The OP already paid Garmin for the capability, just add the cheap sensor.
Yep, overthinking it, but anything is better than getting back to my taxes.
Executive Summary: Would a temp sensor /loop enable me to turn an inflight fire event into an inflight overheat event? IDK. Maybe. But I think I'm gonna run some temp sensors FWF anyway.
****
Long version:
It is helpful for me to think about what precisely we mean when we say "fire detection." We are likely talking about OVERHEAT detection. Overheats can lead to a fire. Fires are proceeded by an overheat. Fire detection is just a sub-range within the overall range of sensor heat detection where things in your FWF start to change state.
With this in mind, I am pondering the following:
-Are there a meaningful number of scenarios in which a pilot could avoid an inflight fire given instrumental evidence of rising and/or excessive heat FWF (but no fire, yet)?
-Do I really care if there is an actual fire FWF if the temps are "much higher than normal"? Which leads me to the question...
-Could I meaningful temp ranges for caution and warning alerts? Ranges like...
"Yes Garmin, Thank you. That's a bit higher than normal, I'll look into it" and "WOW, no doubt that'a $*&@ fire!!"
-Are there redundancy/design elements/procedures that allow for safe continued flight after a "nuisance trip"? If not, am I really gonna treat an indication alone as a fire? A caution? There are HF issues to consider.
These are the some of the questions that I need to answer before installing a fire/overheat detection system FWF.
HOWEVER, the OP got me thinking about adding at least a couple of K-type temp probes to my yet unsealed FW pass-throughs. If for no other reason, I can foresee future tests/projects/scenarios that would make it a good idea to add a couple of cheap probes now vs later. Suspect that your flux capacitor is failing due to heat? move the sensor and check it. Does opening the oil door after flight really help cool that much? You have another data point. And I already paid Garmin for some currently unused monitoring features. Heck, I'm not sure that a pair of unused shielded wires through the firewall now wouldn't be a blessing in the future. I'm estimating a C-note and minor labor for an unsealed firewall?