New high power 405nm diode?

[IgorT]

Ok, so i did more testing.

The diode is set to 214mA, where it does up to 227mW, but normally i get 226mW. Yesterday it was still warm from building, so i got 223-224mW.

Anyway, at a temperature, where it does 226mW through a plastic lens, the power through the custom glass lens is 273mW! So my estimate was close.. When the power through the acrylics was 230mW, the raw output was way over 280mW!

273mW through glass sounds like a lot, but it's actually the lowest increase of all 405nm diodes. "Only" 20.3% more than plastics...

I expected this, because of the unusually good aspect ratio with this diode. I actually thought the increase would be <20%... The no-lens output looks incredibly fat - there is much more power in the slow axis, than with all other 405nm diodes. Acrylic lenses don't collect the entire fast axis, but can collect more than the entire slow axis. And since this diode has more power in the slow axis than most, more of it's power manages to get through a low NA lens - the losses in AixiZ acrylics are the lowest of all 405nm's - around 22% (calculated total losses compared to raw output).

The slow axis almost fills up the acrylic lenses effective diameter, while the fast axis is clipped, making the beam look unusually round. With a high NA lens, the beam is still somewhat flat tho.


I just compared the no-lens output of my 6x and this Nichia diode, and while the 6x is quite wide in the slow axis, the Nichia is definitelly fatter. And the no lens output actually looks slightly oval.


It IS a completelly new diode, and definitelly a good one. How good, i will know soon, altho i hope it's not too soon...  ;D The only question left now is, how long it can keep this up..




So here is what i have on this diode so far:
Wavelength: Normal to slightly higher (but some variation is to be expected)
Forward voltage: ~5V (very low, seems to be good indication of a diode's power rating)
Threshold: ~35mA
Slope efficiency: 1.2mW/mA (= after a plastic lens, actual slope efficiency is higher ~1.6mW/mA).
Beam profile: Oval (good aspect ratio)

The diode actually matches the datasheet!




I now put in the medium FL custom lens, for a more familiar looking, prettier beam, and the power is still 259mW. 259mW with a pretty beam! I really hope it lasts! I had my 6x at a similar power for a few weeks, but it sped up the degradation, so i lowered it back down a bit. It's gonna be a tight competition...

I will plot my 6x next, also up to 220mA, see what i get...


P.S. Keep in mind, that the powers i am talking about can not possibly be safe for this or any other diode available to us right now. I am torturing the poor thing here, in order to find out how it compares to tortured 6x's.

 

[suiraM]

[quote author=suiraM link=1228867955/60#79 date=1229629675]It might be the diode used by Olympus in their fluorescent imaging scanners, which could certainly be a Nichia. If someone has a contact in Olympus, it might be worth checking out. Otherwise, there's nothing to do but to wait for a test.

Why would it be that exact diode?  I don't understand.  :-?[/quote]

Serial number. Olympus uses a diode with a similar serial number in their fluorescent imaging scanners, and it's a Nichia, which translates into a rounder, higher quality beam with more output power. Nichia is king when it comes to short wavelengths.

 

[jayrob]

Great information Igor...

Thanks for your work, keep it coming!
Jay

 

[IgorT]

Thanks Jay..


Ok, so i just plotted my old 6x. I just need to put the data in. Altho the Vf data is barelly worth typing. It overlaps almost PRECISELLY!
I could type it in, but the difference would not be visible in the graph anyway.. This will save me some time.


And that's not the only similarity.. Just a moment..


Ok, look at that.. Yeah, i went up to 250mA.. It's an old diode, and i'm getting plenty more in a few days. I was only saving it for this very graph. Now i will set it high again, and start using it till it dies.

So, i went up to 250mA, where the power after a plastic lens was 234mW! That translates into 288mW after glass, or almost 302mW raw output!


And the similarity to the Nichia diode is... Once again, no kink! The Vf is nearly identical, the slope efficiency is slightly lower (it used to be higher, but the diode is old and degraded), but if i only had the graphs to judge from, i would think these two diodes are one and the same model, from the same manufacturer....

But they are not. The 6x is an amazing diode for sure, it has a rounder beam than most 405nm's but the Nichia is slightly rounder than that. That is the only difference. That, and the high intensity "yellow halo" with the Nichia. Unless 6x's vary in beam profile, which i will know soon...



In any case, this is why i was so interested in this Nichia diode.. It has a known power rating. All the others are completelly unknown, and the datashets we try to attach to them are guesses at best.

But the Nichia is rated for 100mW (120mW absolute max. rating). And the 6x behaves pretty much the same way. It has been suggested, that the 6x is the 105mW Sharp diode. This testing can not proove it is, but it shows that it definitelly could be. Because from what i've seen just now, i believe the power ratings of the 6x and the Nichia are nearly identical. So the 6x might just be the 105mW Sharp diode..


My 6x always had a slightly lower efficiency (and it degraded a bit with use), but the lower efficiency doesn't actually mean anything. Diodes always vary in efficiency, datasheets specify a range for it. I expect to find all sorts of efficiencies, once i get more 6x's.



The only question now is... Where is the kink?!? And how do i find it without killing the diode. They are not made to survive 300mW raw output, not even pulsed. I will repeat the range and do some kink-seeking with the visual method, but i doubt i will find anything under 250mA.

 

[IgorT]

If i put the graphs one on top of another, this is what i get....


One interesting thing, which is hard to notice without overlapping the graphs is, that the 6x has a slightly higher threshold.

Again, this does not necessarily mean anything, because the thresholds will also vary among the same diode type, and besides, when 405nm diodes degrade, it is their threshold, that moves up, while the slope efficiency stays the same - if you increase the current by the difference in the threshold, the diode is back at it's original power.


So this higher threshold might very well be the result of degradation over time. It's possible it was the same or even lower, than the Nichia, when it was new... Unfortunatelly i did not plot it then.


But i have also been saving another (badly) degraded diode. A 4x with a broken window, which dropped in power dramatically in a very short time (30 minutes or so). I have a plot of that one, from when it was still fresh, and i can re-plot it now, to finally see what degradation looks like in a graph.


Anyway, here is the 6x vs. Nichia plot...

 

[jayrob]

Nice Igor...

The two diodes look close enough from this first test comparison, that it wouldn't make sense to pay the extra for the Nichia. Especially since your 6X has already been tortured before the comparison.

On this last comparison graph between the 6X and the Nichia, I'm figuring both doides were tested with an acrylic lens?
Jay

 

[suiraM]

Most likely, this LD is a minor variation from the NDV4313, which is rated for 120mW CW. The threshold, slope efficiency, operating current, voltage, oval beam profile and moderately high wavelength all seem to match. The difference is probably the PD, which makes sense, since a scanner would probebly use a temperature controlled system anyway, making the output linear with power up to at least 140mW.

So long as the voltage stays below 5.5V, I suspect you're just shortening the lifetime when you go above its 140mW absolute maximum. On the other hand, I'm not the one who paid money for it, so I'm not going to ask you to blow it up for us. :-X

 

[IgorT]

Well, the graphs alone don't tell the whole story. Both diodes can easily go to 200mW after a plastic lens, but so can many PHRs. The thing i am interested in the most is, which of them can survive that for longer.


As for the price, if Susie were to sell them one by one, the $75 would have been justifiable, since she would have to pack each and ship 100 packages. Not to mention buying 100 herself first, and taking the risk of not selling them all.

But in a GB, if we collect 100 orders, then she is taking no risk, and has no more work to do than she did with shipping this one sample to me. Because of that, i would expect the price to be much lower, than one by one.


As much as it hurts, i will have to do accelerated aging testing with this diode, and then with some of the 6x's i'm getting. THAT is the really important part for me - which one lasts longer. The rounder spot is obviously a nice thing to have, but if it turns out the 6x's are tougher....



Otherwise, yeah, AixiZ acrylic lens in all graphs. The power of the 6x is 23% higher with short FL/high NA custom glass, and the power of the Nichia is 20% higher with the short FL custom glass.

The Nichia is currently at 214mA, where it does 226mW after plastic, 273mW after short FL glass and 260mW after medium FL glass. I am using the medium for a prettier beam. The power is plenty high as it is.



Anyway, i forced myself to abstain from using my 6x in order not to kill it before i would finally have a minute of spare time, to plot it. Now that this is done, i will set it to 235mA, and use it till it's dead. Oh, right, i need to do visual kink seeking while it's still attached to the Diode Analyzer!

Ok, visually, no kink is noticable either. So obviously, the diode does not have to be set above a kink in order to degrade.. When i realized my 6x is degrading, i thought i would find it is set above a kink for sure...

 

[danq]

so... not to be repetitive, or insatiable... but why not push one further? have you ever lost a 6x by overpowering it?

BTW - Igor, you are a very valuable asset around here!

Thanks.

DanQ


p.s:





Go! Go! Go! Go!
Higher!
Higher!
Higher!
Go! Go! Go! Go! Higher! ;D ;D ;D

 

[IgorT]

so... not to be repetitive, or insatiable... but why not push one further? have you ever lost a 6x by overpowering it?

What do you mean? I am losing my one and only 6x, from overpowering it. I already know it's dying. I was saving it for this plot. I now set it to 236mA, where it's doing 227mW through plastics. It's not gonna last much longer. But damn, it's pretty!  

As for the Nichia, i only have one. I don't know if i'll ever have another. Pushing a diode till it pops is useless, cos it tells you nothing. I want to find a power they can survive not where they die instantly, and i want to compare the Nichia to 6x's in terms of "toughness"..


The best way to go about it would be to set both the Nichia and a fresh 6x to 200mW after plastics (or better yet, same raw output), and put them on the cycler circuit (aka the diode terminator). 200mW is an interesting boundary. One that will be hard to cross while keeping a laser reliable. They would degrade at this power and die, but for a toughness comparison, that is actually desirable, to compare the rate of degradation....


But damn.. >220mW looks so much better on the meter..  : Both diodes are currently blasting against their respective lenses with nearly 300mW of raw power. That translates into 227mW after plastics for my 6x, and 259mW after medium FL glass for the Nichia... (The Nichia stole the glass lens from the previous queen).

I'm scared, but addictions are so hard to overcome, and my tolerance is climbing..  That's my excuse for the risky behavior..


300mW of raw power is bizzare, Dan. How much higher would you have me go? Even the future 12x diode is not made to survive that! Not that that will stop us from setting them to 0.5W.. After a plastic lens of course.. ;D


EDIT: Once i get the 6x's that are on the way, i will push a few until i find the kink. It has to be somewhere! If the Nichia diodes become available, i will do the same thing with those. Then run accelerated aging tests on them all of course.

 

[danq]

What do you mean? I am losing my one and only 6x, from overpowering it. I already know it's dying.

I guess I meant, follow that curve (actually a straight line so far) upward until it starts to level off... or it pops.
Mostly, I was having fun - it's Christmas! and I only have Rosie to play with, but I can't play physically so she's just bored, and I only have short times sitting up at the computater so I'm bored too. But you're doing exciting stuff! and I'm only kidding, in pushing you...

Your specific current recommendations for driving these sustainably are much appreciated - 'cause I will have a few 6x's in a day or so and want to be safe with them.
But: I want that safe level to be as high as possible - safety to the max! ;-)

EDIT: Once i get the 6x's that are on the way, i will push a few until i find the kink.

That's it - I want you to get Kinky with it! ;D

DanQ

 

[IgorT]

I guess I meant, follow that curve (actually a straight line so far) upward until it starts to level off... or it pops.

Yeah, well, i fear the other thing would happen. Kink or no kink, the end facets on these diodes are not made to withstand 300mW CW raw putput! Not even pulsed!

It's a real miracle, what we can do to these diodes, and how forgiving they are. At least for a certain time.......

Your specific current recommendations for driving these  sustainably are much appreciated - 'cause I will have a few 6x's in a day or so and want to be safe with them.
But: I want that safe level to be as high as possible - safety to the max! ;-)

Did i make specific recommendations? I forget... Many of the powers i mentioned were DANGEROUSLY HIGH! Some diodes might die in an instant with nearly 300mW raw output.

I know for a fact, that a 6x at 190mW is more, MUCH more reliable, than a PHR at 150mW. I mean, 20 experimental PHRs at 150-170mW died in a row in the same time the 190mW 6x just kept going.

And yet it degraded with time. Question is, will the degradation lead to death, or just gradially dropping power? With PHRs it could go either way, depending on how high they were set. The ones set high seemed to live and then just popped after dropping ~7mW. Others that were set lower, dropped 40mW, but kept working!

So while a 6x at 200mW after a plastic lens will mostly have much better reliability than PHRs at 150mW, my 6x has clearly suffered some damage from this.

That is why my 6x power recommendation would be 160-170mW for a long life. The current depending on efficency of the particular diode, of course.... At 160mW a 6x still won't last forever, but it could be hundreds of hours still. And that is the minimum i am willing to accept. Altho the fact that mine lasted months even in my hands, means that they could even last a year of normal use or more at ~200mW (plastic lens)...

If only they wouldn't cost so much, i would put more of them on the cycler, at different powers, find out the approximate lifetimes. But as it is, i will have to test one or two powers, so i have to make a smart decision as to which, to get the most out of it.

EDIT: Once i get the 6x's that are on the way, i will push a few until i find the kink.

That's it - I want you to get Kinky with it! ;D
DanQ

Oh, i'll get real kinky with them, once i have more than one..

 

[john_lawson]

hey I want one lets get a group buy going :-?

 

[LarryDFW]

Igor;

I appreciate your testing of the Nichia LD.

Nichia lab people say the effective lifetime of the LD is when the current increases ~30% to maintain original max output power.

I have also been looking for some short focal length lenses, preferably with AR for 400nm.

Here is one by LightPath for Nichia 405nm lasers with a focal length of 4 mm and working distance to the diode of 2.39 mm.

lightpath.com/literature/catalog/LightPath_Catalog_Blue_Laser.pdf


LarryDFW

803T Violet pointer
Several greenies
Several HE-NE reds

 

[IgorT]

Igor;
I appreciate your testing of the Nichia LD.

Nichia lab people say the effective lifetime of the LD is when the current increases ~30% to maintain original max output power.

Yes, but that is when powering the diode WITHIN it's ratings. Even then diodes degrade with time. But very slowly. It can take thousands of hours. We don't need them to go on for so long, so we trade some of this time for more power.

But if you go too far, the diode can either degrade really rapidly or even simply just die in hours.


Which way it goes, would seem to depend on how much it is overdriven. At really high powers diodes rarelly get a chance to degrade by 30%. They degrade just 2-3% and then poof.

I have also been looking for some short focal length lenses, preferably with AR for 400nm.

Here is one by LightPath for Nichia 405nm lasers with a focal length of 4 mm and working distance to the diode of 2.39 mm.

lightpath.com/literature/catalog/LightPath_Catalog_Blue_Laser.pdf

I actually have that lens, but it's not very good. It projects all sorts of crap around the spot. It's not as much the lenses fault, as it is the diodes. 405nm diodes seem to have way more die artifacts than red diodes. As if the manufacturing processes are not as advanced as with reds. And the shorter the FL, the further away from the spot this crap is projected, and it gets real bothersome.

Otherwise, the lens you linked to costs $95 a piece as a sample and $60 in quantity of 500.. As do all their other lenses which are useful for us. But i have identical ones from another manufacturer, and they will be available, as soon as i get some more money together. Won't be long, cos i'm finally getting what i needed for that..

 

[IgorT]

hey I want one lets get a group buy going  :-?

We don't even know what they can do yet. I am currently letting it "compete" against my 6x.. Even Susie said she will wait for the end of the lifetime testing, before deciding what to do..