Looking for new applications fields for novel technology - Direct Laser Soldering (developed by CERN

Discussion in 'General Electronics Chat' started by DLS Soldering, Mar 24, 2015.

  1. DLS Soldering

    Thread Starter New Member

    Mar 24, 2015
    Hi guys!

    We're a project team from the institute of Entrepreneurship & Innovation at the Vienna University of Economics and Business and we're looking for your help in finding new application fields for a novel technology called direct laser soldering (DLS). The technology has been developed and is currently being used at the European Organization for Nuclear Research (CERN) for silicon pixel detectors (Large Hadron Collider).

    some of the advantages:

    + minimize insensitive areas / minimal interconnection areas necessary
    + assemblies which cannot tolerate large temperature variations
    + assemblies with complex shapes
    + silicon chips can be directly mounted to flex PCB without further packaging
    + high quality mechanical and electrical connection (no glue required)
    + chips in mosaic patterns can easily be replaced
    + low cost

    Our goal is to evaluate further fields of application for this technology in order to bring a high value to a broad range of people and make this existing technology applicable for other users as well.
    Do you guys have any ideas where this technology could be used as well? (which Industries? which problems could be solved (or not)? how do we get to potential users? who might be interested as well?)

    Thanks for your time!
    Your information is an important contribution to the success of our project! :)

    best regards
  2. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    Is this system more than just a pinpoint source of heat? How is the solder controlled. Is the solder in any way unique?

  3. alfacliff

    Well-Known Member

    Dec 13, 2013
    how do you get around the problem of reflective metals? solder and tin are pretty reflective, and would not absorb heat well from infra red. I know metal can be worked with lasers, we do it here at work with aluminum, but what power levels are you using, and how long a focus?
  4. BR-549

    Well-Known Member

    Sep 22, 2013
    I've read your post several times and trying to understand your description.

    Are you saying that you are bypassing the IC pins and black plastic packaging, and soldering directly onto the circuit die?

    What exactly are the components that you are soldering?

    What are the dimensions of the components?
  5. DLS Soldering

    Thread Starter New Member

    Mar 24, 2015
    Hi all!

    Thanks for all the replies so far.

    @jpanhalt No, it's only a short impulse of heat to go easy on the chip (heat generation only occurs locally and hence doesn't damage the chip in anyway). SHOULD a chip break you can also easily detach and replace it. "How is the solder controlled?" - I'm not quite sure I understand the question. Do you mean how the solder balls are currently being applied? (They are being "spit" on the solder joint and melt by a quick laser impulse while "in the air"). The solder balls are about 200µ, the solder joint around 50µ.

    Yes - the novelty in the described technology is that a silicon chip can be DIRECTLY mounted to the circuit (no black plastic packaging at all). The components soldered are small high-quality (the chips currently used are 2x2 cm) chips which are used for measuring particles. The CERN uses this technology to minimize insensitivity areas (dead gaps). The collider is built of many of these small chips (mosaic-pattern) which are used for measuring. I.e. colliding particles need to be measured - in order to be able to measure EVERY particle there must not be any dead gaps. (You can imagine a big circular tunnel which consists of linked chips on the inner layer).

    @alfacliff I'm not quite sure about the used power levels + I assume you mean how long are we lasering the solder ball? Thats only a small impulse to melt the ball (less than a second). With regard to that, however, I do believe that these points aren't that important since we are only looking for other application fields.

    I just attached a short pdf with a some additional information if someone's interested.
    Another aspect that could come in handy - mounting one chip takes about 3-5 minutes (!) which implies that this technology will probably only be worth it for expensive chips!

    As said - thanks a lot for the response - however I'd like to remind you that the technology WORKS and is RELIABLE. So there is no need for further development on that side. What we are looking for are additional application areas. A la applications like "Where does heat generation damage our expensive chips?" / "Where do we have problems replacing our chips in mosaic-patterns?" / "How can we minimize dead-gaps in our measuring systems?"
    Our project goal is to evaluate these application areas and find new uses in the industry.

    best regards.
  6. Papabravo


    Feb 24, 2006
    For the vast majority of electronics applications I've never even considered working with dice. It seems like you would have to solve a number of additional problems in order for that to be practical. If you look at a typical pick and place machine you can hardly see the arm move it is so fast. By comparison 3-5 minutes per chip seems agonizingly slow. Most of us here are hobbyists and seldom if ever think of or require advanced manufacturing techniques. That said it would not surprise me if somebody here comes up with an idea. Good luck.
  7. wayneh


    Sep 9, 2010
    I frankly don't see the novelty here. Is there a patent involved? Precisely place solder droplet, blast it with a laser. Yawn.
  8. BR-549

    Well-Known Member

    Sep 22, 2013
    Let me see if I understand this.
    You want us to recommend new applications for this technology.
    But you are not interested in going thru and explaining the technology.
    You just want the recommendations because it works and is reliable.
    There are many things that work and are reliable that we do not use because it would be silly.
    Like taking 3-5 min. to solder 50 micron size connections on a 2x2 cm die.
    Have you researched your competition?
  9. DLS Soldering

    Thread Starter New Member

    Mar 24, 2015
    3-5 minutes IS agonizingly slow which takes away a lot of the industrial possibilities (as we already encountered). The technology therefore only is used for high-quality chips which are VERY expensive (you really don't want the chip to break). The speed could with further development be enhanced by a not insignificant margin - this, however, is not needed by the CERN and hence no further development goes into this.

    Of course I am interested in explaining the technology (here even more than with other potential users or experts that are working with similar things) - nonetheless it's not necessary to go into too much details because we are not trying to change anything tech-specific.
    To be frank on this - the project is ONLY about finding new application fields. +on a side note: the CERN is not trying to generate any profits with their technologies but to create a positive impact for the industry, if possible.

    We do know that this process isn't suited for mass production but for highly specific application fields, which aren't always easily accessible (especially for us without a technological background) - thats why we are trying to research across a broad front. And to answer BR-549's last question - Actually we are looking for the "competition" in analogue markets but didn't get too much information so far - we are looking for the ones that could and would use it if they knew about it.
  10. nsaspook

    AAC Fanatic!

    Aug 27, 2009
    Mounting die directly to circuits without encapsulation is great for what CERN is doing (sensors) but has problems for regular electronics as it makes circuits sensitive to changes in light due to the photo-electric effect on the PN junctions unless you glob something over it.
    http://www.silabs.com/Support Documents/TechnicalDocs/AN0878.pdf
  11. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    Possible applications might be:
    1) Close mounting of existing CMOS optical sensor arrays to create super-arrays, e.g. cylindrical arrays for panoramic imaging. This could be useful in medical imaging, or industrial inspection of the inside of pipes, tubes and the like.
    2) Complex arrays of other transducers, or LEDs.