Topband: Chassis Bonding

[Jim Brown]

Thanks for the update from someone still working in that part of 
industry.  Yes, PF correction was mandated in EU something like 10 years 
ago. It's good news if this ripples back to our continent, even if our 
electrical codes haven't addressed this issue. .

That said, as long as rectifier-capacitor input power supplies dominate 
the landscape, we'll have those harmonics to deal with. BTW -- for those 
following the discussion, the harmonics are generated because current 
flows in pulses at the positive and negative peaks of the AC waveform to 
recharge the input filter capacitor. A switching power supply simply 
takes the resulting DC, generates square waves at a higher frequency 
(typically in the 10-20 kHz range) that's easier to filter.

The harmonic problem is NOT limited to lighting ballasts, data centers, 
etc. Nearly 20 years ago, a major Chicago broadcaster, WFLD-TV, had a 
very serious encounter with the high neutral currents that these 
harmonics can produce (nearly double the current in one phase of a well 
balanced system) and hosted a meeting of our local SBE chapter to talk 
about it. Their neutral hardware got so hot that they nearly had to shut 
down and go off the air.

73, Jim K9YC

On 1/29/2014 12:52 PM, Bill Wichers wrote:
> Triplen harmonics, in most facilities, are caused primarily by lighting ballasts and "electronic" (rectifier-input) power supplies. The newer switching power supplies have a power factor controller in the first "stage" that acts something like a current regulator preceding the actual switching power supply. You can look up data sheets for these devices. The result is a .95 or better power factor in most cases, and almost no generation of triplen harmonics on the neutral. My understanding is that the EU requires power factor correction in computer power supplies and since the manufacturers want to have one power supply to use globally ("100 - 250 volts, 50/60 Hz", we've all seen this:-), they just build to the most stringent standard and keep their volume costs down with the single part.
>
> The result of the corrected power supplies in datacenters is that we don't need to use K-rated transformers (rated for high harmonic content) anymore which saves a*lot*  of money. I have found almost all of the "commercial" equipment (network gear, routers, rack-mount servers) incorporate PFC. Many, but not yet all, of the desktop computer power supplies are also going this route.
>
> Newer electronic ballasts are also incorporating PFC. It's all part of an effort to keep electrical costs down. It makes a difference in large facilities, especially when you consider the long lifetime of most of the commercial equipment involved.