input doesn't equal output?

[CHWflorida]

I have a ups that keeps reporting an overload. I checked amps on each phase and found all 3 to be steady and pretty balanced on the input. But on the output L1-N jumps from ~15 to ~25 repeatedly. This is also the phase that over loads. I don't understand why this change does not show on the input. I have little knowledge of how ups works (obviously) and according to liebert the problem is on my end. Your help will be appreciated.

 

[ron]

If this is a double conversion unit, then the input is not directly coupled to the output. The input will need to provide sufficient power to support the load output and losses. The input is used to create DC, then the DC is converted to three phase AC. So the fluctuations seen on the output, should be seen on the input, but not direct phase for phase, but an overall wattage increase in accordance with the wattage usage.

 

[LarryFine]

CH, have you measured the voltages, both loaded and unloaded? Also, you could try rotating the phases temporarily to see if the fluctuation stays with the load conductor, meaning it really is a load phenomenon.

If it is, you may be able to isolate sections of the load where they branch with a clamp-on, one at a time, until you isolate the offending load. Use a methodic process of elimination: find out what the cause is not.

 

[CHWflorida]

If this is a double conversion unit, then the input is not directly coupled to the output. The input will need to provide sufficient power to support the load output and losses. The input is used to create DC, then the DC is converted to three phase AC. So the fluctuations seen on the output, should be seen on the input, but not direct phase for phase, but an overall wattage increase in accordance with the wattage usage.

Are you saying that the ups essentially splits the load equally across the phases on input even though only one phase shows a significant change on the output? Please explain a little more. I think I'm misunderstanding you.

 

[LarryFine]

Are you saying that the ups essentially splits the load equally across the phases on input even though only one phase shows a significant change on the output?

That's what he's saying. There is no direct phase-for-phase link between input and output like a 3-ph transformer would have. The output is totally independent.

Remember, the UPS can supply 3-ph power even away from the building. There is no need for a synchronizing reference from the utility supply.

The UPS converts incoming power to a single DC voltage, which charges the battery and supplies electronic circuitry which creates a new 3-phase output.

So yes, a load on a single line output is shared among the three phases of the input. Sort of a self-balancing act.

 

[massfd]

If you go looking for your load that is causing the 10 amp jump on L-1 I bet it will be a good size Laser printer or Copier. They have heaters that cycle to keep the fuser warm. On cycle time is about 5 to 10 seconds every 2 minutes or so. The heater cycles any time the unit is on and in standby mode, also runs when printing.