Dedicated Line, Back to the Service Entrance

Occasionally, when power quality problems are related to other loads, to excessive impedance, or to other factors, a “dedicated line” is a reasonable solution.

Other times, when the power issues are related to the utility or the source – not so much. In fact, sometimes power issues get worse when a sensitive load is coupled directly to the source, with no other loads to attenuate higher frequencies.

Here are two snapshot waveforms from a recent site, one from a shared distribution panel, the other from a dedicated line. Can you tell which is which?

Dedicated Run Snapshot 6Mar15

Original Fee Snapshot 3Mar15

The harmonics are visibly worse in the first waveform (3/6/15) – which was a dedicated feed. The shared feeder (3/3/15) still had harmonics but they were attenuated somewhat, presumably by other loads.

The problem here is at the service entrance or facility level – not restricted to a single distribution or sub-panel. The “dedicated line” actually made problems worse!


Hello Datawave, My Old Friend (and Goodbye)

My first introduction to the Liebert Datawave came back in the late 80s – an over-ambitious power conditioner salesman had pushed four of these behemoths, 100 KVA each, to feed two biplane Cath Labs. The X-Ray generator technology at the time (pre-power converter) was a linear power supply that was calibrated based on a steady, predictable voltage drop, and the Datawave output characteristics (attempting to regulate) were anything but stable. I suggested they switch the Datawave devices to bypass, the imaging system started to operate properly – and the facility and the power conditioning sales person were left with the problem of what to do with four used power conditioners.

I used to joke in seminars that I’ve been responsible for the removal of more power conditioners than for the installation of same, and that particular site was foremost in my memory, although there have been many others over the years.

I recently came across a CT Scanner site powered by a Liebert Datawave, circa 1986. Amazingly, nearly 30 years later, the Datawave was still working, although time (and more specifically, the aging and changing of internal capacitance) had taken its toll – with high voltage distortion and load related sags and swells. I’ve gotten pretty good over the years at spotting these devices by simply looking at the power analyzer data; often times, a medical imaging device is upgraded and the installers have no idea what is electrically upstream.


Distorted voltage waveforms in stand-by / low load – typical of medical imaging system.


THD exceeded 5% on all phases, and 3rd, 5th, and 11th harmonics were over 3% on some phases.


Step change in load current, typical of a medical imaging load, causes moderate (+/-10%) variations in voltage.

In this case, the Liebert Datawave was still doing its job, although there was clearly some performance degradation over the years – but the customer wanted to reclaim some space (the footprint of the Datawave was impressive, nearly the size of a minivan). A static, double conversion power conditioner would provide a similar level of protection in less than 25% of the footprint, and with a significant energy savings (improved efficiency, decreased standby load, and less thermal load on the HVAC system) to boot.

The Datwave was a pretty amazing, innovative device back in the day (ferro-resonant power conditioners have always seemed more magic than science), but usually not the best fit for a medical imaging application. There are not too many of these beasts kicking around anymore – so I am perhaps saying farewell to an old nemesis for the last time.