I am trying to use the LM317L to power (2) 650nm 5mW laser diodes from a battery source. Ideally I want this circuit to be as small and compact as possible with the laser able to stay on continuously for an hour or so. So far I'm just trying to get (1) laser diode to work before trying the second. My problem is that no matter what type of battery I use, 9v, coin cells, etc. it seems the battery just can't cut it. After about 5-10 mins the laser is too dim for practical purposes. What can I do to keep the battery from draining so fast?

 

5mW refers to the light output; not the DC input power to the Laser diode...

What is the Vf of the laser diode when operating? What is the operating current?

An LM317 will be wasting about half the power coming from the battery...

 

I don't know what the Vf is off hand, I will have to measure it. The spec says <40mA for operating current but that's a large range. What would you recommend using other than the LM317 for better battery performance?

 

When using an LM317 as a constant-current source, the minimum voltage drop across it is Vdo + Vref, which is about 3V + 1.25V = 4.25V. This means that the minimum battery voltage must be 4.25V above the Vf of the laser diode. This app note says the typical forward voltage for a laser diode is ~1.5V, so with 4.5V across the driver, and only 1.5V across the laser, you have a very inefficient system...

It is possible to build a current regulator that has a minimum differential voltage of about 100mV using a voltage reference, an opamp, a PFet, and a couple of resistors...

 

Hmm I will definitely have to try that out. I see what you are saying about the inefficiency of the LM317. Is there a reference schematic you can point me to that uses what you described?

 

For best efficiency you can use a switching buck current regulator such as this.
That way there's no power loss in a resistor in series with the laser LED, since a resistor is not needed (but the two laser diodes must be run in series).
Ebay may also have LED switching modules that could work for you.

 

I am trying to use the LM317L to power (2) 650nm 5mW laser diodes from a battery source. Ideally I want this circuit to be as small and compact as possible with the laser able to stay on continuously for an hour or so. So far I'm just trying to get (1) laser diode to work before trying the second. My problem is that no matter what type of battery I use, 9v, coin cells, etc. it seems the battery just can't cut it. After about 5-10 mins the laser is too dim for practical purposes. What can I do to keep the battery from draining so fast?

The transistor radio type of 9V batteries are the least energy density for the most money. The coin cell type were never intended for that sort of current drain either.

AA cells have better energy density for a given size (and cost) - you can also select a holder with the right number of cells instead of starting with too high voltage and wasting lots of power in a regulator.

The best efficiency is to be had by switching to a buck type SMPSU regulator.

 

I'm trying to make this as light weight as possible so AA cells aren't exactly ideal. Any suggestions as for other battery types? Aren't lithium batteries better for electronics?

 

I'm trying to make this as light weight as possible so AA cells aren't exactly ideal. Any suggestions as for other battery types? Aren't lithium batteries better for electronics?

Non-rechargeable lithium coin cells are less than ideal for the current draw you want.

Rechargeable lithium cells might do - the ones in some of the entry level E-cigarettes (the ones that look like cigarettes) weigh in at somewhere around 180mAh. You also get a free charger if you strip an E-cigarette for the parts.

Here comes the usual safety warning: Lithium cells can vent with flaming gas and/or explode if not charged correctly.

Some cheap import E-cigarettes have caused fires without having been got at - so work carefully and don't make any mistakes!

 

Yes, lithium batteries have the highest energy density of the common battery types, but they are also the most expensive.
How about AAA cells? The alkaline types generally have a little over 1Ah capacity.
At 50mA load current that would give an operating time of about 20 hours using a linear regulator or a resistor to control the diode current. You could get more if you used a switching regulator.

 

Yes, lithium batteries have the highest energy density of the common battery types, but they are also the most expensive.
How about AAA cells? The alkaline types generally have a little over 1Ah capacity.

About a year or 2 ago, a local shop had solar rechargeable keyfobs real cheap, so I bought a few.

When I opened one up, the solar panel wasn't connected to anything and the solder pads defied any attempt to solder them.

The 2032 cell might actually be rechargeable, but I sort of lost interest in messing about with them.

 

Hmm I will definitely have to try that out. I see what you are saying about the inefficiency of the LM317. Is there a reference schematic you can point me to that uses what you described?

Does the laser diode have to operate with the cathode grounded (heat sink)?

 

No

 

Hmm I will definitely have to try that out. I see what you are saying about the inefficiency of the LM317. Is there a reference schematic you can point me to that uses what you described?

Here is a link to lots of info...

 

You may refer to the pwm circuit as below, it can be adjust the output current, I used 5mA for zener diode and 20 mA for Laser diode, the total current is 25mA, so it won't exceed the laser diode needed.

You can measure the current of Zd1 and Laser diode, and adjust the R2 value to get the properly current and also not exceed the rating current 25 mA of Laser diode, I used 20mA, because it was 80% of 25mA.

You can also using the 555 Astable Circuit Calculator to calculate the frequency.

 

Suggest my version of the laser driver.
I want to warn you about working with a laser.
Exceeding the maximum optical power laser leads
to degradation of its mirrors. As a result,
sharply falling output optical power.
Enough of exceedance within one second,
and after a short time the laser fails.

 

The transistor radio type of 9V batteries are the least energy density for the most money. The coin cell type were never intended for that sort of current drain either.

AA cells have better energy density for a given size (and cost) - you can also select a holder with the right number of cells instead of starting with too high voltage and wasting lots of power in a regulator.

The best efficiency is to be had by switching to a buck type SMPSU regulator.

Why would coin cells not work if I can get cells that are 280mAh, 550mAh, 620mAh etc.?

 

Why would coin cells not work if I can get cells that are 280mAh, 550mAh, 620mAh etc.?

Because coin cells are designed to deliver a small current for a long time; not the other way around...

Listen to Heavy. You need a true current-feedback, constant-current Laser driver; not a pwm constant-voltage source as in post #15.

 

Because coin cells are designed to deliver a small current for a long time; not the other way around...

Listen to Heavy. You need a true current-feedback, constant-current Laser driver; not a pwm constant-voltage source as in post #15.

Then how does a laser pointer work with button cells? Is it because it is supposed to be used in short increments and not continuously?

 

There are the LEDs used in cheap laser pointers, and then there are real Laser Diodes. I thought you were working with the latter...