When they tested the air to air...did they take it off the truck and clean it?? If not...do so. You can not blow crude out while it is mounted...what goes in the front...has to come out the front.

 

The oil sample could be a clue to a head gasket issue. It may be a head gaket in the early stages of failing. A buddy was fighting the same symptoms/problem and replaced rad, w/pump, t/stats etc & didn't help. He ran a long time like that & finlly had a bit coolant in the oil. Ended up doing the head gasket & no more overheating either .

Try doing a buubble test...............Put the overfow hose from the rad into a container of water & with engine hot & the temperature stabilized check for bubbles. Keep in mind the engine is working a lot harder while driving & would push out more bubbles then.

 

Not to be state the obvious but make sure your guage is accurate. They can go out of calibration when they get old. What year and model of CAT is it?

 

Waterpump wise - have seen them fail without leaking. Impeller breaks off the shaft - it simply doesn't spin and move near enough coolant. All been on CATs tho not Detroit.

Headgasket could cause the over heat. You can have a headgasket fail and it will still pass the "bottle test". If they blow where they aren't venting pressure to a water gromet - you have the issue but no coolant/pressure in the tanks. Saw one like that last week. Single turbo C-15

 

I had a simuliar problem. Turned out to be the belt had a bad spot on it and would not allow the pump pully to turn as it just slipped on the sheave. Went to replace the water pump and found the belt being the issue. I like you went through everything in the book. I would have never found the problem had i not decided to replace the belt. Someone from this site suggested it was the belt and i shrugged it off. I replaced everything imaginable also.

Here is a link to that thread http://www.classadrivers.com/phpBB2/...ic.php?t=32751