[shockey`]

There is another possibility, dealer filled radiator to correct level but left overflow bottle empty, therefore when radiator water expands and gets hot during driving the water that has expanded and is excess to radiator has flowed to overflow bottle, then when engine cools down the water from overflow is sucked back to radiator like it is suppose to but leaves overflow bottle to look empty.
Dealer should have put clean coolant in overflow bottle too usually to about a fifth to quarter overflow water bottle level.
To check this need to see overflow bottle straight after hot engine has been shut down and if coolant is in overflow bottle with engine hot then dealer forgot to add coolant to overflow bottle which has happened to people I know. But if overflow bottle is still empty with hot engine then you have an issue to fix. But you did mention overtime you check radiator is full so system can't be leaking and I believe coolant is going back and forward from reservoir to radiator as it should and all you need to do add coolant to overflow reservoir when system cold, try half a litre and see how you go.

I don't think this is the case as it's just an overflow tank, no return line and evap isn't going that fast. but I couldn't help myself but laugh at the... Totally unintentional way to try and explain the basics of entropy lmao.

My guess is mineral corrosion broke off and got stuck somewhere in the rad tubes itself? Something like that. Fluid dynamics is rather complicated so I can't say X did that or didn't plus I'm not looking at it myself.

 

[shockey`]

Overflow tanks are plastic and sometimes developed cracks on the bottom allowing seepage. I would remove Overflow tank and check it for leaks.

I didn't think about that one 👍. Fits the timeline also.

 

[shockey`]

I had a car that the overflow tank would go empty every couple months.

Pulled it and it was wet on bottom. Very slow seeping and hairline Crack on the bottom. They get brittle from heat in engine compartment and develop cracks normally around mounting points etc.

Yeah I'm a material scientist that's autistic so all of the.. atmospheric corrosion (it's an engine -- the basic atmosphere is going to increase o3,no2,Br,Cl, and others levels) I honestly wasn't thinking about polymers. Something that I'm gonna need to do with these new nylon 3.3 type manifolds(anyone know exactly what polymer they're using?), which okay sure resist heat better and certain chemicals.

I hate my life.

 

[shockey`]

Talking about chemicals I use Penrite Ethylene Glycol concentrate good for 8years or 500,000kms which is Glysantin G48 antifreeze and anti boil with distilled water 50/50 mix.
This works in my old RS 1967 race spec Camaro, my 2015 modified WRXSTI, and my other standard Wrx, plus a LS injected Chev V8 I am a car lover and my 7 cars new and old work well with this penrite coolant.
But I like to change coolant every 2 to 3 years even though product says 8 years.
They never over heat or leak and no corrosion to any part of the engine manifolds or damage to hoses or gaskets, but I have been servicing my cars for 50 years but kept up with technology.

Thank you for that information - I'm a physics guy not chemistry of the organic type and unfortunately just use DI water for my race car and the factory specs for everything else - that will change now! 👍🙏

And I guess it's time for me to admit that 8 years may actually be a good thing for some of my cars that are projects-to-be 😜 😁 (c'mon I'm lazy whaddya expect!)