Current Limiting with Resistor only...?

I am very aware of the value of using constant current sources with my diodes, having started off with the venerable Daedal driver several years ago. But my question here is:

If I want to run a simple test setup with a diode from a battery, say for the purposes of the math, a pair of 1/2AAs rated 1100mAh, for a total of 7.2vdc (we'll say 7.4vdc if they're a little over rated voltage as they often are when brand new).

If I put a series resistor between them and a diode that is rated for- say, 1.25A, but my resistor, in combination with the Vf of the diode, is set so that at the battery's MAXIMUM charge, could never pass, say, more than 500mA to the diode - what are the risks I'm not taking into account with this setup? Am I likely to in any way, barring simple carelessness, to pop my diode in this manner?

Thanks for your comments.

Dave

You CAN get a capacitive buildup on battery terminals that spikes a current through the diode on initial connection, but here it is not a well document phenomenon. Main reason is thermal runaway. If you allow the diode temp to increase, its Vf changes to allow more or less current through, and temp to increase further. It can away from you.

The resistor's v drop and current limit setting is not solely dependent on the battery's V, but also the diode's Vf. The resistor drops whatever the difference is between the two. If the diode heats, its Vf causes that difference to increase. The resistor's value doesn't change, so current flow simply changes.

Its essentially controlling the current by applying a temperature dependent voltage across its terminals. Not an easily predictable or consistent method of current control.

Hopefully that made some sense...

Essencially you can use a resistor for current limiting
a Laser Diode if the Laser Diode is not running HOT
the Voltage source is Locked at a predetermined voltage
that will not vary at all.

Then there is the spike problem that Meatball referred to..


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

That makes sense. This is for an application where thermal runaway is very unlikely due to the momentary nature of the operation - seldom for more than 3 to 5 seconds in an already heat-sunk housing. As for capacitive buildup on the terminals, if during the test setup, I have a resistor of some reasonable value (perhaps 10k) that is always across the battery terminals, and voltage is only applied after I hit the test button - that should prevent capacitive spiking...

I could always use the old LM317 driver setup, but I was mainly curious if there were any other significant considerations for such a setup. I'm planning to have the current delivered always be roughly 50% of the diode's rated operation current.

If there are any other thoughts that may not have been covered here - I'm still about a week away from actually doing this

Dave

lasersbee said:

Essencially you can use a resistor for current limiting
a Laser Diode if the Laser Diode is not running HOT
the Voltage source is Locked at a predetermined voltage
that will not vary at all.

Then there is the spike problem that Meatball referred to..


Jerry

Click to expand...

Thanks Jerry, you're much better at explaining it than I!

xanatos said:

That makes sense. This is for an application where thermal runaway is very unlikely due to the momentary nature of the operation - seldom for more than 3 to 5 seconds in an already heat-sunk housing. As for capacitive buildup on the terminals, if during the test setup, I have a resistor of some reasonable value (perhaps 10k) that is always across the battery terminals, and voltage is only applied after I hit the test button - that should prevent capacitive spiking...

I could always use the old LM317 driver setup, but I was mainly curious if there were any other significant considerations for such a setup. I'm planning to have the current delivered always be roughly 50% of the diode's rated operation current.

If there are any other thoughts that may not have been covered here - I'm still about a week away from actually doing this

Dave

Click to expand...

I think your bases may be covered. I would personally never do it for any current >500mA, but that's just for your consideration. The lm317 driver is so cheap, there is almost no good reason not to use one.

I would have to disagree.

If your diode is heaksinked well,

then the resistor has some advantages.

One, it is more reliable.

Two, the resistor will have less voltage drop, as the battery voltage declines.

I am not aware of any capacitive "build up" on the cells I use.

Here is a thread on my Direct Drive 445nm laser:

http://laserpointerforums.com/f65/direct-drive-445-host-52918.html

LarryDFW

Meatball said:

Thanks Jerry, you're much better at explaining it than I!

Click to expand...

I don't know about that.... I clearly understood what you posted
and I basically agreed with your opinion...


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

Meatball said:

...The lm317 driver is so cheap, there is almost no good reason not to use one.

Click to expand...

One reason: Space. That's it. For a variety of reasons, I have a number of things that need to get stuffed into a tiny space - this is one of them. The space difference between the LM317 driver vs. a resistor actually can make the difference of including a function or not. Otherwise - I'd be using a driver without question.

I'd use a smaller driver, like a FlexDrive, but I also need to have the current be potentially variable, and I don't want to have to unsolder surface mount resistors, etc., and replace them with a switchable resistor set. I'm looking into some drivers that can go to 1.5+ Amps, but which have the ability to be modulated and/or "dimmed".... still in planning stages.

Thanks,

Dave

As far as I do know the main reason for not using a simple resistor (like with small LEDs) is that a resistor is a linear (R=U/I) element but a laser diode is a highly non-linear element.

A tiny change in voltage across the LD significantly changes the current. For small LDs an increase of 0.1V may be already be too much...

To get a somehow stable setup a huge voltage drop across the resistor is required (e.g. > 10V). A LED is not as critical as a LD and doesn't suffer if current is too high for a fraction of a second

And finally, without a fast current regulation any spike (static discharge etc.) on the supply may kill your expensive LD and make a LED out of it.

Would recommend to spend some bucks for a small switched regulator.

-mo-

Voltage spike issue can be solved with a capacitor in parallel with the laser diode. This will also give the effect of a slight soft-start, depending on the value of the capacitor used.

This has a disadvantage in that if the LD becomes temporarily disconnected from the capacitor, the capacitor will charge to a higher voltage that will likely kill the laser diode when the connection is made again. (this really shouldn't be thought of as a major downside though, most drivers will exhibit this behavior and some switchmode drivers may actually be destroyed if the load is disconnected during operation)

Since a resistor does not do any active regulation, your laser loses power as the batteries discharge.

This is not a well known fact but the aixiz boost driver is a voltage source,
with a series resistor. not a current source.

using a resistor can work, as long as your calculations are accurate

The surface charge thing is real, and all batteries have it.
it just means that the batteries internal resistance doesnt come into play,
until that surface charge is depleted.