Using a DC/DC converter to power the laser

[Saturn400K]

Here's a little something that could help. It has a minimum input of 1.1V and can be adujusted to be up to 6.5V by replacing RF2 with a pot. I also heard that the output voltage can even be boosted up to 9V by replacing the cap at C2 with one that can handle more voltage. The best part is that this thing is only about the size of a quarter. I ordered one to see if it will drive my LM317 setup

As usual I cannot post a link but you can search for it at sparkfun.com, It is called a 5V DC to DC Step Up - VPack PCB

 

[IgorT]

Here's a little something that could help. It has a minimum input of 1.1V and can be adujusted to be up to 6.5V by replacing RF2 with a pot. I also heard that the output voltage can even be boosted up to 9V by replacing the cap at C2 with one that can handle more voltage.

I took a look, and it seems it could work.. The max. current is 300mA at 5V. I imagine this would go down if you turned the voltage higher, so it might lack some power..


It would be much better to convert it to current regulation, if possible. That way the output voltage needed would be a little less than 3V and it would be capable of supplying more than enough current..


Do you perhaps know which IC is used on this circuit, so i could try to find the datasheet and see if it can be converted to a current source?

 

[woop]

with getting smps's to do constant current. how can you do it without having an extra chip for amplification of a shunt voltage etc...
i found this schematic from this page http://www.edn.com/article/CA6325592.html
does it look like it actually regulates current or just something dodgy with voltage?
If it is so simple to implement constant current, could it be used with any smps feedback loop? like could you get one of those single AA portable phone chargers and modify it for constant current?

 

[IgorT]

Does it look like it actually regulates current or just something dodgy with voltage?
If it is so simple to implement constant current, could it be used with any smps feedback loop? like could you get one of those single AA portable phone chargers and modify it for constant current?

I read your idea in the 317 thread (the part about "phone charger") and didn't understand at first.. Where i live, i've only seen such phone chargers working off a 9V battery..

Only now do i understand what you're saying.. Do you perhaps have a link to one of these 1x AA phone chargers?
This could provide all the necesary parts very cheap, due to mass production. How much do they cost?

Have you ever measured the output of these chargers? If the phones contain all of the charging circuitry, it's not even necesary for the charger to be regulated.. It could just contain a voltage tripler. I've used many old phone charging AC/DC adaptors, and none of them were regulated, so i'm guessing the phone does all the charging, and it just needs high enough voltage, to allow a certain current into the battery.


The circuit you posted should be a current source, since it's meant for driving LEDs.. Unfortunatelly i don't know enough about it, but i can always ask my electronics engineer, if it is and if not, how to change it..


Thanks for the link! I'm gonna forward it.

 

[IgorT]

Woop..

It says: "Switching regulator efficiently controls white-LED current"
and: "the circuit efficiently provides constant-current drive to a high-current LED and minimizes the effects of supply-voltage and temperature variations on the LED's brightness"

So i'm guessing it should.. This could be the perfect solution!


Nice find!

 

[chimo]

At first glance, it does not look like a constant current circuit.  I will take another look later.

EDIT: I read some of the circuit notes from the other thread - this appears to use the diode to compensate for teh temperature drift of the luxeon. That requires the diode to be mounted in the heat path of teh luxeon - not sure if you could accomplish that effectively with a Aixiz module.

with getting smps's to do constant current. how can you do it without having an extra chip for amplification of a shunt voltage etc...
i found this schematic from this page http://www.edn.com/article/CA6325592.html
does it look like it actually regulates current or just something dodgy with voltage?
If it is so simple to implement constant current, could it be used with any smps feedback loop? like could you get one of those single AA portable phone chargers and modify it for constant current?

 

[IgorT]

EDIT: I read some of the circuit notes from the other thread - this appears to use the diode to compensate for teh temperature drift of the luxeon.  That requires the diode to be mounted in the heat path of teh luxeon - not sure if you could accomplish that effectively with a Aixiz module.

So it was easyer to do this trick, than to convert it into a current source? Damn..

 

[woop]

well i bought the phone charger off ebay for 10$Australian
http://search.ebay.com.au/search/search.dll?from=R40&_trksid=m37&satitle=aa+phone+charger
inside them they have a little boost mode SMPS, there is one controling chip, a fet, inductor some caps and resistors, pretty standard boost topology.
its rated for 320mA but the voltage drifts a bit with an alcaline battery and it can supply up to 500mA but with the voltage really drifting, like to 3V. so it is not current limited. however it might ok just to run an LD with a resistor and a cap for protection. or if anyone knows of any low dropout (linear) current limiters?
by the way, it can half charge a phone from a alkaline battery, the battery gets pretty hot though, and when the battery drops below 1.1V the output starts to get really shaky.
I think it would preform a lot better with a lithium 1.5V AA cell

so any electronic engineers here?
can someone explain how that smps i posted before limits current? i can't make sense of it, although i haven't had time to read through all the text yet.

 

[IgorT]

so any electronic engineers here?
can someone explain how that smps i posted before limits current? i can't make sense of it, although i haven't had time to read through all the text yet.

Chimo is an EE, and he allready looked into it..
This particular circuit seems to be good only for driving that specific LED..


Too bad to hear about those phone chargers.. I was afraid it wouldn't be regulated too much, since the phone takes care of the charging process itself..

But if you put an AMC after it, it could probably work just fine.. It has a very low dropout voltage.

 

[chimo]

Here's the meat of how it works (note my added bold and italic emphasis):

[qoute]In this circuit, the LM2852 operates at efficiency of approximately 93% and directly controls a step-down-regulator topology that maintains a constant current flow through LED1, which potentiometer R1 adjusts. Current-to-voltage conversion taking place within the circuit's control loop effectively regulates the circuit's output current. In operation, the LM2852 compares its internal reference voltage with the voltage from the divider formed by D1, R1, and R2 and drives the control loop to maintain a constant 1.2V at its voltage-sense pin. Current through the voltage divider is proportional to the current through LED1, and the ratio of the currents tracks over the circuit's operating-temperature range because D1 and LED1 exhibit approximately the same forward-voltage temperature coefficient of –2 mV/°C. Mounting D1 and LED1 next to each other on the pc board provides sufficiently close thermal coupling for temperature compensation.[/quote]

 

[woop]

so even if they where not coupled, would it still deliver a constant current?
because most circuits for laser diodes that we use don't have thermal compensation.
I don't get how that diode and voltage divider can sense current, i would think that they would just act as a normal voltage divider... and divide the voltage.

 

[chimo]

The driver is configured as a voltage source. The voltage divider is not sensing current.

No, it does not give a "constant current" period.  It will only result in a pseudo-constant current if the diode is thermally coupled to the "real load" (which in this case is a high-power LED running at about a watt).  Recall that there is only about 1mA going through the diode - not very much to generate any heat.

It is a normal voltage divider that sends a voltage to the sense pin of the chip. The fixed resistor, the pot and the diode are all part of the voltage divider.  The characteristic of the diode (and the LED) is that their resistance decreases for a given voltage as they heat up.  That would normally mean an increase in current for a normal (or thermally uncoupled) voltage divider, however, they are using the thermal properties of the diode to alter the voltage divider with temperature and semi-regulate the current.

so even if they where not coupled, would it still deliver a constant current?
because most circuits for laser diodes that we use don't have thermal compensation.
I don't get how that diode and voltage divider can sense current, i would think that they would just act as a normal voltage divider... and divide the voltage.

 

[woop]

damn.
ok, how about using a shunt and an op-amp or transistors to provide feedback?
not necessarily with that chip.
don't some chips have current sense built in? i wonder if there is anyway to use that for constant current.

 

[chimo]

Here are some current regulator schematics to stir your thought process.

 

[woop]

hmm they all look pretty similar.
is that first one from a torch? i wonder how many led torches use constant current regulators that are easily modified.

 

[chimo]

is that first one from a torch?

All of them are. They are all boost regulators.