Taking my cooling system issue a little farther...thanks to others for replies on previous post regarding the temp sensor location.
What I have is a once-in-a-while leak, apparently only when climbing a hill just after I've added water to the radiator or overflow tank. I've not been able to locate the source of the leak, probably because it is on the condenser side of the radiator and out of view.
Or not. Here's what I observed:
Starting with a full radiator and the coolant overflow tank between max and min, and with a new radiator cap tightly closed, under ordinary conditions, the temp gauge stays resolutely at the exact midpoint on the gauge, even when climbing a hill. The most recent test of that, about 6 miles or so up a sustained gradual hill at 50+ mph, confirmed it. It showed NO leak.
I went back down the hill, put a piece of cardboard between the radiator and condenser to see if the temp gauge would climb above the middle (to confirm it is operational) and started back up. The gauge did begin to climb after a few miles on the hill, though never reached the red. When I found a place to pull over to check, there was water dripping steadily (but less than a steady stream) from near the extreme front passenger side, forward of the lower hose and overflow tank outlet. Couldn't see anything though, had no flashlight (my mistake).
Then I removed the cardboard, went home and kept the engine running for 15 more minutes and never saw a leak again.
Once the engine cooled, I discovered that the radiator was still completely full right up to the filler neck, BUT...the reservoir was now below the min mark by an inch or two. That suggests to me that the leak was water leaving from the reservoir, but I can't explain exactly why it would drain below the min mark. Or is this proper functioning of the system? I think I can accept that coolant is leaking from the radiator and being sucked back in from the reservoir. I think that makes sense.
Each time I've noticed the leak, it's been right after I've added water to the rad and reservoir.
I have been told that it takes some time for the system to "stabilize", ie, that there may be too much or too little water in the system and sometimes it gets spilled out or needs to be replenished until it's where it really needs to be.
I'm inclined to replace the radiator regardless, it's probably the original one and the car is just over 150k, so it's probably due. And it would be a lot cheaper to do it myself as a preventative measure, than to have a problem far from home where I need to pay someone else. But if the radiator isn't the problem, I don't want to be distracted from the real problem....like a head gasket issue. I have previously checked the radiator for combustion gases and that seemed to be negative, and the engine runs great with excellent CA emission test very recently. I didn't see any signs of a problem with the water pump, though I think I'll replace that too just to be sure.
I'd do a compression test but it's such a difficult process to reach the rear bank of plugs that I hesitate to dive into that, would probably have to hire someone for that job.
Anyone have any insight on this? I'll be grateful for suggestions.
BTW, this is a 3.3 L engine. 152k, I've put only 1k on it in the six months I've owned it.