I don't know if I'm posting this at the right place, but here it goes anyway. I hope you guys understand what I'm trying to say.
Some people, including myself, have been discussing on brazilian forums (I'm brazilian) about the side effects of having relays opening and closing on the AC line, those relays usually found in UPSses and AC voltage regulators. That annoying clicking sound they make every time a diferent transformer tap is selected to add or subtract a few volts to make the input voltage to the PSU as close as possible to it's nominal voltage ratings and its effects on the AC sinusoidal wave, is the concern here.
Well, it's been said that relays induce on AC power, noise, small surges and causes inrush currents. The latter is what I want to put into discussion here.
Since most if not all computer PSUs use a NTC Termistor to limit the start-up current drawn by the PSU when it's first powered, the question that has been arisen is how would the inrush current be limited when a PSU is working at normal conditions, since the NTC termistor is bypassed a few seconds after the power starts to enter the PSU?
The thing is, the time the relay makes the PSU float with no power until a different transformer tap is selected. The avarage time between this switching process from one tap to another, is about 8 miliseconds, sometimes more, sometimes less. Since the standards for ATX PSUs require that they should not drop load for a time of 16.6 miliseconds after the AC power is removed, this would make me conclude that an 8ms relay switching time would make the power stored on the PSU bulk capacitors drop to half and that to recharge it again its to its full capacity, an inrush current would happen, one now bigger because the hot termistor can't do anything to stop it.
The estimated inrush current for a typical PSU can be as high as 120 Amperes @ cold start and if not limited somehow (usually by a termistor), damage can occur.
Read these articles, their both very similar and from the same source:
So, is it logical that an 8 miliseconds time that the PSU floats without power, would make the PSU cause and inrush current of let's say 60A, since now the NTC Termistor is bypassed? Wouldn't the clicking relays inside UPSes and Voltage regulators cause some kind of situation that would reduce the PSU lifetime and stuff like that? I'm using the numer 60A here based on the 80 to 120A inrush current of a typical power supply.
I know that for the bulk capacitor to reach half of its charge in 8 miliseconds, the power supply must be supplying its full rated power, but it happens sometimes and even if the bulk capacitor drops 1/4 of its charge, it would mean that a current surge of about 15Amps would still reach the PSU input stage everytime the relay clicks and for situations where the AC power is unstable, the relay would click a lot and current surges would be happening during this process, stressing the PSU little by little.
Is it right what I'm exposing here? I hope somebody can give me some intelligent and interesting answers.
I apologise for any english mistake I must have made!
Thanks!
Some people, including myself, have been discussing on brazilian forums (I'm brazilian) about the side effects of having relays opening and closing on the AC line, those relays usually found in UPSses and AC voltage regulators. That annoying clicking sound they make every time a diferent transformer tap is selected to add or subtract a few volts to make the input voltage to the PSU as close as possible to it's nominal voltage ratings and its effects on the AC sinusoidal wave, is the concern here.
Well, it's been said that relays induce on AC power, noise, small surges and causes inrush currents. The latter is what I want to put into discussion here.
Since most if not all computer PSUs use a NTC Termistor to limit the start-up current drawn by the PSU when it's first powered, the question that has been arisen is how would the inrush current be limited when a PSU is working at normal conditions, since the NTC termistor is bypassed a few seconds after the power starts to enter the PSU?
The thing is, the time the relay makes the PSU float with no power until a different transformer tap is selected. The avarage time between this switching process from one tap to another, is about 8 miliseconds, sometimes more, sometimes less. Since the standards for ATX PSUs require that they should not drop load for a time of 16.6 miliseconds after the AC power is removed, this would make me conclude that an 8ms relay switching time would make the power stored on the PSU bulk capacitors drop to half and that to recharge it again its to its full capacity, an inrush current would happen, one now bigger because the hot termistor can't do anything to stop it.
The estimated inrush current for a typical PSU can be as high as 120 Amperes @ cold start and if not limited somehow (usually by a termistor), damage can occur.
Read these articles, their both very similar and from the same source:
So, is it logical that an 8 miliseconds time that the PSU floats without power, would make the PSU cause and inrush current of let's say 60A, since now the NTC Termistor is bypassed? Wouldn't the clicking relays inside UPSes and Voltage regulators cause some kind of situation that would reduce the PSU lifetime and stuff like that? I'm using the numer 60A here based on the 80 to 120A inrush current of a typical power supply.
I know that for the bulk capacitor to reach half of its charge in 8 miliseconds, the power supply must be supplying its full rated power, but it happens sometimes and even if the bulk capacitor drops 1/4 of its charge, it would mean that a current surge of about 15Amps would still reach the PSU input stage everytime the relay clicks and for situations where the AC power is unstable, the relay would click a lot and current surges would be happening during this process, stressing the PSU little by little.
Is it right what I'm exposing here? I hope somebody can give me some intelligent and interesting answers.
I apologise for any english mistake I must have made!
Thanks!
