Converting open drain pulse to TTL signal?

Discussion in 'The Projects Forum' started by SimonGylden, Oct 23, 2009.

  1. SimonGylden

    Thread Starter New Member

    Oct 23, 2009
    4
    0
    Hello Folkens!

    Has anybody got any idea how to convert an open drain flash pulse from a camera to a TTL signal that a Luxeon BuckPuck can utilize?

    I need the light and camera to be in sync, otherwise flicker occurs as I'm using high shutter speeds. It's for a 8mm film transfer system where the camera is pointed directly at the light with the film strip in between.


    It should be doable, but both producers' support guys had no standard solution to the problem, so any help is much appreciated.

    The camera has a 'Flash control (open drain)' pin with this note: Max. 0.2 A (ID) for open drain MOSFET'.

    The LED driver states ' TTL/CMOS logic level on/off control'. A 'Pulse and strobe capable (control input)'-pin.



    Basically, I don't know what the heck I'm talking about, so please explain in layman's terms. :rolleyes:


    Hope somebody can help. Thanks!


    --------------

    The camera is a The Imaging Source 31BF03 - documentation here:
    http://www.theimagingsource.com/en_US/products/cameras/firewire-ccd-color/dfk31bf03/

    The LED driver is a Buckpuck 3021-A-E-700 - documentation here:
    http://www.luxeonstar.com/buckpuck-700ma-ac-led-driver-pcb-mount-p-28.php
     
  2. spacewrench

    Member

    Oct 5, 2009
    58
    1
    I think you can do it just by wiring a couple of things together (no external components necessary). First, I'd use a plain piece of wire to test the BuckPuck: connect the CTRL pin to the LED- pin and confirm that the LED goes on. Disconnect and it should go off. (This is what Figure 13 suggests.) If the LED doesn't go off when the CTRL pin is disconnected from LED-, then you can try a resistor from CTRL to REF. (I'd try about 4.7Kohms.) Repeat the CTRL-to-LED- test. You want the LED off until you connect CTRL to LED- with a wire.

    Once you have that working, then connect the camera GND to the LED- pin, and the camera STROBE_OUT to the CTRL pin. For the camera-to-puck connections, I'd probably start out with a wire for LED- to GND, and a 1K resistor for STROBE_OUT to CTRL, because it looks like the camera is a little spendy and you don't want to blow it up. With the resistor in place, I'd expect the LED flash to be a little weak, but it should at least be timed correctly. If that works, go down to a 470ohm resistor (brighter flash), and finally, you should be OK to just go with a straight wire.
     
  3. SimonGylden

    Thread Starter New Member

    Oct 23, 2009
    4
    0
    Hi spacewrench,
    thanks a lot for your reply. (I'm not sure I get it though..)

    First of all, could you tell me what the characteristics of the TTL and the Open drain signals are? What's the reason for their incompatibility? Maybe this will help me understand the whole thing better.

    Next, I should mention that the camera draws its power from the firewire connection only, so I don't think it will be able to drive the puck & LED. I'm not sure if this is what you suggested or it's implied in the 'open drain' term. Right now the unsync'ed LED draws its power from the modified filmprojector's lamp socket.

    Finally, the engineer at The Imaging Source said it would take an external power supply to make it work. (According to the support guy).

    I don't know the implications of all this, but I hope you will expand on your suggestion.


    Here's a pic of the system:
     
  4. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    After looking at the BuckPuck datasheet, I'm convinced that this should work.
    See attachment.

    EDIT: Camera common is the same as GND on the camera header.

    EDIT: Added R3 to schematic. Transistor(s) will smoke without it.
     
    Last edited: Oct 25, 2009
  5. spacewrench

    Member

    Oct 5, 2009
    58
    1
    I agree that this circuit is a reasonable adaptation of Fig. 14 in the BuckPuck datasheet, but I wonder if this simpler version wouldn't work. According to Fig. 13, you should be able to put a 0-5V TTL or CMOS signal into CTRL, where the signal is referenced to LED-. (Essentially, I think, you short CTRL to LED- to turn the LED on, and open the circuit to turn it off. So why not just use the open-drain output to do that? I think the on/off sense is correct.)

    (In case it isn't clear, I think you can leave out all the stuff in yellow, and just put a wire in where the red line is.)
     
  6. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    According to the BuckPuck datasheet, the Ctrl input has a 1kΩ resistor to GND (LED-). To turn the LED(s) off, Ctrl has to be pulled up to at least 4.2V. An open (NMOS) drain cannot do that. See Fig. 14 in the BuckPuck datasheet for an example.
     
  7. spacewrench

    Member

    Oct 5, 2009
    58
    1
    Aw, rats, I didn't see that part of the datasheet. And with such a small pull-down, you can't even use a passive pull-up arrangement as I'd suggested.

    Simon, I guess you'll have to build the little circuit Ron drew. Fortunately, it's not too complicated.
     
  8. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    I think the reason for the 1k resistor is that simplifies brightness control, and also allows full brightness with no connection to CTRL. The brightness can be controlled with the voltage on the control pin, and a rheostat (pot) from REF to CTRL forms a voltage divider with the 1k resistor. See Figs. 9, 11, and 16 in the datasheet.
     
  9. SimonGylden

    Thread Starter New Member

    Oct 23, 2009
    4
    0
    Hey you guys,

    Thanks a lot for the design. I still don't know what it all does exactly, but as long as it works who am I to complain.

    Building this, is there anything I should be especially careful of in regards to blowing up either the camera, the puck or myself? :p

    I'm super grateful for this!
     
  10. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    I'm glad you asked. I took another look at the schematic, and realized I had omitted a resistor which is needed to limit the collector current of Q2 and the base current of Q1. I also changed the value of R2, which is not critical, so you only need one value for all 3 resistors. I edited the schematic.
    You should not have any other problems, as long as you wire it correctly.
     
  11. SimonGylden

    Thread Starter New Member

    Oct 23, 2009
    4
    0
    On a final note, I can confirm that the circuit works!

    The engineer at TIS ate his words about the external power supply and approved the design.

    I tip my hat! Thanks for all your help. :)
     
  12. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Excellent!:D
     
Loading...