Home made LM-317 Driver

[jupiter8]

Eudaimonium, thank you. That is common sense talking. I recognise it, I've seen it before. Logically sensible. I will take your advise and try it. I like the idea of a resistor selector switch with various values on the out pin.

 

[Eudaimonium]

Eudaimonium, thank you. That is common sense talking. I recognise it, I've seen it before. Logically sensible. I will take your advise and try it. I like the idea of a resistor selector switch with various values on the out pin.

Sure umm, I'd also like to apologise for 'harsh' manner in which my posts were written, the one with pics especially.
I'm glad we understood eachother

Thanks for +rep Jerry, my rep points made enormus jump , over 2k now :beer:

I've seen a lot of selector slide switches, some with more than 10 positions.
You can commonly find them in some AC DC adaptors which have output voltage selectors on them. I have one of those, can even select output polarity on it.

 

[Autist]

@ Eudaimonium,

I am researching drivers for 9mm IR 1 watt and 2 watt diodes, may I ask, how hot does your regulator get after 2 or 3 minutes ?

Also, what voltage is your battery supplying ?

Lastly, forgive my ignorance about electronic components, I thought that the LM 317 was a voltage regulator. This is very puzzling because I have read some say it regulates voltage, and I have seen it written that it is possible to trick it into being a current regulator...
I have also seen circuits that take the output from the ADJ pin with a resistor to the OUT pin, and the other way round too.

I have been using the free software for circuit designing (called SPICE), and (unfortunately, there is no laser diode in their "library" of components, so an equivalent network of diodes has to do) it seems to make an enormous difference from which pin the output is taken, but absolutely no difference whether the sense resistor is 5 ohms or 1 ohm...

This may be a bug in the software though.

best wishes

David

 

[Eudaimonium]

@ Eudaimonium,

I am researching drivers for 9mm IR 1 watt and 2 watt diodes, may I ask, how hot does your regulator get after 2 or 3 minutes ?

Also, what voltage is your battery supplying ?

Lastly, forgive my ignorance about electronic components, I thought that the LM 317 was a voltage regulator. This is very puzzling because I have read some say it regulates voltage, and I have seen it written that it is possible to trick it into being a current regulator...
I have also seen circuits that take the output from the ADJ pin with a resistor to the OUT pin, and the other way round too.

I have been using the free software for circuit designing (called SPICE), and (unfortunately, there is no laser diode in their "library" of components, so an equivalent network of diodes has to do) it seems to make an enormous difference from which pin the output is taken, but absolutely no difference whether the sense resistor is 5 ohms or 1 ohm...

This may be a bug in the software though.

best wishes

David

Hi, well first off, "ignorance" depends on point of view.
You do not know, and you are asking about it. That's good. Ignorant man ignores his ignorance, IMHO.

Next, well, LM317 is called a voltage regulator, because it is it's primary use, but this is how it works:
It'll supply some voltage reference between it's pins OUT and ADJ, and according of resistance it will supply the constant current.

Now, let's say that you use a constant current of 1A to power nothing more than one 5 ohm resistor.
Now, you have 1 Amp, and 5 ohms, do yo know ohms law?
It says :
Voltage = Current x Resistance (this formula nad many it's derivations are widely used).
Now, that means you have exactly the 5 Volts of potential difference on pins of the resistor.

That means that , whatever you hook up in parallel with that resistor, it will be powered by constant voltage of 5V (ideal for example, IC and MCUs which work on 5V).

Now , what we do is not use the final resistor at all, but use the constant current to power out precious laser diodes (speaking of which, your avatar is depressing ... )

Next, 1W 808nm diodes require 1.25 A, which is achievable with LM317 and 1 ohm resistor as current setter.

However, maximum rating for LM317 in TO220 case is 1.5 A.
2W diodes would require double than 1W, and hence, LM317 is out of picture.

If you need more power (one always needs more power! ), you can use the LM350. It is used identically as LM317, same Vref and everything, except it can supply up to 3 A!

The battery pictured few posts up is my "testing" 18650 cell, it's naked - I stripped it off it's regulation circuit and heatshrink tubing, and it's entire surface acts as negative connection except the positive pole, obviously.
I can use to to test green modules or such drivers and stuff, because of large surface of negative connection it's easier to use with springs and alligator clips maybe, then normal batteries.

In any way, lithium ion cells such as this one have a surface charge of 4.2 V when fully charged, and they discharge until 2.8 V.
I always keep it at full charge to power various stuff, like 1W 9mm diode seen in pictures (forgive very sloppy setup, no module or focusing, I soldered it real quick so I can take pics and present my point that driver OP posted is a wayste of components, rudely put).

I think that answers all your questions for now, if you have any more questions about anything at all, do not be afraid to shoot me a PM (personal message) and ask away! I have a lot of free time on my hands

EDIT = Oh, how hot it gets? Considering it's underpowered by therms of voltage (4V is not enough, 2.5 V drop of driver and 2 V diode voltage, I'm 0.5 V short), I can' give accurate anwer, but let's say that you supply exactly the needed voltage at 1A, total power dissipation is around 2.5W, add more watts with more volts supplied.
Driver would produce 2.5 W, and diode around 1W Ironic, driver needs more cooling than the diode.

 

[Autist]

Hi, well first off, "ignorance" depends on point of view.
You do not know, and you are asking about it. That's good. Ignorant man ignores his ignorance, IMHO.

Next, well, LM317 is called a voltage regulator, because it is it's primary use, but this is how it works:
It'll supply some voltage reference between it's pins OUT and ADJ, and according of resistance it will supply the constant current.

Now, let's say that you use a constant current of 1A to power nothing more than one 5 ohm resistor.
Now, you have 1 Amp, and 5 ohms, do yo know ohms law?
It says :
Voltage = Current x Resistance (this formula nad many it's derivations are widely used).
Now, that means you have exactly the 5 Volts of potential difference on pins of the resistor.

That means that , whatever you hook up in parallel with that resistor, it will be powered by constant voltage of 5V (ideal for example, IC and MCUs which work on 5V).

Now , what we do is not use the final resistor at all, but use the constant current to power out precious laser diodes (speaking of which, your avatar is depressing ... )

Next, 1W 808nm diodes require 1.25 A, which is achievable with LM317 and 1 ohm resistor as current setter.

However, maximum rating for LM317 in TO220 case is 1.5 A.
2W diodes would require double than 1W, and hence, LM317 is out of picture.

If you need more power (one always needs more power! ), you can use the LM350. It is used identically as LM317, same Vref and everything, except it can supply up to 3 A!

The battery pictured few posts up is my "testing" 18650 cell, it's naked - I stripped it off it's regulation circuit and heatshrink tubing, and it's entire surface acts as negative connection except the positive pole, obviously.
I can use to to test green modules or such drivers and stuff, because of large surface of negative connection it's easier to use with springs and alligator clips maybe, then normal batteries.

In any way, lithium ion cells such as this one have a surface charge of 4.2 V when fully charged, and they discharge until 2.8 V.
I always keep it at full charge to power various stuff, like 1W 9mm diode seen in pictures (forgive very sloppy setup, no module or focusing, I soldered it real quick so I can take pics and present my point that driver OP posted is a wayste of components, rudely put).

I think that answers all your questions for now, if you have any more questions about anything at all, do not be afraid to shoot me a PM (personal message) and ask away! I have a lot of free time on my hands

EDIT = Oh, how hot it gets? Considering it's underpowered by therms of voltage (4V is not enough, 2.5 V drop of driver and 2 V diode voltage, I'm 0.5 V short), I can' give accurate anwer, but let's say that you supply exactly the needed voltage at 1A, total power dissipation is around 2.5W, add more watts with more volts supplied.
Driver would produce 2.5 W, and diode around 1W Ironic, driver needs more cooling than the diode.

Wow, :bowdown:

Thanks Eudaimonium

I needed that info. :thanks:

Interestingly, when I was using a DrLava MicroFlex at around 900 mA, I had a 1N5392 diode in series with the 9mm IR and the LD was dropping 1.88v, all in a pocket mini, thanks JayRob.


With regard to my avatar, whilst experimenting, I have had a few 'accidents', but two of those were brand new GGW H20 6X Blu-Ray diodes that could not handle 187 mA.

Thanks again for your help. I mentally give you REP+1, but cannot physically, sorry.

Very best wishes

David

 

[Eudaimonium]

Wow, :bowdown:

Thanks Eudaimonium

I needed that info. :thanks:

Interestingly, when I was using a DrLava MicroFlex at around 900 mA, I had a 1N5392 diode in series with the 9mm IR and the LD was dropping 1.88v, all in a pocket mini, thanks JayRob.


With regard to my avatar, whilst experimenting, I have had a few 'accidents', but two of those were brand new GGW H20 6X Blu-Ray diodes that could not handle 187 mA.

Thanks again for your help. I mentally give you REP+1, but cannot physically, sorry.

Very best wishes

David

Only 1.8 V with diode in line? Hmm, interesting...

Sorry to hear about your GGWs, I remember excitement when I built my 6x, I would've killed myself If I destroyed the diode. My diode lives at 190mA, dunno why would your die at almost same current, given no other accidents happened,
But don't confuse two variants of 6x,
GGW diodes can take a good amount of current, like 200mA,
But, SF-AW210 are good to up maybe 170mA MAXIMUM, any higher will degrade it very much and it will die soon.

Don't worry about +rep, styropyro got you covered

 

[Bionic-Badger]

Interestingly, when I was using a DrLava MicroFlex at around 900 mA, I had a 1N5392 diode in series with the 9mm IR and the LD was dropping 1.88v, all in a pocket mini, thanks JayRob.

Why would you put a diode in line with your laser diode? The Flex driver adjusts the voltage as necessary to cause the laser diode to pass current. The diode doesn't do anything more than cause the driver to overwork itself.

The rectifier diode that can be used with the driver is supposed to be connected to the battery end of the driver, not the laser diode end. It's there to reduce the input voltage that the driver receives because it can only operate at below 5.5V.

 

[Autist]

Why would you put a diode in line with your laser diode? The Flex driver adjusts the voltage as necessary to cause the laser diode to pass current. The diode doesn't do anything more than cause the driver to overwork itself.

The rectifier diode that can be used with the driver is supposed to be connected to the battery end of the driver, not the laser diode end. It's there to reduce the input voltage that the driver receives because it can only operate at below 5.5V.

Hi BB,

I was unsure how much voltage and current the 9mm IR 1 watt diode required, so I had communications with DrLava with regard to using his MicroFlex drive. Using a CR123A rechargeable battery at around 4 volts, DrLava wrote that the minimum voltage the MicroFlex drive output was 2 volts, which may be too high, so to put a Schottky diode in series with the LD. The 1N5392 diode was the highest capacity diode I had lying around.
The advice seemed good to me, because I know little about electronics.

Best wishes

David

 

[Eudaimonium]

Hi BB,

I was unsure how much voltage and current the 9mm IR 1 watt diode required, so I had communications with DrLava with regard to using his MicroFlex drive. Using a CR123A rechargeable battery at around 4 volts, DrLava wrote that the minimum voltage the MicroFlex drive output was 2 volts, which may be too high, so to put a Schottky diode in series with the LD. The 1N5392 diode was the highest capacity diode I had lying around.
The advice seemed good to me, because I know little about electronics.

Best wishes

David

What kind of IR diode exactly are we talking about?

Because 808nm diodes, commonly used and found, do have 1.9V - 2.2 V (depends on power and current and temperature) of normal Vf.

I never heard of any lower forward-voltage laser diodes, though just because I haven't does not mean there aren't any.

 

[Autist]

What kind of IR diode exactly are we talking about?

Because 808nm diodes, commonly used and found, do have 1.9V - 2.2 V (depends on power and current and temperature) of normal Vf.

I never heard of any lower forward-voltage laser diodes, though just because I haven't does not mean there aren't any.

I do not know.
I bought two 2 watt and two 1 watt 9mm diodes from Milos. He had bought them from Chris Luebner (?) at Heruurscience.
Milos had no details on the diodes, so I PM'd Chris, but he did not reply.

When attempting to ascertain the Ith with a circuit I borrowed from a 85 mA red LD, the two 2 watt diodes drew 380 mA at 1.66 volts, and the two 1 watt diodes drew 360 mA at the same 1.66 volts.
The supply was just two AA nimh cells and a potentiometer turned down to the minimum, 0.5 ohms. The open circuit voltage was 2.55 volts.

Three of the diodes had one pin connected to the case, whereas the last diode seemingly had both pins case isolated.

I am sorry I cannot be of more help.

Best wishes

David

 

[Eudaimonium]

I do not know.
I bought two 2 watt and two 1 watt 9mm diodes from Milos. He had bought them from Chris Luebner (?) at Heruurscience.
Milos had no details on the diodes, so I PM'd Chris, but he did not reply.

When attempting to ascertain the Ith with a circuit I borrowed from a 85 mA red LD, the two 2 watt diodes drew 380 mA at 1.66 volts, and the two 1 watt diodes drew 360 mA at the same 1.66 volts.
The supply was just two AA nimh cells and a potentiometer turned down to the minimum, 0.5 ohms. The open circuit voltage was 2.55 volts.

Three of the diodes had one pin connected to the case, whereas the last diode seemingly had both pins case isolated.

I am sorry I cannot be of more help.

Best wishes

David

They draw 380mA at 1.6 V , but at 1.9V they'll draw an amp or more, regulation needed at 1.2 for 1W diodes, 2.2A for 2W diodes.

Would sound like my 9mm diodes, all of mine have two pins, one case connected one isolated (exception is 1000+nm diodes, which have three pins, but in general pinout identical to 3pin 5.6mm 808nm diodes),
the one with both pins isolated , can you take a picture of it?

Also the front of it, and tell us what color is the window in the front?
Typical 808nm IR diodes have greenish tint, one I mentioned 1000+nm (don't know exactly), has brown tint.

 

[Autist]

They draw 380mA at 1.6 V , but at 1.9V they'll draw an amp or more,

regulation needed at 1.2 for 1W diodes, 2.2A for 2W diodes.

Would sound like my 9mm diodes, all of mine have two pins, one case connected one isolated (exception is 1000+nm diodes, which have three pins, but in general pinout identical to 3pin 5.6mm 808nm diodes),

the one with both pins isolated , can you take a picture of it?

Also the front of it, and tell us what color is the window in the front?
Typical 808nm IR diodes have greenish tint, one I mentioned 1000+nm (don't know exactly), has brown tint.

This is something I am not sure about. If the efficiency of the diode was approaching 33% (unlikely) then to get 2 watts of light out, it would need 6 watts of eletrical energy into it. That would be around 3 amps at 2 volts.
Similarly, for the 1 watt diode, to get 1 watt of light output, then 3 watts would be needed to be inputted, no ??

I will try, tomorrow.

best wishes

David

 

[Eudaimonium]

This is something I am not sure about. If the efficiency of the diode was approaching 33% (unlikely) then to get 2 watts of light out, it would need 6 watts of eletrical energy into it. That would be around 3 amps at 2 volts.
Similarly, for the 1 watt diode, to get 1 watt of light output, then 3 watts would be needed to be inputted, no ??

I will try, tomorrow.

best wishes

David

Actually no, efficiency of 808nm emitters is around 40-45 %, hence only 1.2 Amps needed to drive the 1W diode to full power.

We'll need PBD to confirm this though...

 

[Autist]

They draw 380mA at 1.6 V , but at 1.9V they'll draw an amp or more, regulation needed at 1.2 for 1W diodes, 2.2A for 2W diodes.

Would sound like my 9mm diodes, all of mine have two pins, one case connected one isolated (exception is 1000+nm diodes, which have three pins,

but in general pinout identical to 3pin 5.6mm 808nm diodes),
the one with both pins isolated , can you take a picture of it?

Also the front of it, and tell us what color is the window in the front?
Typical 808nm IR diodes have greenish tint, one I mentioned 1000+nm (don't know exactly), has brown tint.

Hi Eudaimonium,

At last, I've managed to upload them. The first image shows a slight blue tint, the second without a reflection, and the third the rear, obviously.

I did not know that the efficiency of IR ld was that high. Sorry.

Best wishes

David

 

[Eudaimonium]

Hi Eudaimonium,

At last, I've managed to upload them. The first image shows a slight blue tint, the second without a reflection, and the third the rear, obviously.

I did not know that the efficiency of IR ld was that high. Sorry.

Best wishes

David

Slight bluish tint?
Red little marker dot near the pins?

I'm sure it's a 635/641 nm red diode, power it up with 300-400mA and see for yourself!

 

[Autist]

Slight bluish tint?

Red little marker dot near the pins?

I'm sure it's a 635/641 nm red diode, power it up with 300-400mA and see for yourself!

Yes, all four 9mm IR diodes I have have a bluish tint on the window. I think it is an AR coating.

That red dot was for me to remember which pin was positive. I have run it at 360 mA, but only quickly. Just to see if it worked. :yh:
It got a little warm not in a module or heatsink.

best regards

David