Page 2 of 7 FirstFirst 123456 ... LastLast
Results 11 to 20 of 67

Thread: Bizarre simpledrive II driver problem

  1. #11
    Join Date
    Nov 2008
    Location
    Cleveland Ohio
    Posts
    2,599

    Default

    Here's an idea to solve this forever. Load switching. Make a circuit that hooks to a dummy load and switches over to the diode after a few seconds. This then absorbs spikes. Do the same on turn off. Thinking a dpdt on off switch. One side is power and the other is the switch circuit. One shot 555 and 4066? Then again zener can lock voltage. Is there a currrent version of a mov? I'm sure this can be engineered out.

  2. #12
    Join Date
    Feb 2008
    Location
    Northern Indiana
    Posts
    921

    Default

    Quote Originally Posted by jors View Post
    Do you power with Lab PSU while adjusting? Which brand is your PJ PSU? I got 4 P73 killed by a chinese 'no brand' PSU months before...now all seems OK with Meanwell
    I set-up with a lab supply. The supply in the projector is a meanwell. Although, to be honest, I have had a few meanwell's go up in smoke in LED installations. Not sure if they were genuine meanwell? They were very sparse and cheaply built when I opened the ones that failed.

    Quote Originally Posted by kecked View Post
    Here's an idea to solve this forever. Load switching. Make a circuit that hooks to a dummy load and switches over to the diode after a few seconds. This then absorbs spikes. Do the same on turn off. Thinking a dpdt on off switch. One side is power and the other is the switch circuit. One shot 555 and 4066? Then again zener can lock voltage. Is there a currrent version of a mov? I'm sure this can be engineered out.
    I'm thinking an easier solution may me to add a low inductance resistor in series with the diode. Say the power supply = 15V and the Vf of 2 series blue LDs is ~10v. A momentary short on the driver transistor would put a max of 5V on the resistor (plus the drivers sense resistor). A 1ohm resistor would allow only 5amps to flow. Probably not enough to kill the diodes on a start transient.
    Last edited by logsquared; 07-10-2017 at 08:47.

  3. #13
    Join Date
    Feb 2008
    Location
    East Sussex, England
    Posts
    5,248

    Default

    If this is the project I'm thinking of, I've got the other 10 drivers of the batch in a pair of projectors (used them this weekend just gone actually, still working fine)
    Although they did have a bit of leakage last year where I had to adjust the quiescent current (perhaps you recall the conversation BBE)
    Frikkin Lasers
    http://www.frikkinlasers.co.uk

    You are using Bonetti's defense against me, ah?

    I thought it fitting, considering the rocky terrain.

  4. #14
    Join Date
    Nov 2010
    Posts
    346

    Default

    99% we are dealing with some defective part. Unfortunately, I can not remotely check this ...

    My offer for replacement is still in force. This fall, I launch new drivers for lasershow, I can send you the first of them as a replacement

  5. #15
    Join Date
    Nov 2008
    Location
    Cleveland Ohio
    Posts
    2,599

    Default

    How about a picofuse. Might pop a bunch but diodes will live. Add lasorb for voltage spikes. Then your covered for current and voltage. Maybe add a current shunt in front of the driver. So the driver can't ask for more than you set the max even if the driver fails entirely. I think a resistor might work but It would get hot. Thinking a transistor or get of some kind. Sort of a driver on the driver.

  6. #16
    Join Date
    Nov 2010
    Posts
    346

    Default

    It's all a workaround. A good driver should not kill diodes.

    In most cases, laser diodes do not die due to a current spike, but because of the reverse current spike, generated by self-inductance in the wires. In this situation, only the Schottky diode or lasorb on diode side can help

  7. #17
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,478

    Default

    ... how fast have your drivers to be?

    For a forums-project I'm building high current drivers with up to 9 Amps current and some ten to hundred Kilohertz bandwith.

    Here's a graph from one of my older prototypes with LM338 IC's used for constant current driving - tree parallelized LM338 gives this switching curves at 6,5 Amps current (measured across a 0,1 Ohm-resistor) without any over- oder under-oszillation

    Click image for larger version. 

Name:	Pulsform mit 3x LM338-KSQ-Treibern.png 
Views:	11 
Size:	14.4 KB 
ID:	52642


    Viktor

  8. #18
    Join Date
    Nov 2010
    Posts
    346

    Default

    How fast should it be for laserist needs? The prototype of what I will offer has about 250ns rise / fall responce at ~5A. But this is too fast for regular use and imposes certain restrictions on the feeder line between the driver and the diode. That "wiring" which I often see on the forum, is categorically contraindicated at this speed ))
    Attached Thumbnails Attached Thumbnails 0.05R-~5A-220ns.jpg  


  9. #19
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,478

    Default

    ... the needed speed depends mostly on your application - with laser-engraving or cutting on an XY-plotter-style machine (most laser-cutters are) rising/falling times of some ten Microseconds are more than enough ... but with the actual development with an XY-galvo-scanner it's more like some ten Nanoseconds -- especially with adjustable pulse-widths for "colour-marking" - here I'm dealing with pulse times between 400ns down to 10ns!

    Viktor

  10. #20
    Join Date
    Nov 2010
    Posts
    346

    Default

    Just look at the wiring of any projector in this forum and try using a driver with 200-400ns feedback in these projectors, haha. At best, you will get a RF/noise transmitter, at worst - a burnt diode. For laser shows (in my option), it does not make sense to use the driver with rise/fall faster than 1us. Also need to pay attention to the modulation signal waveform after ~10-20 meters of ILDA cable, it never happens superb.

    And another question, if I may ask you:
    please explain the nonlinearity of the signal rise on your waveform (1);
    can you show this (2) part of the waveform, say, in 1us resolution? As a rule, all the oscillations are in this section )
    Attached Thumbnails Attached Thumbnails Pulsform mit 3x LM338-KSQ-Treibern.png  

    Last edited by Bbe; 08-11-2017 at 18:17.

Posting Permissions

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