UPS Dual Inputs

Status
Not open for further replies.
T

texun789

New member
I have been told by a UPS manufacturer that it is a violation of the NEC to have the rectifier and the static bypass of a UPS to be fed from separate sources. That is done to provide a measure of redundancy.

I have looked through the code book and have not been able to find where that is a violation.

My question: Is it a violation of the NEC to have the rectifier and bypass fed from separate sources?
 
Which UPS mfg is telling you this.

We do this type of install all of the time.
 
Sometimes for fun, you could bring one source to the rectifier/inverter, a second source to the static switch internal bypass, then a third to the maintenance bypass.
Just be careful to ensure that any closed transition to bypass is sync checked.
The output of the inverter is typically timed to sync to internal bypass.
 
ron said:
Sometimes for fun, you could bring one source to the rectifier/inverter, a second source to the static switch internal bypass, then a third to the maintenance bypass.
Just be careful to ensure that any closed transition to bypass is sync checked.
The output of the inverter is typically timed to sync to internal bypass.
Click to expand...

That's a little different that most of the UPS's I've spec'ed. On those, you had to activate the UPS internal bypass before you could activate the external bypass. (At least that's what I've been told.)

I wasn't aware that anyone synched the UPS inverter with the input, but it would make sense.

Steve
 
texun789 said:
I have been told by a UPS manufacturer that it is a violation of the NEC to have the rectifier and the static bypass of a UPS to be fed from separate sources. That is done to provide a measure of redundancy.
Click to expand...


I would ask the rep: "If it's a code violation, why do you make UPS's with separate inputs."

Or does this rep's company only make UPS's with single inputs?

Steve
 
steve66 said:
I wasn't aware that anyone synched the UPS inverter with the input, but it would make sense.
Click to expand...
They all do.

UPS inverters cant provide decent fault clearing current, so when a downstream fault occurs the UPS static switch goes into bypass, so the upstream source provides the fault clearing current. Once the fault is cleared, normal operation is resumed.
 
dbuckley said:
They all do.

UPS inverters cant provide decent fault clearing current, so when a downstream fault occurs the UPS static switch goes into bypass, so the upstream source provides the fault clearing current. Once the fault is cleared, normal operation is resumed.
Click to expand...


Cool, I never knew that but it sure makes sense now that you have brought it up. Thanks!
 
Bob,
It's because of the transfer to bypass during a fault is why I've been after the UPS manufacturers to provide a short circuit rating for the UPS module instead of just telling me that the rating of the input breaker AIC is the rating the whole module takes on. That's bogus.
 
dbuckley said:
They all do.

UPS inverters cant provide decent fault clearing current, so when a downstream fault occurs the UPS static switch goes into bypass, so the upstream source provides the fault clearing current. Once the fault is cleared, normal operation is resumed.
Click to expand...

I'm not following the connection between transfering to bypass during a fault, and synching the inverter output with the bypass. Couldn't they just switch during a zero crossing, or actually do a fast open transition switch?

Not saying you are wrong, but statements like "They all do" kind of make me cringe. Surely with all the different UPS manufacturers out there, one of them is going to do it differently ( or cheaper).
 
steve66 said:
Couldn't they just switch during a zero crossing, or actually do a fast open transition switch?
Click to expand...
They could.... but... if you wait until all the phases of the outgoing supply have reached zero cross before bringing the incoming supply on-line then there is a gap with no power. Given all three phases have to switch it could be a long gap depending on the relative angles.

You can crash switch a thyristor static switch with syncronised inputs, as as soon as the current through the outgoing supply drops below the thyristor holding current then the thyristor will stop conducting. (a thyristor once triggered stays conducting until the current through it is below the "holding current" whereupon it stops conducting - you cant command a thyrister 'off') So its easy for the UPS to drop inverter output at the same time as switching the output to bypass for a clean crash transition.

With an IGBT static switch you can crash switch at any time as you can switch IGBTs off. So you crash switch non-syncronised inputs.

But the real problem with non-syncronised input and inverter is this: Imagine what any motor powered off this UPS is going to do when the motor supply rotates through several degrees (or even many degrees, and in either direction) instantly. I wouldn't want to be in that lift...

Edited to note: When I talk about UPSs I'm generally talking about devices of significant size, and of the double conversion varienty. At least one manufacturer (Hitatchi or Toshiba I think) has a configurable option on the LCD panel that you can run the UPS without mains syncronisation, and then you can configure the output frequency to something other than the input. But in this mode it disables the static bypass and thus if you bolt the UPS and the fault doesn't clear within a cycle or two at something less than 200% full load amps the the output will collapse in short order.

The toy size UPSs only have a single transformer which does all the transforming, and in these boxes the input and output have to be in phase.
 
Last edited:
Status
Not open for further replies.
Top