A little noob help on the DDL driver...

Hi all,

I've been reading the forum for some time now and just recently joined. Thanks for all the info so far.

I've been scouring the web and the forum for a couple of specific things I need some help with. Sorry if I am reasking things that may have been addressed 100s of times. I just can't find it if it has.

I've built my laser driver probably 4 times and can't get good readings. I've followed 2 or 3 different guides.
http://laserpointerforums.com/f67/laser-driver-ddl-schematic-60077.html
Laser driver - It can be done
and more here on the forum.

There are a few basics I don't understand - what is my output voltage supposed to be with this LM317 driver? I consistently get .7 volts lower than whatever I'm putting in. I've tried it with 7,9, and 12 volts. Isn't it supposed to be a fixed output?
Also, what do good output readings look like from a test load? I usually get almost 0 volts. What should I be seeing?

I'm building the standard dvd burner diode laser with the LPC826.

Hi and welcome to the forum,

I would recommend you post detailed pictures or diagrams of the project you are working on and with more details on the test load, input source, etc.

It will be much easier to help you then. One thing I can say for sure is you need at least 3V more on the input than the voltage you will be meeting at the output. :beer:

The DDL driver is a constant current driver. The current is held constant but the voltage varies in accordance to the resistance of the load via Ohm's law. Also, the input must be around 3V higher than the Vf of the diode so that the circuit can properly regulate the output.

Yes, post pictures of your circuit. It may be that your circuit isn't hooked up right. Show us where you're measuring too.

alright, here are some pics and more details.

As stated in the OP, I have breadboarded this multiple times from a couple of different guides. This specific version is based on the physical image daedal posted (here).

This is the circuit with 8v going in from 2 li-ion 14500s. I actually want to end up running it off of 12v.
The leads going off the left are shown more clearly in other pics. I have 8v going in and 7.3 v coming out. When I put 12 in, I get 11.3 coming out. Is that safe to put through my diode?



When i test my amps, they are reading between 0 and 120ish milliamps based on my potentiometer. Seems to work for varying output. My question is if my output voltage is changing based on whatever I put into it (8v or 12v), how do I figure out the proper pot setting to keep my diode at 150mW or so?

These pics are two different test loads. I am going from the ground on my driver to the 1 ohm resistor and the + from driver to the 4 diodes. When i read the multimeter across the resistor i get 0 v and 0 amps...

what readings should i be getting from the test load?

End result is I'd like to run the lpc826 around 250mW from a 12v source.

Any direction would be much appreciated! Let me know if more details are needed.

Merry Christmas!

You want a 250mw LPC 826 so you need to feed around 400ma to that diode (based on Cyparagon's graph).

To find the resistor needed by the LM317 you divide the reference voltage by the output current. That is 1.25v/.4A = 3.125 ohms. The nearest value you can get is 3.3 ohms. Put that back into the equation and you get an output current of 378ma. You need to put the resistor in between of the adjust and output pins of the LM317.

That is all you need to set the output current. Since laser diodes need constant current you do not need to worry about the voltage as the regulator will drop it to the level needed by the diode.

You also need to heatsink the regulator because it will dissipate all the excess voltage into heat. Running it from a 12v source will result in 3.59W of heat from the equation (12v-2.5v)*.378A. 2.5v is the forward voltage of the diode at 378ma. You will need a heatsink that can take all that heat away.

You're supposed to be measuring the voltage across the resistor. Attach the meater in parallel with the
resistor and the volts reading will be within 5-10% of the actual current going through the resistor (and
hence the diodes). It would be more accurate if you had a 1% tolerance or better resistor. Wirewound
is okay on a linear driver like this, but it won't work properly with a switch mode driver if you end up
using one of them in the future. Jufran sells a nice test load in the sale and trade section which has
jumpers for configuration and the proper resistor. The price is decent, too for what you get.

You can also just attach the meter in series and put it in current mode, but this is fraught with possible
problems for a beginner. Just measure the voltage drop across the resistor and you will be fine. That is
unless you have already blown the fuse inside the meter, which could explain why it's not reading
anything. Give the fuse a check and replace it if necessary.

Go and get some safety glasses from Survival Laser if you have not already. At 250mW you are
entering dangerous territory, even if you're just looking at the reflected dot.

Thank you both for the information and tips!

Zeebit: I'll implement the new resistor and keep you posted on how it goes.

What's the deal with resistors? Seems like anything less than 10 ohms is hard to find. I have some 1 ohm resistors and they are huge. Why so big if resistance is so low? 3.3k ohm resistors popped up. We're talking just 3.3 ohm, right? not 3.3 thousand?

Also, if the output calculations are giving me x number of ohms, how does the 100k ohm pot come into the mix? do I have it's resistance all the way down?

I have been reading values across the resistor on the test load. The fuse is fine because I get output when I hook straight into the driver. It's only when I test through the test load that I get 0s.

The formula for setting the current limit of regulators is usually Vref/R=Iout. Rearrange the equation and you get R=Vref/Iout. This is the same with the LM317 in constant current mode. Its Vref is 1.25v.

As I said above you need a 3.3ohm resistor. As you can see from the equation, the higher the resistance, the lower the output current will be. This is where you add a trimmer if you want an adjustable output current. If you only need a fixed output you don't need the trimmer.

The size of the resistor determines its wattage rating - how much heat it can take before its value drifts off and eventually it will burn up.

The formula for power is P=V*I or I*I*R. If you use a 3.3 ohm resistor you get an I of 0.378A. Use that value and you get a P of 0.47W.

You can also test the output of the driver by just simply hooking the output directly to an ammeter.

Don't forget the heatsink. Its a must.

Try DigiKey or Mouser. Search engines don't handle component values well because of the fuzziness.
You have to go into the product index and use the filters to narrow down the results.

Mouser will probably be your best bet for nailing down the values you need. Digikey's website still looks like it's from 1999.

Also, when you take pictures, try to make them as clear as possible. In those photos above, it's hard to see what is connecting where, or what pins, etc. Really zoom down to the main circuit (make sure it's not blurry), and crop out the stuff you don't need. You can even annotate the picture to tell us what is going where, or make a separate diagram you draw on paper.

Alright team, I got it working. I built the driver with the 3 ohm resistor and heat-sinked the lm317. It powered up nicely and gave a great beam. Then I wiggled something....

Now the laser is very very dim. Almost like a normal red led. I checked all my connections. Nothing. Rechecked output, still looks good -- getting about 350ma.

I rebuilt the driver on a breadboard to troubleshoot and it's still dim. I have one other lpc826 in case I fried this one so I soldered it up and tried it. It is also very dim.

What could cause this? I've tried two diodes on two drivers and get the same results.

It's fried. Don't wiggle things. If the diode gets disconnected for even a fraction of a second, the capacitor
will charge up. Then when the diode reconnects, the capacitor discharges into it and fries it. Always
short the leads of a driver before connecting a diode to discharge the capacitors.

I never bothered with caps when I built my first laser from a DVD burner. Never got any spikes and I was lucky enough not to kill the diode on my first build.

LPCs are very cheap though. I suggest you just use the LM317 and the set resistor, nothing more. The regulator is stable enough for the diode.

Yeah, dang. I fried two of them... that's exactly what I did. I had had power in the driver before I hooked up the diode.

So, as long as the laser is hooked to the driver with nothing in the capacitor, I should be fine if I just leave it hooked up and only turn power off and on at the PSU, right?

Hey Bionic-Badger, I noticed you're here in Utah. Do you know of any hobbyist groups around here?