First off this is my HTPC setup:

Celeron 420 (1.6ghz, 1.3Vcore @2.6GHz)
Zotac 7100 Supreme motherboard with July 2008 BIOS
Enermax Liberty 400W PSU
Antec NSK2480 case

cooling specs:
2x 120mm Antec Tri-Cool fans on low (factory)
Scythe Mini Ninja with 80mm DFS fan (original fan that came with the cooler)
CPU and HSF are both lapped to 2000 grit paper, remounted twice for a drop of ~4°C (probably wasn't clean enough)
I use Tuniq TX2 TIM and ArctiClean everytime I mount new coolers

Here's the issue, now that you have the background:
The Celeron 420 is a 35W processor, and it currently idles at 50°C+, it maxes out at 65°C with a 5 hour run of Orthos. What gives? The Mini Ninja is a great cooler and actually held my Pentium 4 3.2Ghz HT Prescott (S478) at a lower temperature in its previous use.

Obvious questions:
Yes, the push-pins are fully seated and triple checked.
Yes, I thoroughly cleaned both the CPU and the HSF before applying new TIM.
Yes, I used the correct amount of TIM, it makes a perfect footprint when seperated.

I was thinking about trying to remove the IHS, but I'm not sure if that would do anything to keep it cooler and possibly OC it higher. I know the processor is within its acceptable operating limits, but the engineer in me keeps wanting to know what's going on to result in such a high temperature. Thanks in advance for your help. Sorry for the long-winded-ness.

It could be a couple things
Either you mobo is reporting the wrong voltage, and its cranked up too high
Your heatsink itsnt clamped down all the way, or your fan attached to it is diying or not blowing enough
You just plain have a borked cpu

My motherboard does not allow any voltage adjustment for the CPU, so it should be running at its VID (1.3V), whether or not the motherboard is delivering that voltage, is not currently known.

Yes, the heatsink is fine.

No, the fan hasn't failed. Nice try though. (Edit: even if it had, the HSF was designed to be passive specifically for this case, and shouldn't have temperatures that high Link - Pentium D 950 @ 55°C passive)

My thought was that since the Allendale and Conroe-L CPUs have thermal grease instead of thermal epoxy between the core and the IHS, maybe it was manufactured poorly, or has degraded. This is the reason I'm considering removing the IHS. And if I destroy the CPU in trying to do that I'll just get a new E5200 to replace it.

Anyone with experience removing S775 processor IHSs that can give me some pointers. I've read a fair amount on other forums, but I still don't know about the mating surface and whether or not I'll need to modify the mounting to have good contact.

Bump, any other thoughts, suggestions, opinions?