correcting a NDG7475

[steve001]

Okay, I just want to make sure I'm reading things correctly. Using the 6x cylindrical lens set, I only need those two pieces and a G2 lens? No lens goes after the 6x set? I see the Ebay listing says 35mm between the lenses that come in the set. I assume that is between the front of one and the beck of the other. Since you have a set, can you tell me how thick each piece is? I'd also like to know about how much space is needed between the front of the G2 and back of the first cylindrical lens.

Alaskan, since I can't PM or email you, can you please contact me?

Note: the final beam diameter after correction will determine whether or not a g2 lens is usable. You should use a collimation lens that has a larger diameter than the beam. Placement of the collimation lens is determined by it focal length. To keep divergence low, a long focal length lens should be used.

 

[Dr_Evil]

Note: the final beam diameter after correction will determine whether or not a g2 lens is usable. You should use a collimation lens that has a larger diameter than the beam. Placement of the collimation lens is determined by it focal length. To keep divergence low, a long focal length lens should be used.

You just kind of confused me. The way you wrote that sounds like the collimation lens should be after the correction optics. CDBEAM shows it before the corrections optics. As for the G2, Idon't think ti will be bad considering DTR even offers it as an optional add on.

 

[CDBEAM777]

OK...yes...the Collimation lense corrects the beam...just after the diode...then the Cylindrical lenses are placed after the Collimation lens. I believe the G2 lens works best for using on the NDG7475....but....I have not yet tested the G7 ??? I will test in the next month or so. The G2 works great with 6X.

CDBEAM

 

[CDBEAM777]

OK...I have tested a set of 6X Cylindrical lenses vs. a set of Opt Cylindrical lenses specified for the NDG7475....And...I just added a set of 6X from Lazereer ( Which are no longer available I think )

Interesting !!! The Opt worked better than I rembered...but not quite as good as the 6X.

See here...Tests we made at 1A current to the NDG7475.

6x Cylindricals/Lazereer Cylindricals

@ 0 Meter Beam width = 2.5mm
@ 7 Meter Beam width = 5.8mm

mRad = 0.472


Opt Laser Cylindricals

@ 0 Meter Beam width = 2.5mm
@ 7 Meter Beam width = 7.0mm

mRad = 0.643

SOooo... the 6X are a little better....and the further down beam one travels... from the diode output to the Far Field beam geometry target...the greater the benefit of using a 6X will present with a lesser width ( more a dot ).

I still like the geometry perimeter of the Opt Laser Cylindrical lenses a bit more than the 6X from LSP/AJAI... in that both lenses are the same size. This same size perimeter makes for a more simplified Cylindrical lens mount

So...you choose !! BOTH mRad metrics are very good....but the 6X is superior.

See attached pic of my set ups.

Note: a 7M ( 23')....when using the 6X lenses...the beam presents something very close to a circle (looks like a circle to me). In reality...the geometry is a round cornered rectangle. I tried taking pics...but the spot is just too bright.

CDBEAM

 

[Alaskan]

Thank you very much for the info Robert, that took some work to set up and measure, not just a five minute check of something. I am surprised 6X is better, if expanding the slow axis, I expected it was too much X and would cause the slow axis diameter to be larger than the fast, at distance. Good thing I have four sets of those 6X pairs Glutton had a group buy on, I will use them in time.

You just kind of confused me. The way you wrote that sounds like the collimation lens should be after the correction optics.

I thought I'd chime in here to explain what is happening differently.

The first lens of the cylinder pair expands one side of the beam, the second lens re-collimates it back to infinity focus.

Normally, almost always, the beam must be fully collimated to infinity focus before going through the lens pair. The beam can be out of infinity focus some, but not much. If it is, a small adjustment of the distance between the pair of cylinder lenses can make up for that, but there isn't much room to do so and if too far out, not going to happen.

 

[steve001]

You just kind of confused me. The way you wrote that sounds like the collimation lens should be after the correction optics. CDBEAM shows it before the corrections optics. As for the G2, Idon't think ti will be bad considering DTR even offers it as an optional add on.

Both ways may be doable. Experiment.
I've read another thread where a beam expander is used after beam shaping optics. I think however using a short focal length lens, G lenses are that, before beam shaping optics will limit the final minimum divergence. If shaping the beam is the entire goal then forget what I said.
I found this free download Ebook via Youtube. Laser Beam Shaping Applications (Optical Science and Engineering) Download eBook Laser Beam Shaping Applications (Optical Science and Engineering)

 

[CDBEAM777]

Alaskan,

Glad to do these things. Nothing like experimenting to see what actually is good, better or best.

I just happen to have the Opt Laser lenses....and the 6X were from an old group buy...they are large....but are 6X. The Copper holder is from recent testing I have been doing to get a perfect friction fit between the 12mm module OD and a CU Heat sink. SO...sometimes all things just come together....and a quick experiment pops out...besides...Dr. Evil was " Tapping his toe " hahahaha

I am a real Hack Optical " Gueroo "....I know enough to be dangerous...and the rest...I just experiment and wing it !!

Steve,

Thanx for the link to optic info...hahaha..." As though his ape's brain could understand the secret's of the Krell "
!!!

Later...CDBEAM

 

[RedCowboy]

Dam, I thought the NDG7475 was about like the NDB7875, but apparently it's much worse, is it slightly better or worse than a NDB7A75?

 

[Alaskan]

Forgive me for bringing the dead alive, but I want to open this to discussion again, have no fear of necro-posting if you answer, as I've just un-necro'd it.

I have a lingering question and want to check my thinking with the group. My understanding is a cylinder pair will only affect the divergence of the slow axis alone and not the fast axis, that it's primary purpose is simply to make the slow and fast axis to more closely match their sizes on the output of the laser to become more symmetrical. Due to only the slow axis being expanded more, only that side will have reduced divergence, and the divergence of the fast axis is uneffected and remains the same, correct?

Or, due to a longer focal length after expansion, does the 2nd collimating lens of the pair cause both the slow and the fast axis to be physically larger and thus reduces the divergence for both? When we talk about the divergence of a specific diode when using a G2 lens etc., we are only referring to the fast axis, not the slow, but the slow axis has it's own divergence, we really should be looking at both polarities. Someone asked me what the divergence would be of a NDG7475 laser diode when 6x corrected and then 10x expanded, but since we have different amounts of divergence for each polarity of the diode output, the answer is more complex when using cylinder pairs to correct the beam symmetry.

 

[trinh hong phuoc]

Forgive me for bringing the dead alive, but I want to open this to discussion again, have no fear of necro-posting if you answer, as I've just un-necro'd it.

I have a lingering question and want to check my thinking with the group. My understanding is a cylinder pair will only affect the divergence of the slow axis alone and not the fast axis, that it's primary purpose is simply to make the slow and fast axis to more closely match their sizes on the output of the laser to become more symmetrical. Due to only the slow axis being expanded more, only that side will have reduced divergence, and the divergence of the fast axis is uneffected and remains the same, correct?

Or, due to a longer focal length after expansion, does the 2nd collimating lens of the pair cause both the slow and the fast axis to be physically larger and thus reduces the divergence for both? When we talk about the divergence of a specific diode when using a G2 lens etc., we are only referring to the fast axis, not the slow, but the slow axis has it's own divergence, we really should be looking at both polarities. Someone asked me what the divergence would be of a NDG7475 laser diode when 6x corrected and then 10x expanded, but since we have different amounts of divergence for each polarity of the diode output, the answer is more complex when using cylinder pairs to correct the beam symmetry.

I have thoughts like this, actually they can reduce the divergence of both axes, but it will depend on the diode you put ,If you set the fast axis parallel to the lens plane, you will be able to adjust the axis quickly, and when you set the slow axis in parallel, everything will be similar.
But of course, we all want to reduce fast axis divergence, and I think you have spent a lot of effort for this,that's just my subjective thinking,

Thank you for digging this useful theard

 

[Alaskan]

My Optlasers 6X cylinder pair is about 12 mm wide, that allows the beam to be much wider than the output of a ~6 mm G2 collimation lens, but not 6X wider for the fast axis, maybe 2.5X wider, for the NDG7475T. I'd need to measure though, I don't have this diode here in Qatar to measure. However, since the slow axis is so thin, compared to the fast axis, it can be increased much more, obviously 6X, if a 6X cylinder pair, so it's divergence will be very low after correction. Looks to me, the divergence of the fast axis might be close to 1 mRad, which agrees with a post I saw early on in this thread. So perhaps, 6X correction and then 10X expansion might indeed approach .1 mRad divergence for the fast axis.

 

[trinh hong phuoc]

It's really what I want now, the person who asked me to do this wants a very long distance (about 5-8km) so the beam requires very tight, really sounds very mythical, but I will try with this build ,I have seen you have a 520nm 1w build with a 3.4inch PCX lens, have you ever calculated their divergence?
edit:this is result"s CdBEam
6x Cylindricals/Lazereer Cylindricals

@ 0 Meter Beam width = 2.5mm
@ 7 Meter Beam width = 5.8mm

mRad = 0.472


Opt Laser Cylindricals

@ 0 Meter Beam width = 2.5mm
@ 7 Meter Beam width = 7.0mm

 

[Alaskan]

You don't really need to correct the beam with cylinder pairs to produce a low divergence output (but the beam will be a rectangle), you can just use a large long focal length PCX lens to collimate the output of the laser diode (No G2 or other lens prior to it). With a 3 inch diameter PCX lens you should be able to get a very low divergence this way, maybe less than .2 mRad for the fast axis.

Edit: but you must match the focal length of the PCX lens so the distance required to fill, for example, 80 percent of the lens diameter is correct for how fast the beam spreads from the laser diode. In other words, the diameter of the beam must not overshoot the collimation lens diameter, so it's focal length is a big deal to get correct. You can shine the uncollimated output from a laser diode on a wall and measure how big the spot is on the wall at different distances to determine the approximate focal length needed for a PCX lens.

When your customer wants the laser to travel 5-8 km, what does he mean? He must specify the spot diameter at a specific distance for you to be able to calculate the amount of expansion you will need to meet his requirement. Even the worst divergence low power laser will have photons travel that far, just that they spread out so much you cannot see the beam anymore, at some distance.

 

[CDBEAM777]

HMMM ??? Maybe get down to a 0.1Rad....WITH a 10X Beam Expander....With NO Beam Expander...I would say no !! CDB

 

[Alaskan]

My concern is how wide the output of the beam is from an Optlasers 6X cylinder pair, if a 10X beam expander is used. They are difficult to find with a wide enough input aperture for that much expansion. For low divergence, I think it would be better to forget about the 6X cylinder pair to correct the beam symmetry and just use a large diameter PCX lens to collimate the output of a laser diode and live with the rectangle shaped output.

You could go the long route of using a cylinder pair to correct the output, then using a beam expander in front for a very nice laser pointer having that corrected beam shape, but the cost goes way up, unless you are fortunate to find something which will work cheap enough on ebay.

 

[RedCowboy]

With your diode rotated so that the slow axis is horizontal..............NOTE : The slow axis is the more rapidly diverging as slow refers to the internal geometry IINM and not the rate of divergence.

With the slow axis placed horizontal he beam out of the diode naturally comes out in the shape of a tall thin rectangle as that's the shape of the emitter and it's diverging rapidly, however one axis is diverging much faster than the other, it looks like the letter " l " and the narrow part from left to right ( horizontal ) diverges ( grows ) much faster than the vertical/up/down axis, so in short order the output turns into a wide rectangle that continues to grow in width in relation to it's growing height. That is your spot is an ever lengthening line seen far field.

So you 1st use a G2 lens so you grab the beam as close to the emitter where it's as small as possible and you use the concave cyl lens to expand that horizontal axis while it's narrow but about to grow fast and expand it, then use the convex cyl to reduce the divergence of that axis, it's like optical leverage, so your beam will still look like a ribbon only it won't diverge nearly as fast giving you a more consistent size beam and a smaller spot size.

For instance the nubm44 with a G2 lens focused to infinity makes a long thin bar about 1.75 inches long at 15 feet, but after using a 6X c-lens pair ( following the G2 ) the long thin bar at 15 feet is now only 5/16 of an inch long. Basically 1/6 of what it was and the beam looks like a long ribbon rather than a pizza slice.

The NDG7475 diverges slower than the 44 and would probably use a 2X c-lens pair, I have not corrected it yet but plan to, it's the exact same principal.

So in short your G2 ( round single lens aspheric ) collimates your diodes output reducing the divergence of both axis, however one axis is still diverging faster, then your c-lens pair is used to leverage that more rapidly diverging axis, there by reducing it's divergence to better match the less diverging axis giving you a better beam with a tighter spot, looks like a pasta ribbon instead of a pizza slice.

My concern is how wide the output of the beam is from an Optlasers 6X cylinder pair, if a 10X beam expander is used. They are difficult to find with a wide enough input aperture for that much expansion. For low divergence, I think it would be better to forget about the 6X cylinder pair to correct the beam symmetry and just use a large diameter PCX lens to collimate the output of a laser diode and live with the rectangle shaped output.



You could go the long route of using a cylinder pair to correct the output, then using a beam expander in front for a very nice laser pointer having that corrected beam shape, but the cost goes way up, unless you are fortunate to find something which will work cheap enough on ebay.

Try both and you will see, I have gone both routes with the 7A75 and 44 and c-lens correction followed by a beam expander is by far the best result for both appearance and far field spot size, otherwise your output lens diameter becomes unmanageable, seriously try both and compare results then decide.

For the most conpact device size and best bean/divergence correcting the aggressive axis 1st is the way to go, nothing good is easy....usually.