Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

NUBM44-81 (V2) current wavelength distribution

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
Hello Guys!

Does anyone know how the wavelength distribution looks like on these "new" V2 diodes?
I need a relatively low 444 or 442nm wavelength for a special project and like to know my chances of getting one.

Anybody recently spectroed a NUBM44 V2?

Thanks
Farbe2
 





Joined
Aug 31, 2016
Messages
720
Points
93
The "new" V2 diodes came out in 2015...... the 47's in 2016......

At what current or output power do you need to run the diodes? The wavelengths typically climb the harder you push them.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
I need as much power as I can get. Preferably >6W @444nm @ 25°C or even hotter.
I would be willing to go to 20°C but not much below that to reach 444nm.

I wasn't aware of a third version. DTR stated that the "new" V2 might be closer to 445nm.

Do you have data on the 47s?

Or a spectrometer on hand to measure a full block to get informations on the WL distribution?
 
Joined
Aug 31, 2016
Messages
720
Points
93
The V2's and 47's are likely the same diode but one comes in a block of 8 and the other a block of 10. There is also a much newer NUBM4F that I carry but I haven't done much wavelength testing. I have a Thor Labs spectrometer but testing diodes that are super high power has been challenging because it seems to overload the sensor a bit and gives more of a broad spectrum reading. I have a diffuser attached but it doesn't seem to help enough. I only know the basics on how to operate it so maybe another user here can help guide me on the best way to get an accurate reading? I tried getting support from Thor Labs but the person I was in contact with was less than helpful.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
That’s easy!
Just don’t shine the full beam on the spectrometer input Fiber defuser.

I do it with a white piece of brick or a back piece of anodized aluminum.
Use that as your target and adjust your spectrometer pickup in such a way that it catches the reflected light that comes of the target.

Look at the spectrometer pickup like it’s a camera: don’t shine the beam directly in it, but rather let it look at the spot of the beam on the target.

That’s way more than enough light to get a reading.

I would be very happy to buy some diodes if you would measure the wavelength beforehand.
 
Joined
Aug 31, 2016
Messages
720
Points
93
OK, that makes sense. I wasn't sure if reflecting it off another material would cause a color shift. I have plenty of diodes to test so when I have a moment I will see what I can come up with.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
It makes sense to use a normal output power measurement setup.

->
Having a diode cooled and collimated in a heatsink. The diode is power tested with a power meter. The reflection of the beam that hits the power meter gets used to measure the wavelength.
Using a power supply to power the diode makes current adjustment easy.

That way you can see the direct relationship between the output power, wavelength and current.
I do it this way.

I assume that most diodes are able to hit 444nn, it just depends at which current / output power.
 

Why_you

Member
Joined
Jun 1, 2021
Messages
30
Points
18
Look at the spectrometer pickup like it’s a camera
Well, it actually IS a camera, just one with only one row of pixels.

Also, bricks or aluminium are kinda shitty white targets, just defocus the beam and point it at some PTFE (thread seal tape/plumbers tape). For the tape, just wrap a few layers of tape around some object and you get a really good white standard. But you have to defocus the beam so you DO NOT BURN IT cause that releases rather toxic fumes.
 
Joined
Aug 31, 2016
Messages
720
Points
93
Well, it actually IS a camera, just one with only one row of pixels.

Also, bricks or aluminium are kinda shitty white targets, just defocus the beam and point it at some PTFE (thread seal tape/plumbers tape). For the tape, just wrap a few layers of tape around some object and you get a really good white standard. But you have to defocus the beam so you DO NOT BURN IT cause that releases rather toxic fumes.
I usually remove the collimator all together and just shine the un-focused beam at it from a distance. This seems to work fine for lower power diodes but certainly not for the high power ones. I will try some sort of white reflection to point it at.... I might be able to get a small sheet of PTFE to use.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
In my opinion, the target doesn't really matter.
Even if it is "shitty" enough to start glowing incandescent, the broad black body spectrum is easy to distinguish (and way lower intensity) than the reflected light.

Thats why I use a power measurement head. It gives me the added benefit of seeing the output power and giving enough reflection to measure the spectrum at the same time.
Its also rated for the energy so there isn't any toxic fumes, burning plastic or vaporizing material.
 

Why_you

Member
Joined
Jun 1, 2021
Messages
30
Points
18
Or just use a beamsplitter... But if you're using a line CCD based spectrometer it doesn't matter anyway since they are only accurate to a few nm due to their shitty resolution.

If you want to do it right, you either use a white target for a reflection measurement or a beam splitter for a direct measurement with a scanning monochromator setup and simultaneous power measurement with the rest of the beam.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
As the diodes change there wavelength heavily with injection current, measuring the wavelength to <0,1nm accuracy isn't really needed.
Just tweek the current and maybe even temperature to get precisely what you want.

A rough idea (<0,5nm) of wavelength @ power is all that I need.
 

Why_you

Member
Joined
Jun 1, 2021
Messages
30
Points
18
Yeah, but with the usual 1024 sensors you won't get that... You can be happy if it's +/- 2-5 nm.

CCD is good for instant measurements, but the resolution is actrocious. New sensors with 2-3k pixles are better, but still not nearly enough for your <0.5 nm. Especially if it's a wideband 200-1000ish nm sensor. Smallband for 2-300 nm would probably work, but those are specialist items cause that's not a common demand.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
Are you sure?

Almost all manufacturers specify much better than +-5nm.
for example:

Thorlabs also specifes <0,5nm for a 350-700nm range which is exactly what a hobbyist would need.

The LR1 is very old and already uses a 3600px wide sensor. (which isn't even in production anymore)
The newer LR2 version is advertised with a much better resolution.

Might be that the "usual 1024" sensors you talk about are long gone history.
The 3600px Toshiba sensor is older than 26 years by now.
 
Last edited:

Why_you

Member
Joined
Jun 1, 2021
Messages
30
Points
18
Don't confuse resolution with accuracy...

Also, those are new units. I don't think many hobbyists are going to blow 3k for a new Thorlabs. Most of the stuff that's currently on ebay uses 1024 sensors.
Like i said, you can use smallband devices for a better resolution (like the Thorlabs 350 nm you mentioned), but as a general purpose spectrometer those are utter garbage: No 808 and especially no 1064. If you only ever want to look at VIS diodes, sure, but there are much, much cheaper ways to do this.

Aseq doesn't even list the accuracy for their units...

In my opinion, it's just the completely wrong tool for the job. Sure, you can kinda get it to work by throwing money at the problem, but a complete scanning Czerny Turner setup is maybe 150 bucks in parts and five hours of work for 0.1 nm accuracy (since you can easily cal it with a two point cal). Compared to 750-3k for worse specs.
 

farbe2

Well-known member
Joined
Oct 3, 2018
Messages
298
Points
63
Yes you are right, the LR1 does not state accuracy.
I also have no idea what's on eBay currently.
I know that the Ocean Optics USB2000 was quite readily available used. Its <1,5nm accuracy and was quite cheap but I don't remember the real cost. (around 300€ maybe?)

Its also only 3xx to 8xx nm or something like that.

I personally never had a use for detecting/measuring 1064 or 808nm, that might be useful for building dpss but I imagine that most people here never build dpss.
If you only ever want to look at VIS diodes, sure, but there are much, much cheaper ways to do this.
Interesting, I do have a need to do that, do you have any informations on what that would be?

I am also quite interesting in the monochromator setup you described. Do you know any sited providing build logs or part lists or something?

I can't really find a grating that would be able to do 200-1000nm and even small vis gratings are >100€.
Maybe the low efficiency of a VIS grating at the outer edges of the spectrum is still enough.
But if your 150 bucks estimate is remotely correct, I would be very much down to building one!
Maybe we could start a thread and collect information?
 
Last edited:




Top