AC-DC rocks the boat for LM-79 and LED replacement lamps

Oct. 8, 2010
The LM-79 testing protocol allows MR16 LED lamps to be tested with either an AC or DC supply, but there can be a large performance difference between the two. Tim Whitaker and Brian Owen report.
[Update: since this article was written, Lighting Facts has responded to the LED-based MR16 test issues described below.]

An apparent loophole has come to light in the way the performance of certain LED lighting products is measured according to the LM-79-08 testing protocol “Electrical and Photometric Measurements of Solid-State Lighting Products.” This also has repercussions for programs such as Lighting Facts, Energy Star and others, which rely on LM-79 measurements.

The loophole concerns LED lamps that are able to operate directly from either an AC or a DC supply, particularly MR16 lamps. However, there can be a considerable difference in performance depending on whether an AC or a DC supply is used.

LM-79 testing allows the use of either AC or DC supply. The standard says that “the SSL product under test shall be operated at the rated voltage (AC or DC) according to the specification of the SSL product for its normal use.”

The loophole came to light when LEDs Magazine was sent test data for an MR16 lamp, bought in a major retail store in the US, that carries a Lighting Facts label. The label states that the light output is 295 lm, the power is 6W, and the efficacy is 49 lm/W.

When the MR16 was tested at 12V DC, the performance was reasonably close to the stated figures on the Lighting Facts label, with light output of 266 lm at 6W, and 44 lm/W efficacy. However, when tested on a 12V AC supply, the output was only 185 lm, the power was 4W and the efficacy was 45 lm/W.

In other words, the light output measured at 12V AC was only 70% of the value when the lamp was operated at 12V DC.

Crucially, in North America, the vast majority of MR16 lamps in “real world” situations are connected to a 12V AC supply powered by either an electronic or magnetic transformer.

This means it is highly questionable whether presenting DC data for an MR16 can fit with the “normal use” requirement in LM-79.

Clearly, a consumer relying on an LM-79 measurement (or a Lighting Facts label) that, unknown to them, shows data from DC testing would have an unwelcome surprise when they connected their MR16 to a 12V AC supply.

Lighting Facts and Energy Star

So how will this affect programs such as Lighting Facts? A spokesperson said that Lighting Facts “will be taking immediate actions to prevent potential confusion in the future” over this issue. “LM-79 states that products can be tested to AC or DC,” added the spokesperson. “It does also state that they must be tested to the product’s normal use. LF accepts products tested under AC or DC, since it is covered under the scope of LM-79.”

Lighting Facts’ position is that the product in question was tested to LM-79, so the information on the Lighting Facts label is accurate according to that test. “However,” said the spokesperson, “we do see that those values could be confusing if the distinction about testing at AC or DC is not clearly disclosed. Going forward, the Lighting Facts label on any MR11 or MR16 product will indicate whether the product is tested at AC or DC so that there is clarity to that element.”

Another potential area of concern is the Energy Star specification for Integral LED Lamps.

It is not clear at present how the AC-DC discrepancy may affect the requirements for minimum center-beam intensity for directional lamps, or power factor.

Alex Baker, Lighting Program Manager for Energy Star, told LEDs Magazine that the EPA will look into the issue. “EPA will thoroughly review the matter on the Tuesday following the Columbus Day holiday weekend,” he said. “Until then, qualification of pending low-voltage integral LED lamp submittals remains on hold.”

In news related to the Energy Star integral LED lamp specification, Lighting Science Group recently said that it had received the industry's first Energy Star label for an LED bulb. This is a PAR38 lamp that runs from 120V AC only.

Hero lamps

Another issue with LM-79, highlighted by several industry experts, is that the performance on spec sheets and LM-79 reports do not always match the actual product performance when customers get their hands on the product.

The problem appears to be that some manufacturers send their best-performing LED lamps (or other products) for LM-79 testing, and then apply this “hero lamp” data to their entire product line. We can speculate that some lamp makers may use the best-performing (and most expensive) LED bins in the lamp submitted for LM-79, but then use lesser bins in volume production.

If manufacturers exploit LM-79 with their “hero lamps” then this will affect the Lighting Facts label and the ability of the products to qualify for rebates. And the only form of market surveillance is the CALiPER testing program, which tends to lag several months behind the product introduction curve.