Page 2 of 12 FirstFirst 123456 ... LastLast
Results 11 to 20 of 114

Thread: ML501P73-2, a capless 638nm 5.6mm 500mW Red Laser Diode - Test results.

  1. #11
    Join Date
    Dec 2009
    Location
    MI, flint, farmington hills
    Posts
    569

    Default

    nice photos, cant wait to see the testing.

  2. #12
    Join Date
    Feb 2011
    Location
    St. Louis
    Posts
    526

    Default

    Great work with the photos. Now lets see some power tests.

  3. #13
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    Kiyoukan,

    Those photos are extremely significant. Finally, a meaningful difference between the 73's and the 71's. Just like the A and M series 445's, the bottleneck appears to be the bond wires and the manufacturer has realized this and in each case just added bond wires and... more power. Even the SLIGHTLY greater divergence of the 73 makes sense. The 73 is a 71 driven to increased current levels and with these multi-mode diodes this increases the divergence.

    Now I know these are small dimensions, but does anyone think it would be possible to add ANOTHER bond wire? Hmmm

  4. #14
    Join Date
    Jan 2007
    Location
    Florida
    Posts
    4,382

    Default

    What is the weak link? Bond wires? Heat? COD? Has anybody tested these to death yet?

  5. #15
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    The problem with testing them to death is that unless you analyses the process of dying you will probably miss the cause and you won't learn how hard you can drive them for a reasonable life. I may be wrong, but there seems to be a shift that began with these micrographs and has been taken up by lasertack with his laser profile images and by danielbriggs with his prism images. We are looking into the physical cause of the component performance. Previously, it was primarily anecdotal, more hit or miss. These double bond wires are a revelation.

    If we did preform a destructive test it should be done in a microscope with a video ( high speed if possible) record. It would be interesting to see what would happen to the diode if one of the bond wires were cut or if a jumper could be added. The emission wavelength varies at 0.2nm/C this is a good way to monitor the temperature as the current is raised. An image of the output facets before and after failure might show if this was damaged. The LOC's would go LED after failure and so there had to remain some conduction ( the bond wires could not have been blown entirely off). Do these higher power diodes go LED?

  6. #16
    Join Date
    Dec 2009
    Location
    MI, flint, farmington hills
    Posts
    569

    Default

    This image is not mine but one from LPF hope pullbangdead does not mind
    Click image for larger version. 

Name:	10x_topdown_2_hires.jpg 
Views:	168 
Size:	90.4 KB 
ID:	31881
    So you can see what happens to a over driven 445nm diode without proper cooling.
    Our biggest problem is temp, if it gets to hot these diodes will just "burn" apart.
    With 2 pair of bond wires we can pump alot of current threw them.
    It seems with my 445nm diodes they never break from broken bond wires but from just burning out the contact point like the above image.

  7. #17
    Join Date
    Feb 2011
    Location
    New Hampshire
    Posts
    3,513

    Default

    This makes sense. The solder point will have more resistance and will be the weakest link in the conduction path, otherwise the solder would never have melted to allow assembly. More conductors, lower resistance and therefore less heat. The copper's resistance only climbs as the temperature rises leading to thermal runaway. But the diode must now be dead. Right? It can't LED now. Right?

    The problem with cooling is the component that burns out is the thermally most distant component. Unless we try to run these under flowing deionized water the only solution is lower resistance.

  8. #18
    Join Date
    Dec 2009
    Location
    MI, flint, farmington hills
    Posts
    569

    Default

    Well even a diode like that can run just barely.
    Yes mineral water or some submerged cooling might help but there are also other gases that could be filled around the diode and lens to help increase the ability for heat to spread.
    So where i think our limit is thermal resistance or ability to cool that top contact point.
    As long as we can keep all items below 25C there is going to be alot less degradation or damage.
    That is why i want to cool the diode down to 15C in hope that the emitter contact point wont vary above 25C, but the only way to test this is going to be a thermal imaging camera.
    So testing this diode or i think any diode one also needs to be able to keep the temperature records of the case and the emitter. so if lets say the case is 15C and the emitter is 30C at a given current you know that you are going to damage the diode. so you turn the current down till you get about to 24C on the emitter this would be max temp, with a 1C variation.

  9. #19
    Join Date
    May 2007
    Location
    Romania
    Posts
    1,041

    Default

    Seam to be going to extreams here fellas, D.I water and gas, comon. we just need to run them at a reasonable power. Not cook them to death.

    First test I did yesterday I actually ran the diode with no cooling at all, it flaked out. So I quickly cut the power.

    Now after it cooled, it was working again in the typical standard these diodes are becoming famous for.


    Today I expect to do a stipe compare test. Just loaded with other stuff to do.

  10. #20
    Join Date
    Dec 2009
    Location
    MI, flint, farmington hills
    Posts
    569

    Default

    i would agree that gas filled or liquid filled might be a bit much, but the temp measurement on the bond wire contact plate on the emitter could get us some great test results as far as how much power and cooling is needed.
    I would be happy with this diode doing 1W, if it can do that then its perfect.
    If it can go higher fine but i just hope 1W is fine, for decent life.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •