Page 1 of 2 12 LastLast
Results 1 to 10 of 13

Thread: Best way to modify a constant current driver for blanking?

  1. #1
    Join Date
    Oct 2009
    Location
    Eindhoven, The Netherlands
    Posts
    921

    Default Best way to modify a constant current driver for blanking?

    I got a few of these modules and the drivers that came with them, which are basically LM317-based constant current sources. Not the best in the business, but they do the job.

    Obviously, they only allow the lasers to run in CW mode without any way to switch them off. I was thinking of a way to implement blanking (or even modulation) with these driver circuits, the first thing that came to mind was to put an N-type MOSFET in the ground line of the LD, making the laser switch off when no voltage is applied to the MOSFET's gate.

    I am a little concerned of damaging my LD with this brute-force method, though. Will switching the whole circuit with a MOSFET introduce overshoot on the LM317's output current which can fry my diodes? Or will it work effortlessly and allow me to get TTL blanking on the cheap?

    Another option is to use a BJT or N-type Darlington stage in the same spot for continous (analog) modulation, but the LM317 might protest when I'm pinching the current. Ideally, the supply stage should only be a current limiter, not increasing the supply voltage when the load resistance increases.

  2. #2
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    shunt the diode, do not put the modulator in series. 317 is a poor choice for a current source for a LD. It can suffer from oscillation and horrible overshoot. Put your N fet across the LD to ground, the current source never sees a major change this way. Yes, you will need to invert the drive, but using P channel has some down sides. Make sure you get a "logic level" fet, as some need major gate drive voltages. This is one case where a TiP41 or similar NPN darlington shines.

    317 is used by people to build cheap drivers and pushed by people who should know better, trying to make a quick buck.

    Its fine for big leds, that have a huge die and are spike tolerent.

    And yes, with some care, a 317K can be set up for a limited range of analog modulation, but in this users experience, I will build the or buy the opamp current source every time.


    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  3. #3
    Join Date
    Jan 2009
    Posts
    1,622

    Default

    I had considered the option of applying a modulation signal to the adj. terminal of the 317. I don't know how fast an LM317 will allow itself to be modulated in this way, or if the idea will even work, but I thought I'd toss it out there anyway.. haven't had time to try it first hand. Some additional circuitry would be needed to set how high the current can ramp up, so that the diode would be protected.

  4. #4
    Join Date
    Oct 2009
    Location
    Eindhoven, The Netherlands
    Posts
    921

    Default

    Quote Originally Posted by mixedgas View Post
    shunt the diode, do not put the modulator in series. 317 is a poor choice for a current source for a LD. It can suffer from oscillation and horrible overshoot. Put your N fet across the LD to ground, the current source never sees a major change this way. Yes, you will need to invert the drive, but using P channel has some down sides. Make sure you get a "logic level" fet, as some need major gate drive voltages. This is one case where a TiP41 or similar NPN darlington shines.

    317 is used by people to build cheap drivers and pushed by people who should know better, trying to make a quick buck.

    Its fine for big leds, that have a huge die and are spike tolerent.

    And yes, with some care, a 317K can be set up for a limited range of analog modulation, but in this users experience, I will build the or buy the opamp current source every time.


    Steve
    Shunting the LD means I do need to dissipate quite some heat through either the MOSFET, the preceding 1ohm resistor, or the LM317 itself, since it will keep cranking out the set current no matter what. And with a supply of about 12V and an LD current of about 300mA, that's quite a few Watts I need to get rid of.

    If there's a good alternative, I'd be happy to give it a shot. I've been looking at Die4Drives but they don't seem to be available for purchase anymore, and I don't have the gear to solder SMD components so I really need a through-hole circuit that only uses components that are available in the EU.

    I can also use a P-type that switches the supply in the positive line, but I don't really think that solves the problem.

  5. #5
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    A p switch means the 317 chip is going in and out of regulation. This is bad. Using the shutdown circuit in the 317 data sheet is contrary to using it for constant current because it is ground referenced. You need to shut down the LD by shunting but not the regulator. Now, you can reduce the dissipation by using a 1 ohm resistor in series with the fet, but a good fet has a RDS(on) of .01 Ohm. Your only heating the fet with 1 watt or less for red or violet diodes, if it has a low rds on, or you could get a lavadrive.

    If you wanted to get complex you could have two parallel sense resistors with the fet in series with one, but I suspect that is going to need a lot more then 5V of drive and a lot of testing for nothing.

    P=IE E= IR 250 mA of diode, E (on) = .250 x .01 RDS(on) = .025V P= IE .025 * .250 = .006 watts , so 1 watt is a reasonable figure. even if the fet is unsaturated , call it 2.5V across the fet .250 x 2.5 = .625 watt. The saturated fet stays cold with even a long m4 screw as a heatsink.

    Steve
    Last edited by mixedgas; 11-16-2009 at 11:03.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  6. #6
    Join Date
    Oct 2009
    Location
    Eindhoven, The Netherlands
    Posts
    921

    Default

    Quote Originally Posted by mixedgas View Post
    P=IE E= IR 250 mA of diode, E (on) = .250 x .01 RDS(on) = .025V P= IE .025 * .250 = .006 watts , so 1 watt is a reasonable figure. even if the fet is unsaturated , call it 2.5V across the fet .250 x 2.5 = .625 watt. The saturated fet stays cold with even a long m4 screw as a heatsink.

    Steve
    It's not the FET (with low RDSon) that I'm worried about, more so the other components in the circuit. The 300 mA at 12V has to go somewhere, as the LM317 will try everything to maintain it. Worst case, it sees a load of only 1 ohm and therefore will drop the supply down to 300 mV across that 1ohm resistor. The 1ohm has little trouble dissipating the 0.1 Watt that's thrown at it, but the LM317 has to gobble up the remaining 11,7V at 300mA. That's 3,5W of power, which means I'm going to need quite a heat sink and a fan club to go with that. The regulator is getting pretty hot operating at its rated current right now.

  7. #7
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Attached is a educational idea.

    download the really fab and free LT SPICE

    http://www.linear.com/designtools/software/

    enter this circuit in.

    Try it, note, you'll have to tie the mod input to a V source or a square wave source in LT spice.

    LT doesnt make a LT117 any more, so a model is here.

    Lets see how long it takes before this appears on one of those circuit collection web sites as a working circuit

    http://www.electro-tech-online.com/a...m317-lm317.zip

    Unzip this file. Save LM317.sub to the SUB folder in LIB. Save LM317.asy to the MISC folder, which is under the SYM folder in LIB.

    You'll soon see why the lm317 is a poor choice for a modulated LD driver.

    Watch the startup and mod transients in LT SPICE.

    USE 4 1n4001 in series with a 1 ohm resistor to model the LD.

    I would NOT try this with a real LD.

    Steve
    Attached Thumbnails Attached Thumbnails DADSBACK.bmp  

    Last edited by mixedgas; 11-16-2009 at 17:14.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  8. #8
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    [QUOTE=mixedgas;123920]Attached is a educational idea.

    download the really fab and free LT SPICE

    http://www.linear.com/designtools/software/

    enter this circuit in.

    Try it, note, you'll have to tie the mod input to a V source or a square wave source in LT spice.

    LT doesnt make a LT117 any more, so a model is here.

    Lets see how long it takes before this appears on one of those circuit collection web sites as a working circuit

    http://www.electro-tech-online.com/a...m317-lm317.zip

    Unzip this file. Save LM317.sub to the SUB folder in LIB. Save LM317.asy to the MISC folder, which is under the SYM folder in LIB.

    You'll soon see why the lm317 is a poor choice for a modulated LD driver. its slow! It overshoots!

    Watch the startup and mod transients in LT SPICE.

    USE 4 1n4001 in series with a 1 ohm resistor to model the LD.

    I would NOT try this with a real LD.

    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  9. #9
    Join Date
    Oct 2009
    Location
    Eindhoven, The Netherlands
    Posts
    921

    Default

    You don't need to convince me that the LM317 is quite a lousy current source. I am well aware of that, and I don't intend to buy one if I were to build a new driver circuit.

    I am only figuring out options that might be useful in converting the existing driver circuit I already built (it came with the laser). Building one from scratch is still an option, if I can get the right schematics and components.

    Getting Flexmods or Lavadrives is still an option, as long as I don't need to assemble any SMD components, and the thing's able to blank my DPSS as well. The green NewWish module is dead slow.

  10. #10
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Getting Flexmods or Lavadrives is still an option, as long as I don't need to assemble any SMD components, and the thing's able to blank my DPSS as well. The green NewWish module is dead slow.[/QUOTE]


    Lavadrives are solder and adjust, thats it... That 317 circuit does work in the sim, but in reality..... If you must build. there is stanham...

    Steve
    Last edited by mixedgas; 11-17-2009 at 19:28.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

Posting Permissions

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