I own a Core i7-3770 (non-K version) with maximum turbo boost of 4.3GHz (actually 4.3/4.3/4.2/4.1) with a Z77 motherboard.
When setting the multiplier to 43, the maximum temperatures when running Prime95 and IBT wasn't making me happy, sometimes would go as high as 90°C, so I decided to make a manual fine tuning in LLC and base/additional turbo boost voltage.
The first thing I had to do is to find the lowest stable VID with turbo boost disabled (base clock of 3.4 GHz), then after find how many additional voltage I have to add to the base VID to make the CPU stable when the maximum turbo boost kicks in.
I managed to find a base VID of 0.880v (24h of Prime95 blend and 50 IBT passes@maximum stable), this makes happy because the maximum temperature was 58°C during the tests, then I were about to proceed to enable turbo boost, bump additional turbo boost voltage and run a new batch of tests, but something caught my attention.
Using HWINFO and AIDA64 to log data from sensors, I noticed that the TDP remained the same (about 58W) no matter how low the VID was, this means that the electric current would just go up as I reduced the VID to keep the TDP the same. What I mean is:
Stock 3.4GHz VID @ 1.220
58W = i * 1.220V, so i = 47.54A
Undervolted 3.4GHz VID @ 0.880V
58W = i * 0.880V, so i = 64.91A
The abnormal increase of current is going to stress or damage my CPU and motherboard VRM running this undervolt 24/7?
I am missing something and the sensors are inaccurate, or my assumption that the current is being bumped up to compensate the lower voltages is right?