DIY Homemade laser diode driver

[Daedal]

Graphed with an Oscilloscope, I have never seen a spike from an LM317 circuit... the capacitor will actually slowly charge and then discharge to the diode... thus will give the diode a relatively soft start... the spike is from your power supply, and not from the LM317 circuit. There is a workaround though... Try getting a reliable controller that can MAKE SURE that the V-drop across the current limit resistor is in fact ~1.2V

GL;
DDL

 

[Gazoo]

Daedal,
I am using a RS 3 amp 12 volt power supply with a cigarette lighter socket. I have a RS switching power supply plugged into the socket and I can set it at varying voltages. I am not measuring current from the capacitor to the laser diode! I have a 1 ohm resistor soldered in series with the rheostat and I am measuring the voltage across it to get my current reading. I will have to go back and take a look at this. I don't have a scope yet..

 

[Daedal]

Hmm... It's probably the initial spike from the supply then... I don't see any spikes at all with my supply...

http://www.mpja.com/prodinfo.asp?number=14602+PS

--DDL

 

[danq]

For anybody building a variable bench supply for testing their GB diodes, I just found what appears to be a great (if perhaps temporary) source for decent digital panel meters... posted about it in the "Deals and Discounts" section, here

DanQ

 

[IgorT]

For most of my applications I use a 47uF 16V capacitor. To be more on the safe side, you can get a higher capacitance and a higher voltage rating, but it all depends on personal preference. A capacitor will work to regulate the line up to the maximum power that it can soak up as fast as it could. Capacitors ADD up when connected in PARALLEL. Therefore, if you want to use mutiple capacitors with different benefits (Max voltage and big capacitance for instance) you need to make sure that BOTH positive and negative leads on the capacitors are connected together. I will explain more detail later on.

Hi Daedal!

I have a question.. I just got a few nice DVD 16x LDs and went to buy parts for your LD driver.

I got them all, but now noticed you are using a tantalum capacitor... I bought an electrolytic capacitor.. Is this gonna work just as well? I think it should, since it does the same thing, it's just larger than a tantalum..

Am i right? Will an electrolytic capacitor work in the place of the tantalum? I just want to make sure..


Thanks!


Igor

 

[Daedal]

Hey Igor;

The capacitor you have should be fine as long as it has a rating of ~47uF and as long as the voltage rating is move than 1.5 times that of the diode power

We use tantalum because they are smaller and can stick right on the back of the diode

GL;
DDL

 

[aaronX987]

I am a little confused. :-?
I thought that 2 cr123 batteries would be enough to power the driver. But now I hear that it is not enough? Is there any way of tweaking it or do I just resort to another method.

 

[IgorT]

Hey Igor;

The capacitor you have should be fine as long as it has a rating of ~47uF and as long as the voltage rating is move than 1.5 times that of the diode power

We use tantalum because they are smaller and can stick right on the back of the diode

GL;
DDL

Thank you for the quick reply! I thought it was just for the size..

One more question.. What voltage does this circuit need? The schematics show a 6V battery. Does this mean the LD is driven by 6V but lower current or what?

Shouldn't these be driven by 3V? I just want to understand how it works...

The LDs i have i tested on my lab PSU, by setting it to 3V and limiting the current to 100mA and then to 200mA.. The PSU drops the voltage just enough, the pre set current is achieved..

Does this work in a similiar manner? Or something else entirely?


What is the lowest voltage this driver willl work with? I was thinking of using three or max four NiMh cells. Would that work?


Thanks!

 

[Gazoo]

The lowest voltage is 6 volts give or take a few tenths of a volt, but I do not recommend anything less than 6 volts. You have to take into account that the LM317 has a drop out voltage of appx. 2 volts, as well as the voltage drop across the laser diode, which varies but is in the area of 2.6 volts.

 

[Daedal]

Just as Gazoo said...

The LM317 itself takes ~3V out of the circuit... and the diodes (GB reds) have been a stable and steady 2.83V for all the ones I've used.

6V input is not what is going into the LD, and I believe that it was explained earlier... The LM317 has a typical dropout of 2.5V... this is the voltage that is used up by the internal pieces of the IC from diodes to transistors to resistors... whatever...

Also, take into account that the resistors will use up some voltage as well, so you need to have some wiggle room to have this circuit work properly.

Your choice of NiMh is not bad, but not good either... NiMh batteries have a Vout of 1.2V, so 4 will give you 4.8V... which is HORRIBLE... Go with 5 or even 6 of these to make sure you're ok. Another solution is to use CR123A, or 14500, or 18650 batteries... etc. (Lithium) they supply 3-3.7V, a couple of these should do the trick.

14500's are almost identical to AA batteries, so you could use a AA flashlight as a host and replace the batteries with rechargeable 14500's and you're good to go

GL;
DDL

 

[Gazoo]

Daedal, as soon as a CR123 begins to sag, the 317 begins to drop out. LarryQ had this problem until he switched over to RCR123's.

 

[IgorT]

The LM317 itself takes ~3V out of the circuit... and the diodes (GB reds) have been a stable and steady 2.83V for all the ones I've used.

Thanks! That's what i needed to know - the voltage drop...

Just yesturday i fried my second LD while hooking it up to a lab PSU set to 3V and limited the current to 100mA.
It worked fine a couple of times but when i was trying to set up a collimator lense the LD suddenly turned into a weak LED...

Does this mean the lab PSU also has a voltage spike at the start, before it reduces the voltage enough for the current to only be 100mA and your driver would work better, or am i just unlucky, when it comes to getting good LDs?

I have two more left untill i get more from the repair shop of a computer company and don't want to kill those too. So i'm building your driver circuit now..


Thanks!

Igor

 

[Daedal]

Igor, remember that there is a capacitance discharge that takes place...

It could be either a voltage spike issue, or that the capacitor managed to discharge into your LD... The discharge can be controlled, but the spike could be unnoticeable unless you can graph out the output at high speed...

GL;
DDL

 

[IgorT]

Igor, remember that there is a capacitance discharge that takes place...

It could be either a voltage spike issue, or that the capacitor managed to discharge into your LD... The discharge can be controlled, but the spike could be unnoticeable unless you can graph out the output at high speed...

GL;
DDL

Daedal! Thank you for answering all my questions! I really apreciate it.


There was this other thing.. Every time i handled the LD i grounded myself first and it survived many tests with my lab PSU (BTW: I wasn't using any capacitor, just the PSU set at 3V and 100mA - maybe that's the problem?)

The last time i tried to test it with a collimator lense, i DIDN'T ground myself, and when i touched the negative lead from the PSU i felt a spark of static charge..

The negative lead of the PSU was connected to the negative lead of the LD, but the positive leads were not connected..

So i was hoping the static discharged into the PSU and not the LD, but then when i powered it up, it was working fine for a few seconds until i tried to put it next to the collimator and then it suddenly stopped working..

Now i'm hoping it's one of the following reasons:
- PSU not good for this application: build the driver circuit (or at least add a capacitor)!
- Static electricity: Ground myself EVERY TIME!
- 100mA too much for an LD without a heat sink: Use a **** heatsink!

And the reason i'm hoping is NOT the cause:
- Bad laser diode, not capable of withstanding even 100mA...


What do you think?

Thanks again!


Igor

 

[aaronX987]

So the best thing to go with would probably be 2 rechargeable cr123a.

Ok, one more thing. Can I use a LED as a dummy doide for testing.

 

[IgorT]

So the best thing to go with would probably be 2 rechargeable cr123a.

Ok, one more thing. Can I use a LED as a dummy doide for testing.

You could use it for testing at low currents, up to 20mA, maybe 30mA.. Over that it would start dying.. But you would at least know if the circuitry is operating correctly (if you put it together right, it should...) if you'd measure everything with a multimeter... (the voltage coming out and the current going through)


BTW: If you wired 5 of the same LEDs in parralel, you cold test it to 100mA... Since the voltage drop in the LED is similiar to the one in the LD it should work. (some leds have different forward voltages tho)