Purpose of Avalanche Transistor and Inductor in Power Stage of PLD Driver

Thread Starter


Joined Jan 11, 2016
Hello All,

I have decided to start playing around with LiDAR for range finding with the eventual goal of measuring wind speed using either the Doppler shift of the laser light or using particulate in the air. In order to get to that point though I need a LiDAR system first so I thought I'd build one, and here I am. I found an app note that has a pretty cool description of power needs of pulsed laser diodes (PLD) and found this circuit for a power output stage that i'd like to build but am a bit confused and it has given rise to some questions, here is my current guess at circuit operation.

When Vcc is applied, C2, C3, L4, and L3 all charge while L1 and L2 remain discharged. When avalanche mode is triggered, current flows from collector to base creating a forward bias on the emitter side diode essentially turning on the transistor. Now that there is a path to ground a current spike can be generated by C2 and the node between C2 and D1 is now negative. This makes D1 forward biased instead of reverse and is now able to flick on.

My questions are now primarily about the inductors. I realize that L1 and L2 are probably there for current limiting but why not just use low value resistors to keep the amps drawn high? Additionally, what is the purpose of L4, a series tank? I have the same question for C3 and L3 because there doesn't appear to be a need for a positive voltage at that node when discharging. Lastly, the only avalanche mode circuits I have seen have been setup in astable mode and don't have triggering at the base, rather the collector voltage is just cranked up, is the trigger there primarily to just "reset" the avalanche conditions or is it actually what causes it?

I've attached an image of the circuit, along with its originally app note, as well as a link to Linear Technology's app Note 47 that has an astable setup for testing o-scope rise time on p.93. Thanks for any help you can provide in advance, I just can't wrap my head around it fully this time.




Joined May 4, 2013
I am probably wrong so take these comments only as a starting point for a conversation...

First, it would help a _lot_ if your circuit had values.

I don't see any use for L1 other than to slow the avananche pulse. I appears in series with the otherwise low on resistance of the transistor.

I am thinking that the inductors are just a representation of the inductances in the circuit.

Frr instance, L1 is the inductance of the transistor. L2 is the series inductance of the Laser diode. L3 and c3 together are the capacitance of the


Joined Jun 19, 2012
...adding to the conversation:

LIDAR is on the hairy bleeding edge of high performance opto-electronics design.
Unless you are already proficient in this rarefied domain, don't expect much in terms of success.

Thread Starter


Joined Jan 11, 2016
Hey guys, sorry for the late post (swamped at work) but I spoke with a friend who is big into RF and he told me that all the inductors are parasitic indcutances just like RichardO said. According to him, this type of circuit doesn't have current limiting built in since they are usually just trying to cram as much current through the laser diode as quickly as possible (i.e. square wave) but since nature and this being real life and all the circuit gives you a ramp. As for C3 and L3, they are most likely caused by they high speed switching on the trace (L3) and the parasitic capacitance from the trace to the ground plane through the board (C3).


Joined Oct 9, 2015
@Mezzer26: Did you get anywhere towards your goal of wind measurement? I have the same goal.
I find it is easier and cheaper to use a low cost micro to generate precisionly formed and timed sub-microsecond pulses for a laser diode. Use a high speed, high current FET to do the actual driving. Then you have the pulse length and spacing easily controlled by your software.