This will sound weird but it is true (from my experience)
Some times when you over clock a system then at some point down the line replace the psu things can get screwy. I dont know why it happens and have never heard of anyone else it happening to. But as said above reset your overclock see if everything is ok then re do your overclock.
Actually, I think it makes perfect sense.
Power supplies rarely output exactly +12.00VDC, +5.00VDC or +3.30VDC. In fact, according to the ATX Form Factor standard, those voltages are allowed to vary up to a tolerance of ±5%. That means PSU voltages can vary as follows:
12VDC ±5% = 11.4 to 12.6VDC
5VDC ±5% = 4.75 to 5.25VDC
3.3VDC ±5% = 3.14 to 3.47VDC
It is safe to assume the voltages of your old PSU and your new PSU were not exactly the same. For example, the 12V for your old might have been +11.6VDC and for your new +12.3VDC. +4.81V for the old, 5.22V for the new.
The motherboard regulator/divider circuits attempt to compensate based on the incoming voltages. Overclocking is done by making tiny adjustments to voltages applied to the overclocked components. When you change the input voltages by swapping power supplies, the resulting output voltages to the various components are changed, albeit very slightly, too. But very slight changes to the voltages are all it takes to move an overclocked component from a stable state to an unstable state. So whenever you change power supplies, you should reset your clocks and start over with the overclocking process.
I agree your new TT PSU is highly rated (or at least the 650W and the 850W in that same line are highly rated) and 750W is more than enough horsepower for your installed components. Of course, until Man can create perfection 100% of the time, even the best maker can have an out-of-tolerance sample that sneak past quality assurance, or fail prematurely.
I am not a fan of using a multimeter for testing PSUs. They just are not conclusive enough. For one, power supplies must
be tested under a variety of realistic loads - from idle to 100% capacity while operating at the rated operating temperature. For that TT PSU, that is 40°C at full power. This means sticking highly conductive and very sharp meter probes into live connectors while the computer is running a variety of tasks. This is a risky procedure for experts as well as amateurs.
Also, the vast majority of multimeters cannot test for excessive ripple
and other anomalies that affect computer stability.
To conclusively test a PSU, it needs to be done by a qualified technician using an oscilloscope or power supply analyzer - sophisticated (and expensive) electronic test equipment requiring special training to operate, and a basic knowledge of electronics theory to understand the results. Therefore, conclusively
testing a power supply is done in properly equipped electronics repair facilities.
Of course, most "normal" users don't have access to such test equipment anyway. Their only conclusive option is to swap in a known good PSU.
So I recommend you (1) verify all your connectors are securely fastened. And recheck all other cables and wires too to make sure they were not disturbed when you swapped supplies. Then (2) reset your clocks and see what happens. If still having problems after that, you need to try yet another PSU.
If still having problems with a 3rd PSU and default clocks, then I would be worried the first PSU that failed may have damaged something connected to it!
Oh, also, make sure your case is doing its job of providing an adequate supply of cool air flowing through the case. This is critical, but especially critical when overclocking.
One final comment. Holding the power button down with the power cord removed does nothing, contrary to popular belief (on PCs anyway - notebooks are a different story). That "myth" came out the ancient past with "AT" Form Factor systems used with the original IBM PCs when the case's power button was actually direct connected via a wiring harness that ran directly back to the "AT" power supply.
The ATX Form Factor standard changed all that way back in 1996!!!! This is when ATX cases started using "remote" (not direct connected) power switches that connected to "momentary" circuits on ATX motherboards. Momentary means once the push is registered by the logic circuit it is connected to, any further input is ignored until the switch is pressed again
. There are no storage capacitors in that circuit to bleed off or reset.
So, just unplugging the PSU from the wall for 10 seconds or so is enough to ensure any residual voltages are fully degraded. And technically, the better solution is to flip the master power switch on the back of the PSU to off ("0")while leaving the power cable connected to the wall. This then provides a path to Earth ground for any residual voltages stored in an capacitors to fully discharge quickly and properly. Of course, this assumes the PSU has not been physically damaged, the power cord is in good repair, and the wall outlet is properly wired and grounded to Earth ground. To verify that, every home owner and computer user should have access to a AC Outlet Tester
to ensure your outlet is properly wired and grounded.