PWM or TTL controlled driver

I want to build a PWM driver for my red and UV diodes. (6X blue ray and open can Red) Has anyone tried this approach? I am using an LM317 with pot to set the current then a 2N2222 to tie the adjust pin to ground when I want to turn it off. Any one have any thoughts on the best) and safest way to do it. It has been a few years but I'm back in the Laser Saddle! Going to use this in a RGV configuration driven from an AVR for generating the PWM and a three or four motor Spiro.

Thanks,

Jules

Jules said:

I want to build a PWM driver for my red and UV diodes. (6X blue ray and open can Red) Has anyone tried this approach? I am using an LM317 with pot to set the current then a 2N2222 to tie the adjust pin to ground when I want to turn it off. Any one have any thoughts on the best) and safest way to do it. It has been a few years but I'm back in the Laser Saddle! Going to use this in a RGV configuration driven from an AVR for generating the PWM and a three or four motor Spiro.

Thanks,

Jules

Click to expand...

If you sink the feedback current on the 317 like that, kiss your diode goodbye. Pretty good chance it will go the 317's upper internal current limit. Remember the 317 works by maintaining a constant 1.24V across the feedback circuit. You shunt the 1.24V to ground and it will open the pass transistor wide open.

If you must do this, with a 317, put a large N channel Fet across the laser diode and when you want no output, drive the gate to saturation. This bypasses the laser diode and runs the constant current into ground, via the fet. You'll have to find a say 5 Amp fet, heatsink it, and drive the FET gate to at least 5-7 volts or more, and have a pulldown resistor on the gate to turn it back off.

Perhaps best to find a flexmod or microboost driver from DrLAVA?

Steve

Ahhh... But that is not what should happen. The LM317 ia actually a voltage regulator. In "normal" use there is voltage divider on the output to ground with the adjust pin holding 1.2 volts between adjust and output. So, what I am hoping will happen (i'm going to experiment tomorrow with a power LED for a test load) is that the output will drop to 1.2 volts plus the drop across the transistor. All in it should be less than 2VDC. This should turn the laser diode off. Your scenario would happen if the adjust and output pins are shorted. I'll post my results tomorrow.

Thanks,
Jules

microboost does ttl too, if buying a driver is an option

hi

i want to do much the same, my idea was to use a transistor to switch in-line a resistor in parallel to my base current resistor (your 2.7 Ohm one) via say a 555 timer, this way i can control upper and lower current limits, could also use a pots instead of resistors.

Success!!!!! I had my schematic drawn incorrectly. I built three of these on one board and will be using them to drive a RGV configuration similar to the White Fusion. I fired it up with a 1W CREE White LED as a test load and it went straight to 1.7amps and would not adjust. Then I realized I needed to connect it to the adjust pin, not the Vout pin. LOL! then I adjusted it for about 120ma and tried the TTL +5V thing with the transistor. Worked like a charm. I am not using any output capacitors as they would slow down the response and potentially feed back into the LM317. After checking the Data sheet, I put a 10uF on the input right at the IC.

Then I connected my new 6X Blue ray and verified it will modulate with the same way. Finally checked it with a Red LPC-815. That works too. Now to find out how fast I can modulate these things. In theory in the meg hertz range should be no issue.

More to follow on this project.

Jules

I have years of experience with LM317 doing things it should not, including crude current modulation. I'm betting its not so simple. I usually do not place bets I cannot win. Perhaps you should look at the current output on a scope before declaring this a success? You also have one other issue, anytime the potentiometer wiper lifts off the rails, while being adjusted, it slams to full current for a fraction of a second, and nothing is protecting the 2n2222.

Lm317 is not rated for 100% unconditional stability, and the wirewound resistors are inductive at HF and VHF where it likes to oscillate. I'm Just asking you to be careful before you submit this to the public, where it can spread like wildfire and cause issues way into the future.

What is the undefined state it slips to when adjust is grounded? I'm curious if current goes to zero or is really just dropping below the LEDs light threshold? Try a couple of different 317 clones as well. Ie On Semi vs National... Some very interesting things happen with the ON/Motorola part.. Try it with a 317K, and its oscillation city.

(Hint, I built the same circuit years ago. Its not as flexible as it seems)

Steve

I hear you. What happens is that the output voltage drops to 1.2 VDC (give or take a little) National has a similar circuit in the data sheet. The transistor pulls the output down to 1.2 when used for protection while using the LM317 as a voltage regulator. Also, if you look at the board design of the FML driver for the white fusion, it uses LM317's also with some transistors. I suspect that it functions similarly. I plan on analyzing this further and I have not run it above 150ma yet either. I am also not selling these or making claims other than as a pure hobbyist.

Jules