Temperature monitoring for storage hot water system (RP-2 connection)

My HWS has a secondary outlet connection I don’t use, and I’d like to use it to install a thermal probe I can monitor.

HWS RP ¾ / 20 temp probe?

Does anyone know of a practical way to install a temp probe of any kind via a RP-2 connector on a HWS that’s kept under mains pressure? (Obviously I’ll isolate it for the install). I’ve found a few options that are $OMGWTF, but nothing that looks very practical.

Has anyone here done anything like that? If so, how’d it work out? Any advice?

The HWS spec says it’s BSP RP ¾ / 20.

But why?

I now have a dumb timer on my power distribution board that lets me limit its boost to solar generation hours. But it’s hard to work out what the best limits to apply are, and hard to adjust. I’d really like to be able to run a temp probe off the HWS and datalog it from a microcontroller or RPi or similar, but that means being able to safely and securely install the probe in the top of the HWS.

Later I’d be able to use PWM / PLL to do solar diversion, but due to our quirky install (2 x 2.5kW inverters on 2 different phases, neither of them the HWS phase) this isn’t as simple as “get EMonPi and OpenEnergyMonitor diverter, done”. It’d need $lots of hardware and likely sw customisations, at which point you have to ask if it even makes sense.

Also, as far as I can tell the HWS exposes no data about its own thermometer, it’s a dumb, simple, reliable mechanical unit. And it wouldn’t work when the HWS is powered off anyway. So I can’t use existing HWS facilities.

Not actuators or thermostats

I’m not looking for a thermostatic actuator, or a thermostat. I’m after a digital readout or preferably an output in SPI (preferable), i2c or 1-wire form.

One option I’m looking at is running a thermocouple wire to an external sensor but I’m quite worried about the pressure integrity implications of that if I don’t have an off-the-shelf fitting.

Surface contact enough?

One option may be to monitor the surface temperature of the fitting under the insulation, or the tank surface. Anyone done it? Work well enough? Easier? I could just bias it to estimate the actual internal temp based on knowing the thermostat max setting.

Otherwise it seems there are thermowells intended for this sort of thing, where you put the sealed, machined thermowell in, then use a regular temp sensor in the well.

Advice?