Re: High brightness white LEDs damaged by custom switcher

is device to device parameter variations being accounted for?

most of these LED's are not anywhere near being matched pairs/triples or
other, unless you get the manufacturer to custom mount them from the same
die and then do a parametric test on each and then sort them and mark them,
(Very costly)

even clusters are assembled from random devices!

MIL specs may offer you some leeway, but even controlled groups will be bulk
tested only for basic params.

if spiked current is failing some devices, they may have beeen borderline to
begin with, as most will just heat up and dim before catastrophic failure
occurs. usually over time.

your blown bonds and similar symptoms tells me that you had EXTREME current.

rise time between units may force some to go into failure befor the adjacent
units absorb the available current also.

as with all design, the simplest is the best. trying to parlay your
education of PICs and neatsy circuits willl yield expensive and hard to
manufacture systems.

take some lessons from the chinese,tiawanese and others, cheap simple and
basic.

stay within some easy set standard limitations and you may have good
results.


"Paul E. Schoen" <pstech@xxxxxxxxx> wrote in message
news:463ad110$0$20580$ecde5a14@xxxxxxxxxxxxxxxxxxx
Some of you may recall my discussions about a 12 VDC powered switching
converter to drive strings of 7 or 13 high power white LEDs, in posts from
1/5/07 (transients), 2/22/07 (Inductor gaps), and 3/10/07 (TC1410 driver
problems). We built the circuit, found a few problems, then tested it and
it seemed OK. Basically this is a switching boost regulator using a 10 uH
inductor and a PIC with PWM at 100 kHz. I did a lot of LTSpice simulations
as well.

The circuit is somewhat unique in that the PIC is programmed to produce
alternate high and low brightness modes, at about 700 and 100 mA, when
power is applied and removed. The PWM is controlled by the voltage on a 1
ohm sense resistor, and it is designed to ramp up slowly and cut back
quickly. There are also undervoltage input and overvoltage output sensing
circuits to shut down the PWM until the next power on reset.

My friend had three different 7-LED clusters (made by Cree), of low,
medium, and high brightness. While testing the medium-brightness cluster,
he noticed a less than normal difference in brightness, and found that two
of the LEDs were not lit. They tested open, but apparently the voltage was
enough to bridge the gap and provide a current path for the remaining 5 in
series.

The manufacturer examined the LEDs and stated: "#1 appeared to have an
evidence of damaged chip which looked to be a burnt metal traces around
the
bond pads, it also had a blown wire bond on the Anode side which had an
appearance of a fused wire. The lamp #4 had the blown off bond pads on the
chip side." They concluded it was a high current surge.

My recommendation to my friend was to perform a deliberate destructive
test
on one or two lamps to see what level of current would cause such damage.
I
think he had previously tried up to 1.5 or 2 amps for a few seconds
without
sign of damage, although possibly not with these exact model LEDs.

Without going into more detail for the sake of brevity, I would like to
discuss possible scenarios, testing suggestions, and design improvements
to
minimize the risk of more failures in the field. Terry Given was kind
enough to analyze my circuit and layout, with many suggestions for
improvement, and I plan to implement them in the next iteration of the
design. However, someone else was adamant that a PIC would never be as
reliable as a pure analog design, and suggested an off-the shelf Linear
Tech or other part, which would involve a complete redesign and loss of
the
dual brightness feature.

Thanks,

Paul




.