly !!

Is not Optics a demanding Master !!!

PM me for more details. I have a design....and an idea on how to use the JetLaser PL PRO as your host. It will involve a custom HSOP unit ( Heat sink optical platform )

This will be a very, very challenging build !!! But....I am positive...it can be done !!:eg::eg::eg:


Note : Glutton is spot on.....the length of the beam path....as the combined beams enter the Cylinderical Telescope set....must be equal !!!!!

 

[Alaskan]

If you rotate both the diode as well as the cylinder lenses 90 degrees, wouldn't that negate the need for wave plate rotation?

 

[CDBEAM777]

If you rotate both the diode as well as the cylinder lenses 90 degrees, wouldn't that negate the need for wave plate rotation?

Warning !!! Long optics explanation....for those that care to expand their minds !!!!! :eg::eg::eg: :tinfoil::tinfoil::tinfoil:

Alaska....YES.......one can rotate the Cylindrical lenses..... Doable but a PITA to align against gravity !!

OK...here we go....
"Down the Rabbit hole " ( Perfect saying ...Look it up)

SOoooo....let's call the diodes ...diode A.....and diode B. They are the same....and the output of each diode , after collimation....has a fast and slow axis of divergence. These fast and slow axis are at 90 degree to each other. These axis... of each diode is rotated.... when we rotate the actual diode can.

I should note that both diode A...and diode B are pointed in the same direction ....for this explanation !! The beam from each diode is at the same elevation....and the beams are parallel with each other.

For the Cylindrical lenses to work....the fast axis of the diode beam...must be parallel with the optical axis of the first Cylindrical lens in the system...the Plano Concave lens.....and the beam must enter the lens right at the center of the lens !

SO....the fast axis of diode A....and diode B ...BOTH must enter the cylindrical lens....with the fast axis parallel to the optical axis of the Plano Concave lens.

SO....let's say....we have the fast axis of diode A going up and down.....12 to 6 on the clock face.....and now.....we rotate diode B....so that the fast axis is going from left to right...9 to 3 on the clock face.

Well....in this condition of diode A....and diode B....the polarization of the diodes is at a 90 degree relationship....great for the PBS to work....BUT....this condition will NOT work for the Cylindrical lens pair !!!

Now....one can rotate the cylindrical lens pair....90 degree....BUT....in the condition above....the cylindrical lens pair will correct ONLY for diode A....and not Diode B !!!

This is where the 1/4 wave plate is an absolute necessity !!!

NOW....let's try this....Both diode A....and diode B....are rotated....so that the fast axis is going up and down...12 to 6 position !! For both diodes !!

The beam from only diode A... goes thru the wave plate filter. This filter....ONLY changes the polarization of the beam....by 90 degree ( 1/4 turn )

The wave plate has NO effect on the orientation of the diodes fast and slow axis...at all !!

SO....we still have the fast axis....going up and down....on both diodes...but now...we have the polarization of diode A....at a 90 degree relationship to that of diode B. PERFECT !!

The beam from diode A.....shoots directly at the PBS cube....and the beam from diode B hits a mirror....takes a right 90 degree turn....and enters into the PBS cube....from the right side!

SO....now....because we have a wave plate in the beam path on one of the diodes beams...the polarization is rotated...( Remember....the fast / slow axis are not affected by the wave plate)....and the PBS will combine the output of the two beams ! !...When the polarization of one beam...is at a 90 degree relationship...to the other beam !!

THEN....the two beams are combined....and have the geometry of one beam.
The two beams now enter the cylindrical lenses....WITH the fast axis ...from each beam....in the correct orientation....fast axis parallel to the optical axis of the Plano Concave lens.

SOooo...now...the cylindrical lenses will correct the divergence of both beams.

Might as well go into this crazy explanation also !!!

The Cylindrical lenses are made so that the slow axis is expanded...with no effect on the fast axis.

Yes....that is correct !! You might think.....whoa....do we not want to reduce/limit that nasty fast axis....optically correct it....mmmm it is too fast....right...it is spreading too fast....the beam is a bar....I I want a dot n(square )

NO....for the best beam dot...what you want is the divergence of the fast axis to be as close to the divergence of the slow axis.

This is what Cylindrical lenses do...they expand/multiple the divergence of the slow axis.....so that the diverengce of each axis is ...as close to being equal as possible.

SOoooo... Cylindricals lenses EXPAND the slow axis....they...multiply it !!!
Remember....we speak of 4X....or 6X Cylindrical lenses !!! When they do this...this gives us a almost square geometry...as opposed to a bar geometry...much, much more of a dot !!

I think all my explanations are correct !! hahaha....I hope I have not used to many !!! marks to devalue my rant on optics !!!!!

If there are any mistakes...please...more educated members...

...in the words Kent from "Real Genius"..." Let me have it " !!!!! :crackup::crackup::crackup:

 

[Alaskan]

Am I understanding you correctly, that you have both beams from the two diodes going into one set of lenses instead of two separate sets?

 

[CDBEAM777]

Am I understanding your correctly, that you have both beams from the two diodes going into one set of lenses instead of two separate sets?

Absolutely...We will only need one (1) set of Cylindrical lenses in this set up !!

The PBS joins the beams....both beams enter the single set of Cylindrical lenses....with the correct axis orientation....and both beams will enter the cylindrical lenses....exactly dead center....

AHHH.....remember....we are using two diodes....each the same !!!!!
The die cavity/ aspect ratio of each diode is the same....SOOOoooo
...we can use one single set of Cylindrical lenses...not two !!

If you are thinking about the Cyan Cannon ll build.....this build will demand two (2) sets of cylindrical lenses....because....there are two (2) different diodes...the Blue and Green....they are different...and will demand a different separation distance between the Plano Concave and Plano Convex...for each color !!

But this build....needs only one (1) set of Cylindrical's !!!

And then...he is going to put a Beam Expander....on the end of all the SchizZZ

Yikes...complicated build !!! His hair will be on fire....at the end of this !!:crackup::crackup:

 

[Alaskan]

OK, sounds better to do it that way, but what if you correct the output of each diode individually or separately before entering the PBS cube, then you don't need a wave plate, but of course, more complex because you need two sets of lenses?

 

[gozert]

That's one hell of an explanation. Sounds doable for sure, although I'd first have to receive the Jetlasers host and then find someone able to create a perfect heatsink for it.

 

[CDBEAM777]

OK, sounds better to do it that way, but what if you correct the output of each diode individually or separately before entering the PBS cube, then you don't need a wave plate, but of course, more complex because you need two sets of lenses?

:crackup::crackup::crackup::crackup: Optics is just a demanding SOB !!!

Yes....one could use two (2) sets of Cylindricals....just like the Cyan Cannon ll Build....MMMmmmmm..not sure what is worse.....for....I have used a wave plate in a Quad Red build in the projector area, They work !!....BUT....
It is not as simple as just slapping a piece of glass filter in the beam path...

Of course it isn't !!! Wave plate filters are always round....typically I have seen
10mm to 25.4mm in diameter...about 2mm thick. They are available with a metal retainer ring around the perimeter....or not.

OK....here is the kicker....the Wave plate MUST be rotated in a specific position...for it to work correctly. I machined a Brass block....with a recess....and used a set screw to clamp the wave plate....when I determined the optimal radial position.....so....just spin'er around....watching your power meter....when ya get to the maximum power....then....the wave plate is in the optimal position !! EZ-Sleezy !!

Now...in the Hand Held format....the same process must occur !! One does Not need a separate holder....but the filter must be in the beam path....and it must be spun to the correct position !!

So...you tell me....what design is easier ??? Dunno ??????

An extra set of cylindricals....or a wave plate ???? Both have good and bad points !!!....MMMmmmm....???? Optics are such a joy !!! :eg::eg::eg:

 

[CDBEAM777]

That's one hell of an explanation. Sounds doable for sure, although I'd first have to receive the Jetlasers host and then find someone able to create a perfect heatsink for it.

Yea...We are optically forging the beams...bang, bang, bang !! haha !!

Things do get a bit complex....but....it is all fun !!!

I can design the heatsink....and machine it to go with the JetlaserPL pro...I think... ??? for I have not seen the Jet Laser unit and do not know the dimensions involved !! This host may be OK...or not have enough room for such a build ???:thinking::thinking::thinking::thinking:

Note: The real challenge is how to align the combined beams...to the center of the Beam Expander...this has my hair aflame !!!

 

[Alaskan]

Thank you for the information, I was researching wave plates this morning though google searches and found the loss can be under one percent, has that been typical for your own projects too?