I believe the LP2950 also has a lower max current compared to the LM78 regs. In fact, they make LM78 regs in the
really big package that are made to be bolted to a heatsink.
Like, when would be a good time to use the LP2950 over the LM78L05 ?
I spec the LP2950 for the relay bypass PCBs I've made, both the
microcontroller based design, as well as the CMOS hex inverter based design. In both cases, the goal is to minimize power consumption as much as possible, because unless you're engaging/disengaging the effect, the bypass circuit isn't doing anything except waiting for the footswitch to be activated.
When I programmed the logic on the microcontrollers, I had them go into their lowest power state ("deep sleep" or something like that) when not actually debouncing the switch or (re)setting the relay. The microcontrollers I used could still watch a pin for a voltage change (i.e. footswitch press) in this ultra-low power state. IIRC, this ultra low power state was in the
micro amps current draw.
Similarly, the CMOS chips (e.g. 74hc14, cd40106) are ultra-low current, they are also in the
micro amps category.
Actually (re)setting the relay will draw 10s of miiliamps, but only for a few dozen milliseconds. The overwhelming majority of the time the circuit is in the "waiting for switch press" state, where we're back down to microamp current draw.
So this is a case where the LP2950 will save significant power versus the LM78. As JTEX said above, even with the business part of the circuit only needing 10s of microamps, the LM78 itself will draw (waste) 3mA, where the LP2950 will be... I can't remember, but significantly less.
