A closer look at the Philips SlimStyle LED light bulb driver

In which an important lesson is learned: The presence of a large transformer-like component does not necessarily mean an LED  bulb is isolated.

I was so charmed at the clever heat-dissipating pc board design in the new Philips SlimStyle LED bulb that I breezed right Philips SlimLine LED replacement bulbover the electronic circuit used in the driver design. My first look showed a large magnetic component with multiple connector pins and I quickly assumed that it was an isolation transformer, which would have made this the first new LED bulb design I’ve seen recently that was isolated. (Because of their savings of space and components, non-isolated designs have quickly taken over the replacement bulb design wins. See, Isolated or non-isolated LED drivers for light bulbs?)

Fortunately for me, an astute reader pointed out that this is not always the case, and you can’t judge a transformer by its yellow wrapping. By looking at the photos, the reader pointed out, you can tell that the magnetic component has insufficient creepage distance to satisfy safety requirements for a non-isolated design.

(Creepage distance is the distance between the primary and secondary windings of a transformer. For example, in a typical stand-alone switch-mode power supply, a rule of thumb is to allow an 8-mm creepage distance between primary and secondary circuits.)

Custom magnetic - Philips SlimStyle LED bulb

So if this is not an isolation transformer, what is it, and what kind of a driver topology does the bulb use?

Both flyback and non-isolated buck boost designs use a primary side-controller. They both require a custom magnetic, but rather than being a transformer, with a primary and secondary winding used for voltage transformation, it’s a coupled inductor: There are still two windings, but one is used for inductance, and one for bias.

Taking a closer look at the backside of the pc board, we can see that the output of the power FET Q1 goes directly to D8, the output of which goes to the yellow wire that powers the LED string, pointing to a non-isolated buck-boost design.

Output of switching FET drives LEDs

Some further information about the LED board: The voltage applied to the 26 LEDs measures just over 78V, indicating the LEDs are in a series string (26 LEDs x 3V/LED = 78V).

One more thing: In the photo below  you can see the tiny component next to the LED at the top of the photo: This is a resistive temperature sensor that throttles back the current to the LEDs in case of overheating – a nice feature that adds to the bulb’s reliability.

Philips SlimStyle LED board

Just to reiterate from my initial review: The Philips SlimStyle is an impressive and clever bulb design.

Comments

  1. Jay Hughes says:

    Why are there what appear to be tracks going to only SOME of the LEDs? They also appear to terminate around the black object.

    Schematics would be nice in your ‘tear-downs’.

  2. Jay, what appear to be traces going to each LED are actually gaps between the large copper pads that act as heat sinks for each LED. There are two traces that go to the temp sensor — the black object — and these traces go to the two gray wires. So, the yellow and gray wires marked + and – power the LEDs, and the two gray wires carry the temperature sense signal.

  3. How about some RFI/EMI plots from DC up to a couple of GHz? Just this past week in Springfield Oregon the FCC filed a complaint against a private business for a LED light fixture that was interfering with Public safety 700/800 MHz and Cellular communications several miles away from his business.

    The owner of the LED lights was given the choice of a $16000 dollar a day fine which came out to close to $112000 dollars total over the length of time that the FCC monitored the interference or no fine if shut the lights off then and there when the FCC confronted him.

    The only rational choice of course would be to shut the fixtures off and toss them out but as the FCC agent related the story, the owner complained and told the PORTLAND FCC agents in charge that he needed his lighting and it wasn’t right for them to shut his lights off and the FCC agent told him tough luck you either pay the fines or shut it off.

    I am seeing more of this interference to radios due to these new LED lights and it’s nice to see the FCC finally cracking down now.