I want to modulate the current through a laser diode with the output of a (FPGA into R2R DAC)source. I am using an opamp to do voltage to current conversion as shown in the circuit diagram. The laser diode is supposed to connect to the feedback loop of the opamp.

However, the laser diode is on an external PCB which has an SMA input. Hence I will connect the modulator to the diode using a 50-ohm coax cable. Since the coax cable is more than 10 cm long I am convinced that I need to treat it as a transmission line. The bandwidth required is 50 MHz. My solution to impedance matching the two systems is to put a resistor RS1 in series with the opamp out so as the net impedance there is 50 ohm and another RS2 near the diode so that the net impedance there too is 50 ohm (shown in the diagram). Is it the right way to do it? Do you have any suggestions?

I am guessing that the increased load due to the additional resistors won't be an issue since the current flowing through the line will be the same. enter image description here Just to clarify, the opamp used here is a high slew rate, high current output (100mA) LM7171. The specs of this opamp is compliant with my requirements.

  • \$\begingroup\$ Please edit your question with the degree of accuracy you need and the current with which you intend to drive the LED -- it bears on how complicated the circuit needs to be. \$\endgroup\$
    – TimWescott
    Jun 25, 2022 at 21:57
  • \$\begingroup\$ And it's probably a bad idea to drive your coax shield -- if it's an issue, it should be grounded. Depending on your needed accuracy, you may want to consider a Howland current pump or a PNP current mirror. \$\endgroup\$
    – TimWescott
    Jun 25, 2022 at 22:05
  • \$\begingroup\$ @TimWescott : I would ideally like to have a bias point of 35 mA and then have an increment of 1mA in both directions. So the range will be from 27mA - 43 mA for my 4bit R2R DAC. What do you mean by accuracy here though? The reproducibility of the current values? \$\endgroup\$ Jun 26, 2022 at 9:18

1 Answer 1


The bandwidth required is 50 MHz

50 MHz has a free-space wave length of 6 metres.

In cable it'll be about 4 metres.

Since the coax cable is more than 10 cm long I am convinced that I need to treat it as a transmission line

Do you really think 0.1 metres of t-line is going to be a showstopper?

The standard rule of thumb says if the length is shorter than one-tenth the shortest wavelength then it's not worth taking special action to deal with reflections. I'd start worrying at a t-line length of 50 cm.

  • \$\begingroup\$ Thank you for the answer. Earlier, I had made a voltage source earlier to control a laser modulator (EOM) and it had problems with signal integrity when connected through the same coax. That is why I tried to get some reassurance. A current-based source will also work the same? \$\endgroup\$ Jun 25, 2022 at 19:21
  • \$\begingroup\$ @RaviPradip If you really worry about the opamp stability, then place a footprint for a small capacitor from the opamp output to the inverting input. That will allow you to restrict useful opamp bandwidth to a level that is stable. \$\endgroup\$
    – tobalt
    Jun 25, 2022 at 19:22
  • \$\begingroup\$ @RaviPradip really good simulators are free. They will eat this problem and spit out the answer in less than 1 second. \$\endgroup\$
    – Andy aka
    Jun 25, 2022 at 19:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.