Why not UPS with GC ?

[thohry]

As I know, in some locations, power is not stable and can be shut-off suddenly. I think it's safer to use a UPS so that the user can shut down the instrument properly. But I just read in the safety manual of Agilent 7890 that:

If the area where the GC is located suddenly looses power, an unsafe condition
can result if the GC remains powered on. Do not use the GC with a UPS.

I am puzzled ! Who can explain this ?

Thank you.

ps. The safety manual can be googled with the PN: G3430-90013

[Jetjamnong]

Really? why the Agilent engineer installed all my GC/MSD with UPS. maybe I think this have a wrong words "with" or "without" or it does not really need a UPS ?

Jetjamnong

[dblux_]

...

If the area where the GC is located suddenly looses power, an unsafe condition
can result if the GC remains powered on. Do not use the GC with a UPS.

IMHO only for liability reasons.
Agilent reduces it's responsibility for any damages connected with power failures.

[Peter Apps]

Seems strange to me - and it will kill sales of Agilent GCs anywhere that does not have reliable mains power.

Peter

[bluejay]

I asked Agilent Tech Support once about a UPS on a 6890. They said that they did not recommend it becuase the power coming out of the UPS was not terribly "smooth" (for lack of a better word). The GC would function, but the power draw for the oven would fight with the UPS and would wear out the oven (electronics? heater? I'm not sure...) faster than normal.

[Karen01]

I had UPSs (a 5kVA one) on 6890s and 5890s for years without issue and no agilent engineer ever had an issue with it.

As long as the UPS is rated to support the draw of the GC, I can't see why it would be an issue with a quality UPS.

- Karen

[dblux_]

(...)
As long as the UPS is rated to support the draw of the GC, I can't see why it would be an issue with a quality UPS.

- Karen

Not all UPS devices are pure sine wave.
Chromatographs or rather their electronic circuits can tolerate power supplies with approximated sine waveform, but probably no deep tests were performed to be sure for reliable and stable results when supplied with AC approximated by square waveform.

That's in my opinion why Agilent stays on safe side and takes no responsibilities for functioning their GCs supplied with UPS of unknown waveform.

[Chenosuke]

I received this email from my distributor some time ago.

Agilent Technologies does not recommend the use of a UPS or other isolation device to supply power to
a GC, GC-MS, or ICP-MS system. The Agilent system creates a highly varying load to these isolation
devices. The oven of a GC, for example, is controlled by either turning the power on or off each cycle of
the line. This creates a varying load condition on the UPS or isolation device of approximately 100 Watts
when the oven heater is off and 2000 Watts when the oven heater is on. Since the heater may be
powered off for one cycle and powered on for the next cycle of the line voltage, this creates a highly
repetitive level of load variation. Many isolation devices require several line cycles to recover from high
load variations. These isolation devices do not have sufficient damping and capacity to correct for the
level and frequency of load variation created by the Agilent devices.
Some isolation devices create undamped oscillations when subjected to repeated cycle-by-cycle, highlevel
load variations. These units will actually create very high levels of transient voltage supplied to the
load. These undamped oscillations often exceed the specifications for the Agilent system or the isolation
device. This may cause early or unexpected failure of the Agilent Technologies system, the isolation
device, or both.
Additionally, the high variation in load is created at a 50/60 Hz rate. Use of UPS or other isolation
devices utilizing Ferro-resonant transformers, regulators, or supplies is specifically NOT
RECOMMENDED since the large load variations at the resonant frequency of the transformer may cause
the transformer to operate unpredictably. This operation often causes the Ferro-resonant transformer to
overheat or malfunction.
Please contact the manufacturer of the UPS or other isolation device if you wish to use these to supply
power to Agilent Technologies GC, GC-MS, or ICP-MS system. Inform the isolation device manufacturer
of the varying cycle-by-cycle load behavior of the Agilent system to obtain a recommendation from the
manufacturer for the proper size, capacity, and type of UPS or other isolation device.

I know of a 7890 system that operates on a UPS not far from where I work that has blown 2 FID circuit boards in just under 2 years. That UPS could probably be an area of concern. I guess the take-home point from this is to inform the UPS manufacturer/distributor about your equipment when purchasing one.