Adjust your laser's output on the fly(digital pot)

[691175002]

Re: Adjust your laser's output on the fly(digital

Here is the diagram for a TTL red module:
http://www.sparkfun.com/datasheets/Components/General/TTL-Laser.pdf
http://www.sparkfun.com/commerce/product_info.php?products_id=8654

I would personally put the cap after the transistor, but otherwise pretty much how I expected.

 

[Benm]

Re: Adjust your laser's output on the fly(digital

What is the point of PWM if you put the cap after the transistor?

 

[691175002]

Re: Adjust your laser's output on the fly(digital

What is the point of PWM if you put the cap after the transistor?

The capacitor will not flatten out the PWM completely, 0.1uF will charge and discharge at a far higher frequency than the PWM signal.  My main concern is what happens if the cap is before the transistor, in a normal driver if you allow the cap to charge up and then connect it to a laser diode, the burst of current from the capacitor can fry the diode.  I foresee the capacitor potentially saving up the current from the off cycle and dumping it all at once.

Finally, in a current limited driver, even high capacitance will simply flatten the signal out into a constant voltage/current so nothing is really lost.

I don't know much about electronics though, so I would probbably test both ways to check if they work.

 

[danq]

Re: Adjust your laser's output on the fly(digital

What is the point of PWM if you put the cap after the transistor?

Ive done it with a 10uf cap on the output, and it dims very nicely. Didn't have a scope to look at it to seee whether it was averaging or still fully pulsed though...

if you allow the cap to charge up and then connect it to a laser diode, the burst of current from the capacitor can fry the diode. I foresee the capacitor potentially saving up the current from the off cycle and dumping it all at once.

you should have a bleeder resistor across that cap - say, 1K to 4.7K - then there won't be a danger of that. Or else you should not connect a LD without first shorting out that cap...

DanQ

 

[691175002]

Re: Adjust your laser's output on the fly(digital

if you allow the cap to charge up and then connect it to a laser diode, the burst of current from the capacitor can fry the diode.  I foresee the capacitor potentially saving up the current from the off cycle and dumping it all at once.

you should have a bleeder resistor across that cap - say, 1K to 4.7K - then there won't be a danger of that. Or else you should not connect a LD without first shorting out that cap...

DanQ[/quote]

What I meant was that the transistor in PWM mode would almost be like disconnecting and reconnecting the diode over and over if the capacitor is before the transistor.

 

[Benm]

Re: Adjust your laser's output on the fly(digital

[quote author=Benm link=1231947558/20#33 date=1233108025]What is the point of PWM if you put the cap after the transistor?

Ive done it with a 10uf cap on the output, and it dims very nicely. Didn't have a scope to look at it to seee whether it was averaging or still fully pulsed though...
[/quote]

That depends on the PWM frequency and laser current, but i'm sure the cap will smooth out the pwn signal to some degree - not desireable if you ask me. Also, i depends on wether your driver can only source current, or also sink it. In the latter case, the capacitor will form an effective short circuit.

A 10 uF capacitor has a reactance of 1.5 ohms at 10 kHz - low enough to blow your driver transistor if the driver can also sink current, or to average out the pwm if it does not.

 

[691175002]

Re: Adjust your laser's output on the fly(digital

A 10 uF capacitor has a reactance of 1.5 ohms at 10 kHz - low enough to blow your driver transistor if the driver can also sink current, or to average out the pwm  if it does not.

Isn't reactance only for AC current? If it never goes negative the capacitor will always be charged somewhat.

 

[pseudolobster]

Re: Adjust your laser's output on the fly(digital

I'd be interested in this, though I think your estimate of $10 each is a bit optimistic... It'll be a fair bit of work to print boards, flash mc's, solder everything, test it, package it, and mail it, when your parts cost is already probably 75% of your price... Seems like a lot of work for a dollar or two, I'm sure once you figure in transit costs you're going to be not only doing all the labor for free but probably losing money overall. It's a fantastic idea, but I personally wouldn't be willing to pay the $20 one of these is worth...

If you manage to write the code for a pic12f6xx and draft a working circuit diagram I'd be interested in a group buy on the required digital pots for $6 each or something... I have a programmer and a couple pic's laying around somewhere from a li-po alarm I never started.

 

[Benm]

Re: Adjust your laser's output on the fly(digital

[quote author=Benm link=1231947558/20#37 date=1233152177]A 10 uF capacitor has a reactance of 1.5 ohms at 10 kHz - low enough to blow your driver transistor if the driver can also sink current, or to average out the pwm  if it does not.

Isn't reactance only for AC current?  If it never goes negative the capacitor will always be charged somewhat.[/quote]

It doesnt have to go negative, but it does have to sink current for this problem to develop. If the driver sinks current, the capacitor will discharge into the drivers output on the low parts of the pwm cycle. Typical led/kd drivers have an open collector or emitter output and dont suffer from this, but if you are driving pwn directly from a logic output or push-pull endstage, you're in trouble.

For the AC/DC aspect of this: switching 0 to 5v and back is equivalent to 2.5V top-top AC, but superimposed on a 2.5V DC bias - the latter is of no consquence for reactance.

 

[erdabyz]

Re: Adjust your laser's output on the fly(digital

I'd be interested in this, though I think your estimate of $10 each is a bit optimistic... It'll be a fair bit of work to print boards, flash mc's, solder everything, test it, package it, and mail it, when your parts cost is already probably 75% of your price... Seems like a lot of work for a dollar or two, I'm sure once you figure in transit costs you're going to be not only doing all the labor for free but probably losing money overall. It's a fantastic idea, but I personally wouldn't be willing to pay the $20 one of these is worth...

If you manage to write the code for a pic12f6xx and draft a working circuit diagram I'd be interested in a group buy on the required digital pots for $6 each or something... I have a programmer and a couple pic's laying around somewhere from a li-po alarm I never started.

Nah, i finally can't do it with digital pots, it'll be a pot in paralel with the existing one, and a mosfet to connect it or not. That way we have a pwm control to the current.

I have no tiem to develop it, i have my new soldering iron but no time. I think the prototype will be done tomorrow and i'll show you pics and stuff if it works.

Student's life sucks...