Sharp 638 nm 185mW diode

[epyn]

I bought 8 pieces of sharp SM diodes to my new home build 4W RGB compact module - incl. beam corrected 1W DI greenie + 465 nm diodes. I like the sealed modules with external beam overlapping adjustments
Any suggestion from the experienced laserists to not kill any of them by overcurrent?
I wanna use all the 8 diodes to get at least 1.5W red beam .
I've already tried the given amperage 230mA for one diode by the datasheet. Nice and bright 5mm beam with G2 lens at 39 feet/12 meters

Who has a long term stability test yet please?
Ferenc

[ZeroPoint]

I haven't used any of these diodes so I can't speak with any certainty but if you bought them from DTR there should be some sort of testing with them. With 8 diodes at 200mW each im not sure you'll get 1.5W total after all of the correction and combining but it could be close, especially if you can get more than 200mW out of them. The key is to monitor the input current along with the output power. After reaching the threshold current there should be a steady correlation between the two up to a point. Once the output power stops increasing linearly with input current then you know you've hit the limit.

For instance, say from 150mA to 200mA your output power increases from 90mW to 140mW (an average of 1mW per mA) and from 200mA to 250mA it increases from 140mW to 200mW (an average of 1.2mW per mA) but then from 250mA to 300mA it only increases from 200mW to 220mW (an average of .4mW per mA) then you know you've already hit it's peak efficiency and for every mA you increase the power above that current it will become less and less efficient until it eventually dies. For maximum power/efficiency you want to increase the current until the output power decreases per mA. When you can find the maximum ratio of input current to output power then you know you reached the maximum output power while still maintaining a long lifespan.

If a particular diode can produce 250mW at 300mA and that same diode can produce 300mW at 500mA then you're better off with the 250mW at 300mA because anything higher will cause the diode to burn out much faster. See if DTR has any testing regarding your diode and stick to the peak efficiency measurement. Just because you may be able to squeeze a few more mW out of it doesn't necessarily mean it will be worth the reduction in lifespan.

Once you've found the sweet spot you will better know how many diodes it will take to reach your 1.5W goal. Depending on how many optics each laser has to pass through this may be more or less than your estimated amount of 8 diodes. Trial and error is the best method, even if it costs you a few diodes in the long run.

Cheers
Luke

[jors]

Great explanation Luke,
Sorry, not long term stability test yet.
However Some data (with G2 lens as well):
I've measured about 187mW @230mA and about 204mW @240mA. This is 1,7mW increase per 1mA in this range level, not bad. I wouldn't recommend go past 240mA however.
You can get better efficiency/stability/more lifespan if you cool with a TEC, but even this way, I wouldn't pass 240mA for 200mW out (probably a bit more if you cool down below 25C)
That said...I'm a bit surprised with your beam profile measurements, mine are worse.
I get a 12x7mm dot at 12meters and 8x5mm dot at 7,8meters. Taking into account NF profile is about 2x2.5mm (G2) this is 0,8x0,3mrad aprox. Also, mine have wings, the most pronounced into horizontal axis.
I've compared results with another PL member and he get almost the same beam specs+wings as well... So..Do you really get 5x5mm dot at 12m? and no wings?
Cheers!

[epyn]

Thank You Luke this is a really lot of help,really.
I'll make this test and do an amp/Power graph.
I have a simple calorimeter DIY LPM not certified but shows numerical data.
Comparing some diodes with Jordan's (DTR) measurements i've got nearly the same results with my LPM.

Others have also this little diode and did tests but got different results than me.
I've said i used G2 lens. No,i'm sorry about confusing anyone .The collimator was 3 element lens. This and underdriving the diode (~150mA) gaves that good result - 5mm beam spot.

[jors]

Oh ok, this makes sense

Thank You Luke this is a really lot of help,really.
I'll make this test and do an amp/Power graph.
I have a simple calorimeter DIY LPM not certified but shows numerical data.
Comparing some diodes with Jordan's (DTR) measurements i've got nearly the same results with my LPM.

Others have also this little diode and did tests but got different results than me.
I've said i used G2 lens. No,i'm sorry about confusing anyone .The collimator was 3 element lens. This and underdriving the diode (~150mA) gaves that good result - 5mm beam spot.

[epyn]

This will be my first diode array build.
Now i'm using oclaro red diode but not satisfied with even with cyl lens correction. 21x10mm on that 12 meters with the G2 lens. (Chinese-Techhood lenses) Twice bigger than blue M462 and 3 times than my 532 nm .

I'm expecting much better results from SM diodes.

[kecked]

I see the faint wings in his picture. After an aperture trim all gone. Why not stick to the mits single modes? They do 200mw all day and have very nice beam. Stack two and two and pbs them with rotator. Do that twice and stack the result as a pound sign. Bet it will be about the same size in the end as the poor 700mw after combining and optics. No free lunch. Only real way around is dpss opsl or ion. OR really aggressive filtering. Good luck it's not easy but many have done it.

[jors]

You can get good results with multimode as well, but you need to fine-tune your optics.

It seems your dot have a bit of "S" shape aberration. This indicates one of your cyl is slightly rotated vs the other one, and/or your beam doesn't enter completely perpendicular to the cyls.
Just place a shim (very thin foil) below Convex CYL side to rotate a bit until getting a "-" spot shape instead a "S" spot shape. Maybe you need also to rotate diode a bit.
With Mitsu P73, I get about 16mm wide Spot at 12meters. This requires to expand the beam with 6x cyls until 5mm wide after cyls. Going with G2, Just place the CV cyl at about 35mm far from diode and focus to infinite with the CX cyl.
This way you can get in between 0,9-1mrad which is pretty good.
Oclaro, on the other side, have a bit less wings than P73 but slightly more divergence (probably due diode window aperture). If you expand Oclaro at about 5mm you will get 1,2mrad. You will need to expand Oclaro up to 6mm (after cyls) to achieve 1mrad.

This will be my first diode array build.
Now i'm using oclaro red diode but not satisfied with even with cyl lens correction. 21x10mm on that 12 meters with the G2 lens. (Chinese-Techhood lenses) Twice bigger than blue M462 and 3 times than my 532 nm .

I'm expecting much better results from SM diodes.

[epyn]

Unfortunately both my ocla diodes (since i have 2 projectors) didn't sit in the 100% proper way in their nuts. I had to put some aluminium foil under the PCX as You've explained until i got the best results. On the test i have a perfect corrected beam " - " but after gluing the one of the lenses moved around and you can see the final result a bit " S " shape. I was using epoxy ,hard to align again. See picture


When i have time i will try the P73 with the same optics. Many others said the mitsu diodes have far better beam than the oclaros.
Couple years ago i have PBS'ed P73s with Dr Lava cyl lens but i don't remember how was the beam divergence-shape-FF diameter. But served an 1W CNI DPSS green + 450 nm well!

[epyn]

I've noticed when the cyl lenses are more far form the diode i get better divergence but the Near Field beam diameter is much larger then my scanner mirrors can handle...
Lets say i put the PCV at 50mm i got 0.5mRad but 6-7 mm at the aperature...
When i put the PCV next to the Diode i got 1mRad or worse but the beam -almost- fits to the galvo mirrors.
I've had to make a deal you know.

Many thanks for your help anyway i'm eating your words

Ferenc