Exact frequency range for a DPSS 532 nm and a 445 nm lasers.

I'm starting to order to a big company in Italy some new lenses for my five Laserglow beam expanders.
All the lenses of my expanders were scratched and full of imperfections from origin, so I decide to buy new sets built much better of the originals.
I do not understand why a serious company like Laserglow insists to sell very expensive expanders with the lenses ruined at factory.
The last expander that I bought about one week ago show the same defects of all the others expanders.
I asked always to check the lenses before to ship them with the same results ...
Take a look of the horrid surface of the last lenses that I have bought last week:



If You think that this expander costs 349 dollars plus courier plus duty taxes,
it is a totally unacceptable issue!
So now I need some infos about the diode frequency to make an exact coating to apply to the new raw lenses.
I have a NDB7875 9mm 445nm diode @2W, a NDB7A75 3.5W 445nm diode
@5W and a Hercules with a 532nm DPSS diode.

To obtain the best coating performance I need two parameters: the central frequency and the range before and after the central frequency.
So there is anyone that can help me to find those specs?
For example: the 532nm DPSS diode is not overpushed, so the central frequency is really 532. It is enough to make a +-5% gap?
For the 445nm ones I have some doubts: in fact overpushing the diode surely will increase the frequency, but how much? It is important for me to know the central frequency of the two 445nm diodes to make a correct bell centered at that frequency.
Please remind that a closer bell centered to a wrong mid frequency can cause a big loss of power, but also a thicker bell can cause a noticeable loss of power due to the increased reflection area of the lens.

Im not sure....!
but would the wavelength change as the diode warms up (for the 445nm lasers) for example a cold start may be 448nm in the first 15 seconds or something and after 45 seconds maybe 450nm? which would require a bit wider wavelength coating?
Just me guessing really, i have no idea and i dont have a spectro to test either

Minamoto Kobayashi said:

I'm starting to order to a big company in Italy some new lenses for my five Laserglow beam expanders.
All the lenses of my expanders were scratched and full of imperfections from origin, so I decide to buy new sets built much better of the originals.
I do not understand why a serious company like Laserglow insists to sell very expensive expanders with the lenses ruined at factory.
The last expander that I bought about one week ago show the same defects of all the others expanders.
I asked always to check the lenses before to ship them with the same results ...
Take a look of the horrid surface of the last lenses that I have bought last week:

If You think that this expander costs 349 dollars plus courier plus duty taxes,
it is a totally unacceptable issue!
So now I need some infos about the diode frequency to make an exact coating to apply to the new raw lenses.
I have a NDB7875 9mm 445nm diode @2W, a NDB7A75 3.5W 445nm diode
@5W and a Hercules with a 532nm DPSS diode.

To obtain the best coating performance I need two parameters: the central frequency and the range before and after the central frequency.
So there is anyone that can help me to find those specs?
For example: the 532nm DPSS diode is not overpushed, so the central frequency is really 532. It is enough to make a +-5% gap?
For the 445nm ones I have some doubts: in fact overpushing the diode surely will increase the frequency, but how much? It is important for me to know the central frequency of the two 445nm diodes to make a correct bell centered at that frequency.
Please remind that a closer bell centered to a wrong mid frequency can cause a big loss of power, but also a thicker bell can cause a noticeable loss of power due to the increased reflection area of the lens.

Click to expand...

DPSS will be much narrower than 5%, even 1% would be wider than I'd expect for 532nm DPSS - DPSS is not just a diode, you also have an Nd:YAG/YVO4 crystal to produce 1064nm from the 808nm diode and a KTP crystal to double that to 532nm. The 808nm output from the diode could shift by a few nm either way, but that wouldn't change the 532nm output.

For diodes - it's hard to tell, the wavelength will shift with temperature - the datasheet will tell you how much it will vary per unit temperature - but the diodes won't all be the same wavelength at the same temperature, you'd need to test yours with a spectrometer.

I wouldn't expect your coatings to be so narrow that a 10nm shift in either direction would cause an issue though.

Well, since I will use the laser after warmup, it is important to know the center frequency that the diodes emits when it is warm.
For the 532 nm, a +- 2% could be enough, so the bell will start at 521,3 nm and end at 542,6 nm, that is 21nm gap, more than enough in my modest opinion.
For the 445nm I'm really in doubt, because I do not know the central frequency after I overdrived the diode.

Oops I replied when You was writing Mr. Diachi . So Do You think that 1% will be good for the 532nm one?
Due to the high power involved with the 445 nm ones (5W), if I create a too narrow curve there is the risk that too much power could
be stopped by the coating, with a consequent burning of the coating itself. This is why it is so important to know as close as possible the central
diode frequency.

Minamoto Kobayashi said:

Oops I replied when You was writing Mr. Diachi . So Do You think that 1% will be good for the 532nm one?
Due to the high power involved with the 445 nm ones (5W), if I create a too narrow curve there is the risk that too much power could
be stopped by the coating, with a consequent burning of the coating itself. This is why it is so important to know as close as possible the central
diode frequency.

Click to expand...

I think you are over thinking things, I'm not sure what coatings you are planning on using but <2% reflectance for 400-700nm broadband AR Coatings for 0-45 degrees incidence is a perfectly reasonable value. At 5W that is only 0.1W reflected with 4.9W transmitted - Very little of that will be absorbed and the damage threshold on your coatings should be more than able to cope with that.

I am looking at some optics just now (with the above mentioned parameters) that have damage threshold of 3J/cm^2 for a 10ns pulse - There is no way you'd damage a coating like that with 5W CW of 445.

See attached images - 532nm AR on left of image 1, broadband AR image 2.

The performance loss from 532nm specific AR to broadband isn't that huge. 1-2% should be reasonable for 532nm if you want to go that way - I'd go for more like 5% for the 445.

Personally I'd use broadband...

Edit: Just realized how small those images are... Here is a page with a list of AR coating options: http://eksmaoptics.com/optical-components/coatings/anti-reflective-coatings/

Thanks!

Thanks for the hints!
So do You suggest to create a coating with +-1 +-2% gap from the center of the main frequency for both 445 and 532 nm?

Minamoto Kobayashi said:

Thanks for the hints!
So do You suggest to create a coating with +-1 +-2% gap from the center of the main frequency for both 445 and 532 nm?

Click to expand...

If it were me I'd just use a broadband AR coat covering 400 to 600 or 700nm - I'd expect reasonable performance with that, if you're running at 5W then you won't even notice the losses.

If you want to go narrower and get a slight performance increase then I'd go 2% for 532nm and 5% for 445.

Thanks Diachi, now I ask to the company to create for me the coating with the parameters that You have suggested me.

Minamoto Kobayashi said:

I'm starting to order to a big company in Italy some new lenses for my five Laserglow beam expanders.
All the lenses of my expanders were scratched and full of imperfections from origin, so I decide to buy new sets built much better of the originals.
I do not understand why a serious company like Laserglow insists to sell very expensive expanders with the lenses ruined at factory.
The last expander that I bought about one week ago show the same defects of all the others expanders.
I asked always to check the lenses before to ship them with the same results ...
Take a look of the horrid surface of the last lenses that I have bought last week:

If You think that this expander costs 349 dollars plus courier plus duty taxes,
it is a totally unacceptable issue!
So now I need some infos about the diode frequency to make an exact coating to apply to the new raw lenses.
I have a NDB7875 9mm 445nm diode @2W, a NDB7A75 3.5W 445nm diode
@5W and a Hercules with a 532nm DPSS diode.

To obtain the best coating performance I need two parameters: the central frequency and the range before and after the central frequency.
So there is anyone that can help me to find those specs?
For example: the 532nm DPSS diode is not overpushed, so the central frequency is really 532. It is enough to make a +-5% gap?
For the 445nm ones I have some doubts: in fact overpushing the diode surely will increase the frequency, but how much? It is important for me to know the central frequency of the two 445nm diodes to make a correct bell centered at that frequency.
Please remind that a closer bell centered to a wrong mid frequency can cause a big loss of power, but also a thicker bell can cause a noticeable loss of power due to the increased reflection area of the lens.

Click to expand...

Do you know this company? Edmund Optics (Italia) They carry a wide variety of off the shelf ready AR coated lenses http://www.edmundoptics.eu/?site=WE&countryid=108&_ga=1.268125326.1687617163.1428871053

Minamoto Kobayashi said:

I have a NDB7875 9mm 445nm diode @2W, a NDB7A75 3.5W 445nm diode
@5W and a Hercules with a 532nm DPSS diode.

To obtain the best coating performance I need two parameters: the central frequency and the range before and after the central frequency.
So there is anyone that can help me to find those specs?
For example: the 532nm DPSS diode is not overpushed, so the central frequency is really 532. It is enough to make a +-5% gap?
For the 445nm ones I have some doubts: in fact overpushing the diode surely will increase the frequency, but how much? It is important for me to know the central frequency of the two 445nm diodes to make a correct bell centered at that frequency.

Click to expand...

Do you want to order custom made coated lenses of a very small quantity? That might cost a lot.

As suggested look what is available at other companies like EO, Thorlabs, Newport, CVI/Melles Griot.

As to your question:
532 nm +/- 1 nm. Even 0.5 nm.

445 is more tricky.
NDB7875 435-455 from datasheet
NDB7A75. I don't have a datasheet. A similar NDB7K75 has 440 nm (min) 448 nm (typ) 455 nm (max)

Thanks for the hints, I will try!
Unluckly the lenses that I'm looking for has very special features:


Plano Convex Lens:
base curve outer (convex): +9.00 diopters
base curve inner (concave): -1.25 diopters
dioptic power: +10.65
thickness at the center: 4.15 mm
diameter: 30 mm

Convex lens:
base curve outer (convex): +12.35 diopters
base curve inner (concave): -6.5 diopters
dioptic power: +6.75
thickness at the center: 3.3 mm
diameter: 30 mm

532 will always be 532. Your 445's won't dip below 440, and certainly won't go above 460 (based off the 21nm bandwidth you mentioned). Why do you need such exact ranges? I can't imagine you buying a few lenses coated with such a narrow band AR coating- they would be outrageously expensive.

Hi.
The major cost is not the coating (stretch or narrow) ... the big cost is to build two new lenses!
I already tried to decoat and recoat the old lenses, but the result was not so good, because the resultant power loss was much bigger than the original coating.

Minamoto Kobayashi said:

Plano Convex Lens:
base curve outer (convex): +9.00 diopters
base curve inner (concave): -1.25 diopters
dioptic power: +10.65
thickness at the center: 4.15 mm
diameter: 30 mm

Convex lens:
base curve outer (convex): +12.35 diopters
base curve inner (concave): -6.5 diopters
dioptic power: +6.75
thickness at the center: 3.3 mm
diameter: 30 mm

Click to expand...

This is not a common way to describe lenses for lasers. Most often surfaces radii are dimensioned in mm, and focal length in mm (not dioptic power)

The first lens is confusing too
Plano Convex
but then you write about convex and concave. :thinking:

As both lenses are +, you'll build a Keplerian type expander. Any reasons why? The Galilean type is shorter at least.

Hi.
Those parameters were found by a company here in Italy. Perhaps in Italy there is another way to measure parameters.
The lenses were unmounted from Laserglow Beam Expander, and this expander is very short. It uses also another very small input lens.