what would you say would be the top end of psu temp is for good reliability
Very tough question. See this thread: http://forums.silentpcreview.com/forums ... php?t=2301
Short anwer: I don't know, and I don't know how even to measure. Again, see the post above.
The large HS in the Zalman should help with lower than intended airflow, and it does seem like a good quality product. Keep in mind that compared to Q-tech or Seasonic models reviewed, its fan control provides more airflow at a lower power level. IE, the fan speeds up sooner in response to thermistor temp rise. This suggests:
1) The Zalman folks want to keep the internal PSU temp lower & even at the cost of a bit more noise
2) The Zalman PSU components need more airflow to remain as stable (unlikely?).
3) Both of the above?
My own PoV and approach to PSUs:
1) Noise is paramount. I mod them with super quiet fans -- Panaflo at 5V, NMB at 7V.
2) I am willing to take take the risk of hgaving the PSU blow & the cost of replacing it if/when this happens.
3) I have killed 2 PSUs with low airflow (high temps):
- an old generic 250W died after nearly 2 years with a 4.5V Panaflo in a low power system
- a 350W Enermax died after a year with similar mod in a mid power AMD system.
4) Both of the above failures were preceded by a couple of weeks of audible whining that was not there before. Seems to be a sure-fire indication of imminent failure. I pulled them out of their respective systems and avoided potential damage to other components. They failed soon thereafter during routine checks (not torture testing) on the testbench.
5) I always have a complete PC that duplicates the power/functions of the main one with a complete HDD backup of the main.
if I find a PSU that is as quiet as my modded ones, I would stop hacking and buy that instead for the assurance of the warranty and the comfort of knowing(maybe) that the unit was designed specifically to survive the low airflow.