Ripple
Despite the Corsair SF750 making very little of its ripple results, our test results found that it gave a very impressive performance. The 3.3V and 5V rails results were both impressively low. While the 12V does get a little more excitable (particularly at 80%+ loads) they are still massively within acceptable remits.
It is, once again, remarkable that Corsair has been able to not only get high-efficiency but also clean and consistent power delivery.
| Load (%) | 3.3V Ripple | 5V Ripple | 12V Ripple |
| 20 | 3 | 4.2 | 12.8 |
| 40 | 4.2 | 5.8 | 13.4 |
| 60 | 4.2 | 4.6 | 14.4 |
| 80 | 5.6 | 4.6 | 19.2 |
| 100 | 6.2 | 5.6 | 19.6 |
The ‘excessive’ glue is on purpose, it quiets down components which could vibrate and cause coil whine. Sure it doesn’t look pretty but all manufacturers do it (and at least corsair used black so it was less obtrusive). Look at the JonnyGuru photos of the AX1500i (arguably best PSU made) and this amount pales in comparison.
I understand the purpose, but honestly, not all manufacturers do it or, at least to this quantity. I probably see the insides of around 100 different power supplies a year, some have lots, some have none, this definitely falls into the ‘more than usual’ category. I do, however, credit the design for using black glue to make it much less obvious than the standard white and it’s certainly not a criticism of the power supply as a whole.
Hi,
Thanks for this review but have you forgot about noise measurement?
Please add actual DB numbers.
Thanks
Hi Astro, We don’t tend to noise test power supplies for two reasons. Firstly, excessive noise in them is rare and usually only when they are overly dusty/hot/faulty (but when it does exist, we do comment upon it) and secondly, the machine we test them on is AMAZINGLY loud.
I will, however, look to make noise a factor in future reviews. Just got to think of a good application for this.