Mitsubishi ML520G71...Red Holy Grail or Flashlight Fail ??

[Stoney3K]

also anyone notice o-like seems to allegedly have the 500mW version of these?

http://www.o-like.com/index.php?main...kslcpfs8pn1ag2

The 500mW variant is an open can, so it may be a lot more vulnerable than its 300mW little brother.

[CDBEAM]

First, Thank you Planters and Solarfire for your complete explaination on divergence. I still have MUCH to learn about optics. When I get things back up...I will measure again....and WILL use a mirror fold to acheive a much improved accuracy. How do I measure 8.25mm ?? Well...Paper and pencil at the target to mark the pattern...and digital calipers to measure the result.....it is an approximation...for sure !! Measurement at aperature is even more of a challenge...but dialing down the current to minimum cuts the glare.

Doesn't the beam pattern rotate 90 degree...like the 445. So...would you measure the smallest axis at the aperature...and the largest axis at beam termination....therefore...1.5mm and 8.25mm. Having no training in optics is big disadvantage !!! Hopefully...I just have to learn what to measure and how to use the Pseudonomen137's mRad Calculator !!. Just proves...A little knowledge can be dangerous !! Thanx again for your help !!!!!!!

CDBEAM

[Solarfire]

Doesn't the beam pattern rotate 90 degree...like the 445. So...would you measure the smallest axis at the aperature...and the largest axis at beam termination....therefore...1.5mm and 8.25mm.
CDBEAM

The beam pattern, yes. From the collimator it goes form this | to this _____ , provided the fast axis of the diode is vertically oriented.

PS
Turning down the power of multi-mode diodes can give a smaller beam at the aperture due to mode hopping which can occur at different power levels. The 445nm diode is a good example, at lower power levels it can even go single mode. That’s why I always measure beam diameters @ approximately half power thru a measuring grid.

[planters]

CDBEAM,
Probably most of us have little or no training in optics. But, if you really enjoy this you will pick up a LOT over a pretty short time.

[Xytrell]

If you’re using prisms for 4$ and you’re getting 1.17mrad then you are also probably using a long fl collimator.

Correct. It was the 3-element aixiz glass lens.

Have you actually done any power measurement to backup your claim of only losing 30%? I got that during testing with a long fl lens (really good lens) and good AR coated prisms and that left me with a divergence of 1.8mrad and > 30% loss. So I seriously doubt that you are getting what you claim.

Sorry for not being clear. There was 30% loss on the one prism, alone. End result was ~400mW at ~900mA.

about the $ values, they were hypothetical (I'm pretty sure I mentioned that). I'm just pointing out there is a nonlinear return on investment.

[mixedgas]

Beam Profiling.

The classical quick way to measure divergence is a knife edge translated across the beam and stopping at the places where the beam decreases to the 13% and 87% power points, aka (1 -(e^-2)) points. The math is then a approximation. A caliper and a razor blade and a cheap power meter will do this. The problem is that math assumes Gaussian beams with uniform distributions.

I use a half inch ccd, a meter long rail to slide the CCD on and a Spiricon software setup. At least one other PLer has a Spiricon as well. The CCD I use has a known spacing between pixels. If you know the CCD model, you can use just about any frame grabber to get a rough idea about the beam.

You can also use a bare B&W CCD and a Lens, or calipers and a lens (lense for UK PLers) to do the measurement.

I have a special evaporated chrome variable attenuator to cut the power down so I dont burn the CCD, but crossed polarizers and using microscope slides as pickoffs also works.

The variations happen when you decide to include the Gaussian math, and if you use FWHM.

Half inch CCd cameras are on ebay for as little as 15$. Frame grabbers for NTSC or PAL are around as well. Keep it below 10 mW on the ccd, less if the beam is focused or the beam waist hits the CCD.

Webcams wont cut it, you cannot look up the true pixel spacing for most of them. Pixel spacing differs in the X and Y axis, so you must take that into consideration.

http://www.uslasercorp.com/envoy/diverge.html is one way.

Yes I know about the freeware beam profiling app for cell phone cams. NO, it does not do a good fit, the CCD pixels are too small and the code does not do divergence. Contrary to belief, phones use Bayer color arrays and the math intensity fits can really mess up measurements.

Some place I may have a list of a few common CCDs with dimensions.

Steve

[kecked]

steve just got one of these and it real hard for me to beleive they could ever be made useable. It's a line generator. Bright yes, usable no.

[CDBEAM]

Wow Steve....I suppose If we are to do any real science....we must understand...and possibly use the tools of real science. I guess it depends to what level of accuracy we need to drill down to ??? After giving this some thought...it is a logical progression that just a power meter may not be only measurement tool that is necessary. I would surly be nice to have a method to do some form of accurate beam measurement. Again....need to learn more about what is available...at a reasonable cost !! Oh...thought you would enjoy the attached pic !! haha Thanx !

CDBEAM

[kecked]

has anybody tried to model this in software? there is optic software out there that can help out quite a bit. I have my physics department looking at diode as a favor to me. So far not much. They did have success with a microlens arrangement but "we" can make such a setup and it involved opening the can. This still left the bad side wings. Once you do all that and cut off the sides you are not much better off than using 642nm 170mw diodes slightly over driven.

[LaNeK779]

has anybody tried to model this in software? there is optic software out there that can help out quite a bit. I have my physics department looking at diode as a favor to me. So far not much. They did have success with a microlens arrangement but "we" can make such a setup and it involved opening the can. This still left the bad side wings. Once you do all that and cut off the sides you are not much better off than using 642nm 170mw diodes slightly over driven.

the 500mw is open can, so one less problem there.
looking forward to your findings