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

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

New 9mm 445 diodes

This is the famous SORAA 4W 450nm? Are these the ones that were mentioned in the white paper a few months ago? Getting very very excited.
A couple of my ex-students up at LG, whom some of which were involved with C@s!o and SOR@@for the DLP projection made mention that lasers for projectors would get a major power boost
in the coming months.
S*R@@ will source diodes for future LG and Samsung DLP projectors. This was the overall census.
 
Last edited:





No it's likely this a 1.6 watt nichia diode. From the way it seems these should be able to do well over 2 watts. We have no idea if they can hit 3 or not. I'm sure the 4 watt dioides are coming one day too.
 
I would be happy to take macro shots of a decanned diode and return the diode to the owner. :)
 
We can only assume that the bond wires can only hold as much current as previous bond wires have been able to hold - my guess is that they wouldn't make them thinner. So if we figure that the bond wires break in a 4-wire situation at 2A, then if there are 6 wires here, then we can probably assume that, at 3A, this is SURE to break.
 
Wolfman29 said:
We can only assume that the bond wires can only hold as much current as previous bond wires have been able to hold - my guess is that they wouldn't make them thinner. So if we figure that the bond wires break in a 4-wire situation at 2A, then if there are 6 wires here, then we can probably assume that, at 3A, this is SURE to break.
Click to expand...

indeed, i am thinking of it as a kind of loose guide. it cannot be very accurate.

would need approx length, width & material(if possible).

Then we know each wire can deliver approx x miliamps.

Seam to remember these were gold, and likely most are similar length.
 
There is no need to calculate based off of the width, etc. All we would need is to know what current these fail at in other diodes. I am sure the bond wire specs are all the same.
 
Of course but nobody has done it.

I would rather know when a diode has say 6 bond wires what the max is going to be instead to kill a diode to find out, as far as I know its the bond wires that fail first.






Wolfman29 said:
There is no need to calculate based off of the width, etc. All we would need is to know what current these fail at in other diodes. I am sure the bond wire specs are all the same.
Click to expand...
 
Actually, I believe it's usually COD (catastrophic optical damage) that results in failure - i.e. an obliteration of the mirrored facets.
 
Wolfman29 said:
We can only assume that the bond wires can only hold as much current as previous bond wires have been able to hold - my guess is that they wouldn't make them thinner. So if we figure that the bond wires break in a 4-wire situation at 2A, then if there are 6 wires here, then we can probably assume that, at 3A, this is SURE to break.
Click to expand...


I also have looked at the aspects of number of bonding wires.. The only problem with
that in this situation, is you may have a diode from a completely different manufacture.
With this, not only can the gauge be different, but they also may have more, and still
break under the same conditions as the current 5.6mm package.

I have seen 635nm cmounts that have 16 or more bond wires, but are very thin.
The amount of bond wires also do not designate a higher current diode, rather they
are done this way for other reasons.

What would be safe to say, would be finding additional bond wires on a diode from
the same manufacture. Not knowing the true origin of either keeps you guessing.

Only one way to find out :D
 
Wolfman29 said:
Actually, I believe it's usually COD (catastrophic optical damage) that results in failure - i.e. an obliteration of the mirrored facets.
Click to expand...

From My experience, older smaller diodes fail exactly as you put but the newer higher power ones seam to be failing more on bond wires. Its just my experience but am not expert. I mounted only 600 diodes last year.
 
Well i have decanned my diodes the hard way and picked up the tool from thorlabs to do it the right way.
Now the older smaller laser diodes seemed to die from the emitter being fried, just due to the bond wires being able to carry so much more current.
But i have gone and open up other of my 445nm dead diodes and so far most of them have a broken set of bond wires.
Heck i was able to tap one of the wires into place and it was working again.
So the bond wire gives us a 100% max current then we have to look at the emitter.
 
Exaclty.

kiyoukan said:
Well i have decanned my diodes the hard way and picked up the tool from thorlabs to do it the right way.
Now the older smaller laser diodes seemed to die from the emitter being fried, just due to the bond wires being able to carry so much more current.
But i have gone and open up other of my 445nm dead diodes and so far most of them have a broken set of bond wires.
Heck i was able to tap one of the wires into place and it was working again.
So the bond wire gives us a 100% max current then we have to look at the emitter.
Click to expand...
 
Now i say 100% max current yet even still you should never run at that 100% max
as the breaking of the bond wires depends on a few things it can vary for some.
But it gives you a ceiling current so to say.
Then based on output and temp generated via altering currents its very easy to make a graph and see when the power starts to become less and less.
It will still take us a lot of dead diodes to get a "long life" max current.
The current im going to test at will be above 2A.
Start from there and step up in 100ma and make a graph once i get to the point where the graph starts to even out i stop as we are hitting the max and are just going to make more heat and shorten life.
 
I think I know were these are from,
So are these from a projector with only 2000 lumens?

How many of these diodes per projector?
 
Last edited:
You must log in or register to reply here.


Back
Top