Direct modulation of laser diode using high-speed logic

I require any info/paper/article/book about direct modulation of semiconductor lasers (laser diode) for telecommunication applications using high-speed logic.

Especially using ECL or CML (output) logic.

The problem is about DC! interfacing of logic levels from high-speed device to a laser diode, which has to be driven by variable bias current i.e. 50-150mA...

Circuit examples of this interface would be nice.

Purpose of this is OPLL design - interface between laser diode and phase detector.

Thanks,
rfmw

You might want to take a look at Figure 2 from this article. It is using an ECL device (XX89) to drive a VCSEL laser diode.

http://www.edn.com/article/CA307867.html

The only thing is that the article's laser diode is a VCSEL and judging from your modulation current yours is probably a FP laser diode.

The ECL/CML line drivers with a variable output swing was a poor man's laser diode driver. You might want to see if anyone has an application note for it to drive a laser diode. ONSEMI and Micrel had these variable swing parts in their portfolios.

Hi krashkealoha,

you are right of course. I have a FP-type laser diode. For laser-diode (variable) currents of 100-150mA I need to have somekind of bias circuitry. Also for modulation purposes I need to connect somehow a high-speed logic output. The connection has to go down to DC, so the bias shouldn't interfere too much the logic output and vice-versa.

Thanks for your help,
rfmw

Hi..i just want to say a few words...the dc current here..its value is important corresponding to the threshold value of the laser to lase. so the value of dc injected current must be specifeid for proper work as times of threshold current of the laser diode. for instance, I(injected)= x I(threshold). the value of x must be large than 1. if the higher values of x have many advantages and also some dis advantages such as high power, low resonance frequenccyand ...etc. the disadv. is like the higher temperature of operation. and also must take into account the noise driven with injection current..this will effect the turn-on time delay and the line width. mich thanks