Question about voltage regulation for GPSDO and controller + heat

[cdev]

I have a precision frequency standard (a TruePosition GPSDO) that seems to work with any supply voltage above around 9v. It has internal voltage regulation.  Also, although a number of people, especially @Pigrew and @Texaspyro have done a great deal of work to reverse engineer the device's available commands and meaning of the output message stream, I think its realistic to say that a significant body of relevant information about it's operation, command set, etc. remain little known compared to most other GPSDOs. Because its sold to telecoms as part of a system for locating cellphone emergency callers by TDOA that it appears is EOL.

I have a nice STM32F103C8T6 controller for it which was designed by @Pigrew and its soon going into a case with the GPSDO. It generates some, but very little heat. Similarly, the OLED display runs on almost nothing and generates almost no heat.  However, voltage regulators do emit heat. And I think the device should have additional voltage regulation given that its going to be powered by a wall wart continuously, 24/7. I may use two voltage regulators, one for 12v (or 9v) and one for 5 volts. If it can run successfully of 9 volts with good accuracy that would be better because going from 9 to 5 volts is likely to generate significantly less heat, enabling a higher degree of accuracy perhaps for the oven. .

I don't know if I'm getting across exactly what I want to say, I'm kind of exhausted, but you probably see what I am getting at.  Its actually very basic stuff.

The GPSDO has a command $TRAINGPSDO which literally entrains the GPSDO when conditions which could impact the performance change. It seems to work.

The final form of this device is likely to need one or perhaps two voltage regulators, and my plan is to mount them on the back of the device's case to hopefully have the heat not impact the oven.  I want to make sure they don't interact with the GPSDO's temperature control.

I just don't want to make a stupid mistake with the power supply that decreases the device's accuracy or life..

Does anybody know of a document specifically on GPSDO "hardware integration"