Why a GC or GC-MS Should Not Be Backed Up on a UPS

Agilent does not recommend a UPS (uninterruptible power supply) to provide power to our GC and GCMS systems. Shown below is a letter describing why this is the case. Essentially, the GC oven turns on and off rapidly, creating a load that switches quickly between approximately 100 and 2000 Watts, taxing the ability of the UPS to keep up.


Here is some further insight:

The problem for a UPS, or any other power delivery system, is the GC oven.

The GC oven shroud for 120 volts, is about 9.25 ohms, and draws about 13 amps.  (1550 watts)


The 120 volts, (or 230 volts) to the GC is sinusoidal, A.C. (Alternating Current) power.


The oven control is through a triac and is turned on, on a half cycle, by half cycle, basis.

On each zero crossing cycle start, the triac turns "on" if oven heating is required.

On each zero crossing cycle end, the triac turns "off".  (And immediately back on again, if heating is required.)

This triac (oven) on, or not, happens 100 (50 Hz) or 120 (60 Hz)  times a second.


Because of this, on a fully configured instrument, one half cycle may be pulling a few hundred watts

from the line voltage, and the next cycle is pulling full power of up to 2250 watts.


Because of this varying load it can cause a lot of problems to the external power delivery system.


UPS systems are particularly sensitive to this varying load. For example, some UPS systems use battery

back up (direct current) and feed the D.C. into an oscillator creating a square wave, which is then

rounded to resemble a sine wave.  The key word here is RESEMBLE a sine wave.

These electronic UPS systems are particularly susceptible to varying loads.


The UPS manufacturer needs to be aware of the varying loads and over compensate accordingly.

The "best" kind of UPS would be a "BIG" motor / alternator with a big heavy metal flywheel.

This would be a scaled down version of the utility company.



Was this helpful?