Page 2 of 2 FirstFirst 12
Results 11 to 13 of 13

Thread: LambdaPro problems, any ideas?

  1. #11
    Join Date
    Dec 2006
    Location
    Netherlands
    Posts
    983

    Default

    I wish I still had my notes on the PSU controls but I must have thrown them out after I sold it and moved (sorry).. I cannot say with any certainty which way to turn the pots. If the red led is on in your PSU when it runs and leave the current limmit alone you can turn the current set pot safely (as current is still limited) any direction until the red led goes off and you should get reduced power output. If the red led is not on you want to be more careful.

    I do remember, and it can be seen in my gallery, is how the laser cavity is built out of aluminum rings, each holding an optical element. I did not take it apart further so cannot confirm exact configuration but there are four rings probably holding: grin lens, vanadate, ktp, OC. In this case it might not be a bonded crystal or am I overlooking something?

  2. #12
    Join Date
    Dec 2006
    Location
    Texas, US
    Posts
    31

    Default

    It looks to me like its multi-moding pretty badly (TEM01 or worse). Chances are its temperature related, but how to fix it... would be hard to say. A regular heatsink may not be doing all you need to keep the laser cool, and you might want to consider incorporating your own TEC. Turning down the power? I'm not really sure. Normally if a laser is having mode problems, you RMA it as its not something you easily fix yourself, but since you're past the warranty...

  3. #13
    Join Date
    Mar 2006
    Posts
    2,478

    Default

    Heat...
    From the post where Pit quoted the guy whose laser it is: "However the laser and PSU are both incredibly cool at the time it fails".

    In this case, there is no real need of a heatsink beyond what is there, and even the fan at the back inside the head isn't doing much. That's there on the assumption someone might be running it at risk, but the thermal bonding between that block and the head's baseplate is MUCH more important, and that's not going to fail unless you start getting out the heavy screwdrivers.

    If the laser runs cool on a modest heatsink, the only important role the heatsink has is a thermal mass, to make sure that ambient changes cannot happen faster than the TEC and its controller can keep up with.

    Don't be tempted to turn up the TEC to cool the diode, because you'll only put it out of efficient wavelength coupling, and possibly cause a thermal runaway where the TEC sensor detects a rise caused by the entire case temp, but the TEC controller doesn't know this, so it works to cool locally, raising case temp further... This is another reason that a thermal mass is wise. So long as that mass is there, and the laser is cool enough, it does not need it; but if it is NOT there, and IF, for any reason, the laser head temp rises, you again need that mass to dissipate heat and stabilise the rate of change. This is one of the reasons why setting a new TEC control point on RwT is such a lottery. It takes a long time between observations to get it right, and there could actually be MORE than one correct point for TEC temperature optimisation, if the circuit isn't based on an absolute value. It should be, but given the varying equilibria that LamdaPro lasers sometimes find themselves in, it may not be, or at least not be accurate enough.

    The only real way to avoid overheating the composite crystal (assuming it even is one in the UGA's) is to back off a bit on drive current for the pump diode.

    I suppose you could shake it in a very quiet room to see if something sounds loose in there, but if it was, I think the effect would be more dramatic than a flared or multimode beam.

    My guess is that this laser has a flawed crystal, turned for best output, but was a fine balance, and has settled over time. The adjustment needed to fix it is beyond anything but the kind of elaborate dissection and rebuilding Zoof did on one of his. Unless this was done you couldn't even reach the parts affected. If it really is very finely balanced, it might be possible to use controlled heat or cooling while the laser is NOT powered up, to try to see if thermal stress can countaract whatever settling has occured. Totally risky move though, best for last ditch attempts. Limit upper temperature to 90C, or 85C to be safe, in case 93C indium based solder is used in construction, and a minimum of -10C (not lower than -20C anyway, because the UV cure epoxy used is brittle, even more so at low temp, and won't match the shrinkage of aluminium, which might be too much beyond -20C or so). If I were trying this I'd try cold first, sealed plastic bag with laser head inside, and put in a fridge right under the ice shelf for a couple of hours then back to room temp for same before removing and trying to power it. I won't try to predict the results, but I doubt it could make it any worse.
    Last edited by The_Doctor; 03-11-2007 at 07:18.

Posting Permissions

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