Here is a 1MHz laser modulator that uses the LM317 and the LT1215. The schematic is attached here. However, the LT1215 is fairly expensive. Would an LM318D work instead? Also can anyone shed some light on how the current limiting resistor value (8.2 ohm) is calculated?

 

I was all set to tell you the TI op-amp wasn't as fast, but it has an identical slew rate specification. It does have a lower gain-bandwidth product but that may not be a problem at "just" 1MHz.

The TI chip common mode range spec shows that it can't sense within 3.5V of the negative rail. That might knock it out for use in this circuit. The Linear op-amp is not rail-to-rail either but much closer.

I think the Linear op-amp has a higher output current spec, which could be important at VR1. The TI output requires making a calculation I'm too lazy to do.

Oh, and R11 is likely calculated for a specific voltage drop at a specified current. Would it matter if it was 10Ω or 6.8Ω? Doubt it. You can compensate at VR1 to get the desired output.

 

I'm thinking R11 is related to the words, "Gangs extra regulators." about paralleling LM317's.
I'm also thinking, "How do you use two inverting amplifiers with no negative supply voltage?"
My failure to understand that is why I didn't answer.

 

I'm thinking R11 is related to the words, "Gangs extra regulators." about paralleling LM317's.
I'm also thinking, "How do you use two inverting amplifiers with no negative supply voltage?"
My failure to understand that is why I didn't answer.

Yeah, I struggled with understanding how an input signal that goes to zero can be handled by an op-amp that doesn't sense anywhere near the negative rail.

 

The LM318 has a tendency to oscillate if not handled properly. I don't know about the LT1215.

Note that the schematic does not show power supply bypassing. The PCB does have a bypass cap, C3. C3 should be a 0.1uF ceramic cap. Much better power supply bypassing must be done such as having a large cap on the supply.

What makes you think the LM317 is useful at 1 MHz? You might be able to tease the information from the data sheet but I can't.

Bottom line. I would not expect this circuit to work and I would not waste my time building it.

 

The link in #1 addresses the 1MHZ issue. See what you think.

 

Thanks folks. This helps. Would substituting the LT1215 with a rail to rail Opamp help? Also I simulated this circuit using LTSpice and as expected I get weird results (attached here). After a little more digging, it seems this was built and is working according to this post.

 

The link in #1 addresses the 1MHZ issue. See what you think.

Still skeptical.