cgminer is stopping work because one of the cores is too hot. I find that will happen if I haven't done a good job of mounting the heatsink; I'll see most of the chips sitting at 60-70 and one will shoot up to 100. The back and forth with cgminer is something we'll fix in a FW update, but the root cause is bad heatsink contact.
Will look into it but since reported temperature starts at 35 and go to 66 I don't think this is it... Unless we don't see highest temperature reported...
Replaced thermal pads... And problem fixed... But why I didn't see this hi temperature reported in cgminer? And got speed for 37.2 to 39.8

I'm actually not sure, obviously it's getting reported to cgminer since it is throttling. It must be displaying just one of the temps and not the other.
As for reporting all of them, it's in the works probably through a custom Z*X command. We just need to implement it, and then get support added to the mining software.
Throttling should be reduced in the next firmware release. Part of the problem was that if Chili was idle because the host mining software was not sending any jobs (perhaps it's not started yet or there is a problem with the pool connection), then Chili would cool off and notice that it could crank up the frequency and voltage. This is bad when work unit start coming again because it is now running too fast & hot.
New firmware will not change frequency or voltage when there are no work units in the queue. I'm testing that and other fixes now.
Edit: The temperature that is reported is the temperature of the hottest ASIC chip.