X-Mailer: Mozilla 4.75 [en] (Win98; U)
Subject: Re: Parallel LEDs calculation
References: <3D7B8B88.4ED5A338@bellatlantic.net> <3D7CA990.F0432D5C@rsccd.org> <3D7D6315.CE90B627@bellatlantic.net>
Date: Wed, 11 Sep 2002 04:40:05 GMT
NNTP-Posting-Date: Wed, 11 Sep 2002 00:40:05 EDT
"Watson A.Name" wrote:
> In article <3D7D6315.CE90B627@bellatlantic.net>, firstname.lastname@example.org
> talked about...
> > Lizard Blizzard wrote:
> > >
> > > Martin Riddle wrote:
> > >
> > > > Actuall you should derate the resistors according to temperature.
> > > > If this is a 85c enviroment, then the resistors should be derated.
> > > > Check the ohmite web site, the info is in their catalogs.
> > >
> > > Of course derating the resistors still doesn't take into account the
> > > possibility that the series string of LEDs might get shorted, and the
> > > full voltage would be across the resistor(s).
> > Anything is possible.
> Exactly, and that's the whole point.
And there is a reasonable point, beyond which you don't go,
unless willing to bear additional cost. You DON'T build
a circuit that will be bulletproof against every possibility.
That said, your point below about chafed cables shorting out
is a very real possibility. How it is handled is a matter
of choice, not addressed in my post. I thought Blizzard
was indicating the LED string shorting, not the cable
shorting, so I was puzzled. You cleared it up for me.
Frankly, however, I can't envision choosing to run 100
wires to the tail light. I'd include the resistors in
the taillight. I suppose it falls under "anything is
possible" that there might not be enough room in a
standard tail light that can display 500 LEDs and yet
can not accommodate 100 or 200 resistors behind the LEDs.
> Suppose the topology of the circuit
> is that the LEds are located at the tail light and the resistors are
> located some distance from them, upstream in the wiring. If the wiring is
> subject to an automobile environment, there could be a short in the wiring
> to the frame and the resistors would then be exposed to the full battery
> And as you say below, the probability of the LEDs shorting is very low, so
> if the resistors and LEDs are on the same circuit board, the probability of
> the resistors being exposed to the full battery voltage would be much
> In either case, it would be prudent to use a fuse of the right amperage
> upstream to protect the circuit. We've all seen people driving around
> with a broken taillight, probably caused by backing up too close to the
> barriers around the parking space, so these kinds of things aren't all that
> As for maximizing reliability in your last paragraph below, I would guess
> that the regular incandescent lamp is not all that reliable, lasting only a
> thousand hours or so. But if the light, whatever it's made of, causes a
> fire, then I'd say that the whole product would be recalled, and that could
> cost a whole lot of money!
I wasn't thinking of incandescent lamps. LEDs are far more
reliable. I was talking about maximizing reliability of
the LED circuit. For example, I mentioned he'd get better
reliability by using 2 470 ohm 1/2 watt resistors in parallel.
The cost is an additional part per string, the expense of the
additional part, the extra space it takes, and the extra
time to build the circuit. As I envisioned things, the
minimum requirement for a reasonably reliable and well operating
circuit required a single 1/2 watt resistor. That already
includes a 700% derating (which Mr. Riddle does not seem to grasp)
for the resistor. Adding a second resistor increases both
resistor and LED reliability by doubling the wattage and
reducing the LED current ~5%. One would have to decide whether
the increased reliability was worth the increased cost.
The same is true for changing to fusible or flameproof
resistors or adding 100 fuses. One has to decide where to draw
the line, or whether a different approach is called for.
Your point about a product that is capable of causing a fire
is right on target, and carries a lot of weight in the decision
point. As I see it, the possibility of fire would dictate that
the topology would be single wire supply to the light, fused at 3
amps at the source, with the 100 resistors and 500 LEDs all together
at the tail light. Of course if this thing was being built
for the military or other outfit that requires extreme
reliability, your "cost budget" is increased greatly,
and you could design the thing with flameproof 5 watt resistors,
fusible links, and a built in automatic fire extinguisher. :-)
> > But do you really think 5 LEDs in a string
> > are each going to short? Regarding LED failure in general, open
> > is a more likely failure than shorted. Regarding the string,
> > if one shorts, the probably of failing open for one of the
> > remaining good LEDs goes up, and increases with each succeeding
> > shorted LED, afaik. Please correct me if I am wrong.
> > > If so, then plan for the
> > > fault by either using higher power resistors (E^2/R) or else flameproof
> > > fusible resistors or a fuse.
> > >
> > You could do that, or use the approach I mentioned:
> > two 1/2 watt resistors. Or use 2 1 watt units if you feel
> > the need. But at some point there must be an engineering
> > decision as to how far you want to go. You can maximize
> > reliability at the expense of parts count, expense, volume,
> > cost, whatever.
> > >
> > [snip]
> > >
> > > --
> My email address is whitelisted. *All* email sent to it
> goes directly to the trash unless you put NOSPAM in the
> Subject: line. alondra101 hotmail.com
> Don't be ripped off by the big book dealers. Go to the URL
> that will give you a choice and save you money(up to half).
> http://www.everybookstore.com You'll be glad you did!
> Just when you thought you had all this figured out, the gov't
> changed it: http://physics.nist.gov/cuu/Units/binary.html