原创 The union of AC line and CFL/LED lamp

Talk about parametric sensitivity: those are fairly significant exponential-driven variations around a nominal operating voltage setpoint (the Wikipedia entry has many reputable citations and links on this).

I’m sensitive to this problem because I have a small Datel (now Murata/Datel) AC-line voltage meter always in place on one of my outlets (Figure 1). It’s handy because it’s small, unobtrusive, and has no settings or controls; it does only thing, and that is to read out the AC voltage. Mine line is usually at 123 to 135 VAC, thought I have seen it run as high as 128 to 129 VAC, and as low as 118 to 119 VAC, and both extremes are bad, for many reasons. In both the high and low cases, my local electric company blamed it on a problem with my AC line’s neutral connection at the meter box, not their source. (I rejected that rationalization but still paid for inspection by an electrician, who verified the integrity of my power feeds.)

This Datel AC voltage cube is a handy way to keep an eye on the AC line voltage, it truly is a "plug in and play" unit (note that the segments of the digits are starting to "wear out" and dim after 20+ years).

 

Still, the low and high-line scenarios made me think about the effect of line voltage on compact fluorescent lamps (CFLs) and light-emitting diode lamps (LEDs), Figure 2. In contrast to the incandescent lamp, these light sources do not run directly from the AC line, of course, as there is an intermediate AC/DC driver for each of them. I’ve seen many of the reference designs from reputable vendors of the ICs which are at the core of these lamp drivers, and most specify a fairly wide AC-input range over which the driver design will maintain tight output regulation, typically to about 1%.

Three sources, with three very different power issues: the incandescent bulb operates directly from the AC line; the CFL needs an AC/DC voltage driver; while the LED lamp needs an AC/DC current driver.

 

But wait: isn’t 1% roughly the same variation that the incandescent bulb sees when the AC line is well-managed? That brings me to the core question: what is the performance and lifetime sensitivity of both CFL and LED bulbs themselves (the bulb is the actual illuminating source) to variations in their DC drive? I haven’t been able to find any solid data on this, and I doubt that any would be meaningful anyway. The reason is that it is so dependent on the specific manufacturer of the bulbs, which varies from lamp vendor to vendor and may even vary with a single vendor's between production runs.

There's another thing which bothers me about ratings for CFL: and LEDs. Since their performance is heavily dependent on the performance of the driver, consistency in the manufacture of that driver is critical. Yet we all have read credible stories of substandard and even counterfeit parts being used (knowingly or not) in circuits of this type. The overall unit may appear to even function to spec, at least for a while, but its tolerance for input-voltage variations and temperature-induced changes may be much less than the first-run batches which used quality parts, were fully tested, and then submitted for certification.

Do you have any insight into the performance of mass-market CFL and LED divers with respect to input-line variations or driver output? Have you had any experience with batch and unit variations among what should have been identical units from a given vendor?