Yes, this is what I was asking.If you're asking have I tried using the CPU that was originally under heatsink A before it was lapped with heatsink A after it was lapped, the answer is no.
I wanted to be sure that the contact surface on the cooler wasn't an issue (both if there really is a noticeable difference in the CPU's; that is, with the improved contact surface, that one of the CPU's will still run more than a degree or two cooler in the same socket). That's why I was wondering if you'd done a CPU swap after lapping the cooler in question.
Now if one does run hotter, then place each one in the best location in order to keep them cool (i.e. hottest CPU in the cooler that gets the most cool ambient air <should be CPU A IIRC>).
I'm not saying that you've a problematic CPU, but the post cooler lapping swap will prove it, and also allow you to see which CPU may run cooler under it, even if it's just by a couple of degrees (67 - 72C should be fine anyway, as there is a little margin; just not a lot). There's more variance for limited use (i.e. can run hotter for short periods of time, and Intel has thermal data on this if you want to read up on the data sheets for your CPU family). But the higher temps aren't meant for 100% duty cycles (24/7 operation at load). This is where the 67C figure comes from, as that's what is stated as the max temp value for 24/7 operation.
85C is too hot for 24/7 operation. Even for extended periods of time (past Intel's thermal profiles), will cause damage (presuming that you will exceed Intel's thermal profiles for higher temp/duty cycle operations as a means of playing it safe).I just decided to eliminate as many factors as I could in one go. So I removed both heatsinks, swapped around the processors and the known cool running CPU in socketB/HeatsinkB was put into SocketA\Heatsink A and is it turned out the CPU's had nothing to do with it, both run at near same max (67c -72c diode temp) temperatures under Heatsink B and both run hot under heatsink A (diode temperature of 85c, heatsink temp of 73c or so as reported by hardware monitor)
Which is why I'm trying to help you find a way to get CPU B as cool as possible (tricks like lapping cooler B + finding the CPU of the pair that runs the coolest under load placed at location CPU B).
As mentioned, it would also be a good idea to get software to ramp up the fan speeds as a means of getting the load temp where it needs to be (you'll have to deal with the additional noise for the additional air flow), as both methodologies are cheaper than replacing the cooler for CPU B (and certainly cheaper than having to replace a CPU).
BTW, I'm assuming the CPU's aren't actually that different from one another, so the cooler lapping and fan control software are the biggest things you can try. But as the lapping hasn't had that much of an effect, swapping CPU's to find the coolest one for socket B isn't a bad idea either.
Hopefully this will give you an idea of my train of thought anyway.