Laser driver problems

[Bigbondo007]

New to lasers and driver builds guys, plz help me out. Ive built 3 drivers from 3 different designs and still have the same problem. My driver will run an led but wont run my diodes. I love building things and am fascinated by lasers but am ready to give up on this one. Ive built three ddl drivers btw.

 

[nztdm]

Ive built 2 drivers and failed. What laser diode are you using? And what power source.

I made a driver with an LM317T regulator and 5ohm resistor (and 2 RB caps). But I don't understand how they work. The regulator in one of my builds outputs exactly 260mA maximum but the output voltage is the same as the input which will kill the diode i presume (not gonna try it with my M140).

Could someone clarify how to get the correct voltage output AND constant current?

 

[DrSid]

Generally it works like this:

You apply voltage to a component .. and based on it's inner resistance, some current will go through it.
Many components do not react in linear way to voltage, diodes are one of them. Anyway for working ranges of laser diodes, the reaction is almost linear. You increase the voltage, the current will increase in linear fashion .. and so will the power.
There is usually some minimum current - called lasing treshold. There is usually some maximum voltage, and maximum current, neither of which should be exceeded. There is usually also some maximum temperature, which should not be exceeded.
Then there is some 'sweet spot' .. current at which the diode works best.

Now why you need constant current driver and how it works ? Diode changes it's 'inner resistance' based on temperature. As it gets hot, the inner resistance goes down. If you applied constant voltage to it (which for example common AC/DC converter do), the more and more current would go through the diode .. increasing temperature, decreasing resistance and increasing current again .. up to the point the diode is killed. This is called thermal runaway.

Also diode output depends on current, not voltage. So to get safe and constant output, you need constant current driver. Such driver has feedback loop inside it, which will lower the output voltage, if output current is higher then preset current .. and increase the voltage, if it's lower.

But unless the diode is near it's runaway state, any preset current will be associated with specific voltage. For example 445nm diodes, if you want about 1W of power, you need about 1A of current, which is usually about 4.5V of voltage. So it's all interconnected.

The driver has to be constant current. Has to be able to generate specific current .. but in specific voltage range. You can have constant current driver, which can generate 1A, but it's output voltage could go only up to it's source voltage. And you want to use Lion battery with 3.7V. Such driver wont be able to generate 4.5V needed to get 1A from 445nm diode, thus wont be able to feed the diode. But it could be used on red diode for example, as their working voltage is lower at the same current.

You need driver which meets both current and voltage requirements, with given source. Also the power should be considered, but that depends on voltage and current, so if some driver is designed to given current and voltage, it's also designed to respective power. This could imply colling policy might be necessary though. That should be part of driver design .. like for this and this power you have to use heat-sink this and this big.

As you have to test both voltage and current .. for testing and setting the driver you will need what's called test-load. It's simple circuit which simulates the specific diode reaction to voltage. It consists of diodes and resistors, but it's different for each diode type. Setting the driver with resistor as test load, or different diode (like your LEDs) wont work. It could work if the LED was same color as intended laser diode, and could take same current, as laser diodes are basically LED with specific construction. But it's unlikely that such LED would exist if you are building high power laser.

 

[nztdm]

Generally it works like this:

You apply voltage to a component .. and based on it's inner resistance, some current will go through it.
Many components do not react in linear way to voltage, diodes are one of them. Anyway for working ranges of laser diodes, the reaction is almost linear. You increase the voltage, the current will increase in linear fashion .. and so will the power.
There is usually some minimum current - called lasing treshold. There is usually some maximum voltage, and maximum current, neither of which should be exceeded. There is usually also some maximum temperature, which should not be exceeded.
Then there is some 'sweet spot' .. current at which the diode works best.

Now why you need constant current driver and how it works ? Diode changes it's 'inner resistance' based on temperature. As it gets hot, the inner resistance goes down. If you applied constant voltage to it (which for example common AC/DC converter do), the more and more current would go through the diode .. increasing temperature, decreasing resistance and increasing current again .. up to the point the diode is killed. This is called thermal runaway.

Also diode output depends on current, not voltage. So to get safe and constant output, you need constant current driver. Such driver has feedback loop inside it, which will lower the output voltage, if output current is higher then preset current .. and increase the voltage, if it's lower.

But unless the diode is near it's runaway state, any preset current will be associated with specific voltage. For example 445nm diodes, if you want about 1W of power, you need about 1A of current, which is usually about 4.5V of voltage. So it's all interconnected.

The driver has to be constant current. Has to be able to generate specific current .. but in specific voltage range. You can have constant current driver, which can generate 1A, but it's output voltage could go only up to it's source voltage. And you want to use Lion battery with 3.7V. Such driver wont be able to generate 4.5V needed to get 1A from 445nm diode, thus wont be able to feed the diode. But it could be used on red diode for example, as their working voltage is lower at the same current.

You need driver which meets both current and voltage requirements, with given source. Also the power should be considered, but that depends on voltage and current, so if some driver is designed to given current and voltage, it's also designed to respective power. This could imply colling policy might be necessary though. That should be part of driver design .. like for this and this power you have to use heat-sink this and this big.

As you have to test both voltage and current .. for testing and setting the driver you will need what's called test-load. It's simple circuit which simulates the specific diode reaction to voltage. It consists of diodes and resistors, but it's different for each diode type. Setting the driver with resistor as test load, or different diode (like your LEDs) wont work. It could work if the LED was same color as intended laser diode, and could take same current, as laser diodes are basically LED with specific construction. But it's unlikely that such LED would exist if you are building high power laser.

Could you direct me to a test load circuit for the M140 445nm diode?

My driver outputs 260mA with a 5ohm resistor.
What value resistor would I use to get 1.5A output. The LM317T can handle 1.5A as it is rated for that and I have placed a huge TO3 heatsink with fan on it.

Also, if I connect a 12V lead-acid battery to the driver, I get 12V at the output. Therefore, if I connect 8.4V from 2 fully charged Li-ions, I would get 8.4V output. I thought this voltage would kill the diode, even with limited current. Hasn't the diode got a specific voltage that it is designed for?
And you said about using the test-load to set the desired voltage and current. I set the current by changing resistor, but how do I set the voltage. I just get out what I put in.

Btw, when I measure the voltage output, it is without a load. Is that right?

Thanks
JD

 

[DrSid]

M140 works around 4.5V.

You can find the test load mentioned here for example:

http://laserpointerforums.com/f65/2w-445nm-first-build-finsihed-75686.html

Simply go to 445nm section (445nm Lasers - Laser Pointer Forums - Discuss Lasers & Laser Pointers) and use search function with 'test load'.

For voltage/current graph, see this thread:

http://laserpointerforums.com/f65/3x-445nm-ld-piv-plot-53927.html

It's the read line with scale on the right side. Both diode and test load should react the same way.

 

[nztdm]

M140 works around 4.5V.

You can find the test load mentioned here for example:

http://laserpointerforums.com/f65/2w-445nm-first-build-finsihed-75686.html

Simply go to 445nm section (445nm Lasers - Laser Pointer Forums - Discuss Lasers & Laser Pointers) and use search function with 'test load'.

For voltage/current graph, see this thread:

http://laserpointerforums.com/f65/3x-445nm-ld-piv-plot-53927.html

It's the read line with scale on the right side. Both diode and test load should react the same way.

Im still confused how I would get 4.5V from my LM317T driver circuit. It just puts out the voltage I put in, with current limiting.

 

[DrSid]

Only way it can limit current is to lower the voltage. AFAIK this driver cannot increase its voltage over its input voltage. So if you input less then 4.5V it simply won't work.

 

[nztdm]

Only way it can limit current is to lower the voltage. AFAIK this driver cannot increase its voltage over its input voltage. So if you input less then 4.5V it simply won't work.

Whenever I measure voltage across output of the driver with multimeter, it reads a little below the input voltage. What voltage is it meant to read if the driver was made correctly?

 

[Bigbondo007]

My driver seems very weak..little or no output, so I built another and then another. Same thing so I changed the pot style and still no power increase. I got irritated and decided to short the pot terminals since I had no poweR anyhow. when I jumped the pins my test led turned into a blinding light that left me seeing stars for a bit. somehow after soldering all the pins together I miraculously got an adjustment from the pot whereas I had none before. What is happening with my drivrr and how can I have an adjustment when all the pins are jumped?

 

[DrSid]

Whenever I measure voltage across output of the driver with multimeter, it reads a little below the input voltage. What voltage is it meant to read if the driver was made correctly?

It depends on the load. If there is no load, it depends no driver .. unfortunately I don't have experience with your driver.

 

[nztdm]

It depends on the load. If there is no load, it depends no driver .. unfortunately I don't have experience with your driver.

This is where I got my driver schematic from. It is built exactly.
zero fidelity CNC: Laser diode driver circuit

If I connect the test-load you mentioned earlier (4x1N5404 and 1R0 resistor) and measure across the resistor, will that give me the proper output voltage?
Do I want that voltage across the resistor to be ~4.5V? (M140)

 

[Bigbondo007]

My driver is the same as well.

 

[Bigbondo007]

Think i found my problem. I looked at my resistor package fr rom radio shack which very clearly says ten ohms and cross ref'd the banding on the resistors......10,000 ohm packaged in a ten ohm sheet. Should have checked first huh

 

[billg519]

A current regulator will regulate the current to the load, allowing the voltage to rise only to a level that allows the set current to be drawn. The voltage will rise no higher. If you measure the voltage output of a current regulator that has no load on it, the voltage measured will be nearly the input voltage. As the set current is not being drawn, the current regulator has raised the voltage to maximum in order to try to get a current flowing. An unloaded current regulator will always show an output voltage similar to the voltage input. Hook up a test load of a one ohm resistor and several 1N4007 diodes in series. With this hooked up to the output of the current regulator, measure the output voltage across this test load. It will be less than the input voltage. The current drawn can be verified by measuring the voltage drop across the one ohm resistor in the test load. Your laser diode will function similar to the test load. The regulator will only provide the voltage required to cause the diode to draw the set current. The rest will be converted to heat bay the regulator. Therefore, try to use an input voltage no higher than necessary. For the LM series, usually allow 3V plus the Vf of the LD in question.