|
The objective is to make a DIY replacement for fluorescent
light tubes using SMD5630 LED strips. SMD5630 LED strips are
available from the Internet stores for about US $10 for five
meters of strip, 60 LEDs per meter, and require 12 Volts at
about 1 Amp per meter. They produce about the same amount of
light, per meter, as a single fluorescent light tube. (For
replacing fluorescent shop lights, see the Addendum.)
SMD5630 LED strips:
SMD5630 LED strip colors:
- Cool White: 8000K to 10000K, (probably not advisable
except for special purpose applications-the light has a blue
tinge color)
- Daylight White: 6000K to 7000K
- Nature White: 4000K to 4500K
- Warm White: 2800K to 3200K
Daylight White is considered best by most commercial users,
Nature White considered best for residential.
SMD5630 LED strip characteristics:
72 / 300 = 0.24W / LED60 * 0.24 W/led = 14.4 W/m = 1.2A/m @ 12
V60 * 24lm/led = 1440
lm/m
By comparison:
- For an example kitchen, a single fluorescent light, (one
of many):
2250LM/3ft =
2461LM/M, (Philips Alto T12, IBC 4667720559,
819465,) and assuming 50%, (instead of 33%,)
2461 / 2 = 1231LM/M, (because
lumen specification is total around the tube, and only half
the tube's light emits down.)
- Or one 1m strip of 60 led/m SMD5630 stip would be
equivilent to a Philips Alto T12 fluorescent tube in the
example kitchen.
Note that PVC tubing is considered non-flamable and
self-extinguishing: PVC
(Polyvinal Chloride) and is not conductive, which may be
an attractive alternative for mounting SMD 5630 strips.
Each strip, for replacing each Philips Alto T12 tube, would
require 1.2A @ 12V. 2.4A for dual Philips Alto T12 tube
replacement, which is a 2X saftey factor of a 5A power supply
for the two tube replacement.
The SMD5630 LED strip, 60 LED/m, White, 78237, 1599803,
BB11204, S0131-02-5*/5630 300 LEDS, made in China, probably
water proof, was tested with a 10A 12V power supply and the
current into the power supply measured with a "Kill A Watt"
meter:
0.67 A * 115 V = 77.05
W0.67 PF * 77.05 W = 51.6
W
Since the SMD5630 LED strip is DC, the power supply is
dissipating about 77 - 50 = 27
W, or the efficiency is 50 / 77
= 65%.
The SMD5630 LED strip color is more white than the Feit
Electric Performance LED, Dimmable A19, 800 lumens, 25,000
life, 60W Replacement, (13.5 W power consumption,) Omni
Directional, Natural Daylight, 5,000K, or, Cree
SA19-08150MDFD-12DE26-1-14 Led 60W Replacement A19 Daylight
(5000K) Dimmable Light Bulb 820 Lumens 8.5W 25,000 Hours High
Color Rendering: 85+. The color is PROBABLY 6000K to 7000K
daylight white.
Note that 1m of the SMD5630 LED strip would require
77.05 / 5 = 15.41 W, or 1.28A. A
dual strip would require 2.57A, almost a safety factor of 2
for a 12V 5A power supply.
Note that there are 2 X 5m rolls of water proof SMD5630 LED
strip, 60 LED/m, white, in the author's existing inventory,
and each is made up of 4 to 6 pieces, soldered together, and
will suffice for a test case of mounting on PVC tubing, with a
12 V 5A power supply for the example kitchen fluorescents.
If the example kitchen light box, for example, is 48" in
width, and the opening at the bottom is 43". One meter is
39.37", which would leave 1.815" at each end if centered,
assuming the SMD5630 LED strip is exactly one meter long, and
the PVC extends to the internal length of the light box.
The SMD5630 LED strip has 60 LEDs per meter, cuttable every
3 LEDs, or twenty 3 LED sections. Each section is
39.37 / 20 = 1.9685" long. To
extend the entire length of the light box opening, 2 sections
would be added, for a total length of 39.37 + 2
* 1.9685 = 43.307", and the strip length would
be 43.307 / 39.37 = 1.1
meter.
The PVC tube will be custom fit, and be about
48 - 2 * 3 / 4 = 46.5" long.
If two SMD5630 LED strips are to be mounted on the PVC
tube, they should be mounted at +/- 60 degrees from the center
of the PVC tube, (i.e., 120 degrees apart from the center of
the tube.) Meaning, straight down would be 60 degrees off
axis from each strip, which is the half power plane of the
strip, and when added together, is a power of unity. This
means that full power radius of the strips would be 120
degrees, and the half power radius would be 240 degrees.
Setting:
x = 60;y = 2;d = 48.0;s = 39.37 / 60.0;
in isotropic.array.py,
yields the limits of light intensity under the SMD5630 LED
strips:
- 100.89533186228801044024
- 113.86642719083893546286
or about +/- 7% over the length of the strips to a 4'
distant work surface.
The PVC tube and SMD5630 LED strips should be mounted as
close as possible to the diffuser plastic.
A one meter of 60 SMD5630 LED strip water proof LEDs was
cut and a 12V 5A power supply, with an amp meter in series
with the positive +12V DC terminal, and the power supply was
in series with the Kill A Watt meter, which was used to
illuminate the one meter strip, and measurements taken:
- From the Kill A Watt meter, (AC measurements):
- 116.2 VAC
- 0.25A
- 16W
- 28VA
- 0.58 PF
- From the amp meter, (DC measurements):
- A digital Lux meter was used to measure the light
output of strip at approximately 1 meter distance from the
strip, (in a dark room):
- The digital Lux meter was used to measure the light
output of 8 X Philips F40T12/DX/ALTO, 2325 Lumens, each,
24,000 hour life, color temperature 6,500K at
approximately 1 meter distance from two tubes, in the
center of the tubes, (not a reliable measurement since all
8 tubes were on, and the tubes are old-but seems a
"representative" number):
- The ambient temperature was 82.5F, and an IR
thermometer used to measure the temperature of the SMD5630
LED strip, (after on for 30 minutes):
- 115F at each LED
- 107F interstitial between the LEDs
Note the free air temperature rise of 115 -
82.5 = 32.5F.
The measurements of the F40T12/DX/ALTO are consistent with
previous measurements, above, and the electrical
characteristics were consistent as calculated, above:
- One Philips F40T12/DX/ALTO, 2325 Lumens, each, should be
replaced with one strip of SMD5630 LED strip, of equal
length:
- A 3 foot length of SMD5630 LED strip will require 1A
for each strip.
- A 4 foot length of SMD5630 LEd strip will require
1.22A for each strip.
Since two Philips F40T12/DX/ALTO tubes, (both in one
receptacle,) are to be replaced at once, the maximum current
will be 2.44A, (for 4' tubes,) and a safety factor of 2 will
be used requiring a 12V 5A DC power supply for 2 SMD5630 LED
strips.
Whether to use 2 PVC tubes to mount the two SMD5630 LED
strips, individually, or, one tube with the two SMD5630 LED
strips mounted at an inclusive angle of 120 degrees, (i.e.,
straight down is the half power plane of the two strips, or,
straight down has the same luminance as +/- 60 degrees from
straight down,) is yet to be determined. Note that 2 PVC tubes
would have twice the luminance, straight down, (if mounted
close together,) and would equal the luminance of the Philips
F40T12/DX/ALTO tubes-however, the half power plane would be
+/- 60 degrees, where mounting both strips on one PVC tube
would be +/- 120 degrees.
Mounting issues are yet to be determined, and will probably
have to wait for trial-and-error comparison.
Total light in the room:
- For a single 4' Philips F40T12/DX/ALTO tube, 2325
Lumens, one meter of the tube would produce
2325 * (39.37 / 48) = 1907 lm /
meter, and the light is emitted equally in
all directions from the axis of the tube, (only about half
will be usable since the light emitted up will not
contribute to lighting the room.)
- For three strips of SMD5630 LEDs, arranged equal
distance around the circumference of a 1 meter length of PVC
tubing, the total light would be
3 * 60 * 24 =
4320 lm, emitting equally in all directions,
since 360 / 3 = 120 degrees,
and 120 degrees is the half power plane along the axis of
the PVC tube, which has half power from one SMD5630 LED
strip, and half from the adjacent strip, on either
side.
- For a fair comparison, assume both the F40T12/DX/ALTO
tube and the PVC tube with 3 SMD5630 LED strips have 120
degrees of light blocked, (i.e., the shroud on the
F40T12/DX/ALTO tube blocks 120 degrees of light emitted from
the tube, and one strip of the SMD5630 LEDs is omitted,)
then the F40T12/DX/ALTO would produce
(2 / 3)
* 1907 = 1271 lm, and, the SMD5630 LEDs
would produce (2 / 3) * 4320 = 2880
lm, emitted light into the room. Two
F40T12/DX/ALTO tubes would produce 2 * 1271 =
2542 lm, so, two F40T12/DX/ALTO tubes would
be about equal two strips of SMD5630 LEDs arranged at 60
degrees off of straight down on the PVC tube. Note that 120
degrees of emitted light from both the two F40T12/DX/ALTO
tubes and three SMD5630 LED strips, (actually, with one
strip removed,) is "blocked," and the two F40T12/DX/ALTO
tubes produce a total light of 2542 lm, and, the two SMD5630
LED strips produce 2880 lm, by comparison of emitted light
in the room, about the same. But it is not clear how much of
the light, 240 - 180 = 60
degrees past horizontal, on each, is useful.
- This comparison is actually an "apples and oranges"
comparison since it is not clear how much emitted light
produced by one F40T12/DX/ALTO tube is reduced by the
proximity of the other F40T12/DX/ALTO tube, (the assumption
is zero.)
- Additionally, this comparison assumes that the light
emitted by the two SMD5630 LED strips beyond +/- 60 degrees,
(the half power plane of the two SMD5630 LED strips,) is
useful in illuminating the room, (the assumption is it is
useful.)
- The two assumptions tend to cancel each other, but it is
not known how much the assumptions cancel. All that can be
said is that the estimates are probably reasonable.
- Additionally, it is not clear what the 55,000 hour life
time specification of the SMD5630 LED strips mean; 55,000
MTBF? Is it permissible that some individual Leds fail, but
the strip still works? 55,000 MTBF on EACH Led, (and if
there are 60 Leds per meter, this is not a very good
reliability for the entire strip.)
- Additionally, note the measured temperature rise of the
strip of
32F = 18C, and the
reliability of semiconductors is reduced by a factor of two
for every 10C, (i.e., whatever will fail, will fail in half
the time for every increase in temperature of 10C.) Comment:
The temperature rise measurements were made on water-proof
SMD5630 LEDs, which are covered with a water resistant
plastic sheath, possibly decreasing convectional air flow,
thus possibly increasing temperature rise; non-water-proof
may be a better alternative, except where required. Good
ventilation is advisable.
- Assuming the 55,000 hour life time is reasonable, (and
the author doesn't!) and individual Leds can fail, as long
as the strip emits light, (and the failures are ergodic,)
with a 20C temperature elevation, the MTBF would be
55,000 / 4 = 13,750
hours. Since there are two SMD5630 LED strips per PVC tube,
the MTBF for both would be 13,750 / 2 =
6,875 hours, (for either SMD5630 LED strip
to fail, whatever failure means.) 6,875 hours operating at
8 hours a day, (1 / 3 the time,) would be 20,675 hours MTBF
operational, or about 2.36 years MTBF, per two SMD5630
LED strips.
Because of these ambiguities, accessibility to replace
SMD5630 LED strips is advisable.
Power supply:
- It is not clear what the MTBF of the power supplies
are. No MTBF data is available, (an intense effort to obtain
the data resulted in failure.)
- The assumption is that the power supply was designed for
2,000 hour MTBF, at full rated power, at non-condensation,
(
0 C = 32F,) to
44C = 111F, (human skin
necrosis.) This is a common design specification for
consumer, retail, electronics. The assumption is that a
company would do AT LEAST this, (since it is the easiest
reliability requirement to meet.)
- Note that the power supply was derated by a safety
factor of 2, (the engineering tradition when one does not
know,) to operate at half the rated current. The reliability
increases with the square of the safety factor, (if the
devices is designed right,) or 8,000 operational hours MTBF
would be a reasonable estimate under the
circumstances. 8,000 hours operating at 8 hours a day, (1 /
3 the time,) would be 24,000 hours MTBF operational, or
about 2.75 years MTBF, per two SMD5630 LED strips.
- The power supply, as measured, was operating at
(
12 * 1.1) / 16 = 82.5%,
meaning it is requires 16W to produce 13.2W, or 2.8W will be
dissipated by the power supply. The two SMD5630 LED strips
dissipate 13.2W, for a total of 13.2 + 2.8 =
16 W, per two strip SMD5630 LED strip PVC
tube, and power supply. (This estimate ignores the Power
Factor, PF, measurement, and presumes the power lost by the
current phase and voltage phase is returned to the power
lines.)
- Additionally, note the measured temperature rise of the
strips of
32F = 18C, and the
reliability of power supply components is reduced by a
factor of two for every 10C, (i.e., whatever will fail, will
fail in half the time for every increase in temperature of
10C.) Good ventilation is advisable.
- Because of these ambiguities, accessibility to replace
the power supply for each PVC tube of two SMD5630 LEDs is
advisable.
Note that the calculated reliability of an MTBF of 13,750
hours for each SMD5630 LED strip, and, 8,000 hours for the
power supply for both SMD5630 LED strips, (if the power supply
fails, both SMD5630 LED strips do not produce light,) is
comparable with the 22,000 for each Philips F40T12/DX/ALTO,
2325 Lumens, each, 24,000 hour life when the reliability of
the ballast is included, (creating a failure of two
fluorescent tubes,) and both have reasonably comparable
luminescence in a room.
As a worst case estimation, the SMD5630 LED solution will
require about 3 times the remove-and-replace service as a
Philips F40T12/DX/ALTO, (ignoring ballast replacement,) which
can probably be made comparable through:
- Derating the power supply, (another factor of two would
probably suffice.)
- Increasing ventilation to lower operating temperature of
the SMD5630 LEDs and power supply. (Perhaps non-water-proof
SMD5630 LED strips may suffice.)
The MTBF estimates for the SMD5630 LED solution may have
been made under somewhat pessimistic assumptions, (how
pessimistic is not known,) but both criteria should be
considered critical design parameters.
From a cost perspective, the current price for two 4'
Philips F40T12/DX/ALTO is US $25.97, or US $13, each.
The SMD5630 LED strips cost US $9.99 through the Internet
stores for 5 meters = 16.04', or
US $2.50 for 4', and the power supply for two strips is $9.99,
or US $5.00, for each strip, for a total of US $7.50 per Led
strip, or almost half as much as the fluorescent solution.
From an operating cost perspective, each 4' F40T12/DX/ALTO
tube requires 40W. (40W * 22,000 hours = 880
KWH, and at US $0.125 per KWH is US $110 over
the life of the tube.)
Since a single SMD5630 4' requires about 16W, (including
power supply efficiency,) to produce the same mounted
luminence, the cost savings would be a factor of 2.5. cost
savings could be substantial, or US $44 for an equivalent
22,000 hours of operation.
Ignoring the reliability of the ballast, and, the issue of
a power supply failure disabling two SMD5630 LED strips, for
24,000 hours of operation, (comparable to the F40T12/DX/ALTO,)
the SMD5630 LED strip cost would be US $2.50 *
(24000 / 13750) = US $4.36, and the power
supply cost would be US ($9.99 / 2) * (24000 /
8000) = US $ 14.99. $14.99 +
$4.36 = US $19.35 in component costs for
24,000 hours of operation compared with US $13 for the
F40T12/DX/ALTO for 24,000 operation, for a US $4.35
differential for 24,000 hours of operation.
Note: There are drop in replacements, (requiring removal of
the ballast from the fluorescent fixture,) SMD tubes rated at
45,000 hours lifetime. They run about US $23, each, for a 4'
equivalent, making them more expensive than the fluorescent
they are replacing, by about a factor of 23 / 13
= 1.77. They are as efficient as SMD5630
LEDs. They are not compared here because of the high initial
cost, but note that the initial cost, amortized over the
lifetime, is competitive with the fluorescent, (45,000 hours
vs 24,000 hours, a factor of 1.875.) Each SMD replacement bulb
will replace one fluorescent bulb. This may be a viable
alternative where low maintenance is a requirement. This is
considered a backup alternative if the cost of the reliability
of the SMD5630 LED strips and associated power supply can not
be surmounted. The reliability degradation due to increasing
temperature of the SMD solution may be similar to the SMD5630
LED strip solution, but this is not known at this time.
Bottom line comparison of the SMD5630 LED strip solution
vs the F40T12/DX/ALTO:
- pro for the SMD5630 LED strip:
- Much lower cost of ownership due, primarily, to
energy efficiency.
- Lower initial cost.
- Lower cost of replacement.
- con for the SMD5630 LED strip:
- Possibly, three times as much service, but this is
addressable through adequate ventilation design, and,
power supply derating, to, probably, be competitive with
the F40T12/DX/ALTO solution.
Author's comment:
Product QA in the electronics industry is a thankless
job. QA is always in the way of moving product into
distribution. When a company becomes a world class supplier,
the QA department has bragging rights-largely through their
efforts; it couldn't have happened without them. When they
won't brag with MTBF data, one has to wonder why. The
pessimistic numbers used in the analysis were justified for
this rationale.
The example kitchen light box measured 46" internal, (the
box contained 8 3' fluorescents which were being replaced,)
and a 0.8" ID, 1.05" piece of PVC was cut to 45.5" length and
the 1 meter string of 60 SMD5630 LEDs attached to the PVC,
after the PVC was cleaned with water, then TSP solution, then
cleaned with acetone, and allowed to dry in the sun, then
sanded with fine sand paper. The string of 60 SMD5630 LED
strip was attached by laying out as straight down the PVC as
possible, under very light tension, and pressing down.
The SMD5630 LED strip was inexpensive from EBay, and had
several issues along the string-pathetic QA of many joined
sections. Photographic documentation before installation is
maintained, and available on request. This is a major
reliability issue, but was supplied by the cheapest available
vendor.
The second string of SMD5630 LEDs were attached to the PVC
pi * 1.05 / 3 = 1.1" around the
circumference,(measured with a clothes tape measure,) of the
PVC tube from the first string, (edge-to-edge
measurement.)
Measurements, two SMD5630 LED strip mounted on 1.05" OD PVC
tube, located +/- 60 degrees from vertical, (i.e., directions
aimed 120 degrees apart,) after 1 hour of operation:
- From a digital thermometer:
- The ambient temperature was 86.5F
- The temperature inside of the power supply, which
was on its side, (convectional perforations not
impeded,) 107.1F, a temperature rise of
20.6F ~ 10C, or a
reduction in MTBF by a factor of 2.
- From the Kill A Watt meter, (AC measurements):
- 112.3 VAC
- 0.46A
- 32W
- 52VA
- 0.61 PF
- From an IR thermometer:
- The PVC tube was 125F
- A specific LED was 134.8F, a temperature rise of
48.3F = 27C, or a
reduction in MTBF by a factor of about 5.
- From a Lux meter, the light intensity at one
meter:
- Vertical 204 lm
- Vertical +/- 60 degrees 204 lm
Comment after several hours of operation on a work bench:
The complaints about the adhesive on the back of the SMD5630
LED strip losing adhesion is true. The SMD5630 LED strin/PVC
sat on the bench overnight, and no adhesion problems. After
powering on for only a few hours, the adhesive was losing
adhesion in many places. The temperature of the adhesion was
about 125F.
Conclusion:
The alternatives:
Ready made LED strip light tubes, like https://hyperikon.com/product/t8t10t12-3ft-led-light-tube-14w/,
which is not ballast compatible, (i.e., the ballast will
have to be removed from the light fixture, and the fixture
rewired-which is not difficult after the first one.) The
cost is about US $17.50, and one LED strip light tube will
replace on fluorescent tube, (i.e., two tubes replaces two
fluorescent tubs in a two tube typical fixture.) It is not
clear whether Daylight White, (6,000K to 6,500K,) is
available. The lifetime is 45,000 hours per tube. In other
words, US $35 per two tube fixture. Two strips of SMD5630 LEDs, 60 LEDs per meter,
each, on one 1" OD PVC tube, about one meter, (~3',) long,
each angled 60 degrees from vertical, (120 degrees
inclusive angle,) to replace TWO fluorescent tubs in one
fixture. The cost is about US $4 for the two SMD5630 LED
strips, and $10 for the power supply, (operating at 1/2
maximum rating.) In other words, US $14 per two tube
fixture.
But be advised, there is unknown reliability risk with
this alternative, (see above.) It would be inappropriate
to choose this alternative without a known manufacturer
and distributor that is standing behind MTBF numbers.
Without that, this alternative would not be advisable.
The author is not comfortable with the accuracy of the
calculated MTBF numbers, above. As bad as they are, they
may even be optimistic. If it is assumed they are
reasonable, (and the author is not confident of that,) an
8,000 hour MTBF for having to fix something, (replace
power supply, replace SMD5630 LED strips, etc.,) would be
24,000 hours MTBF at 1/3 of the time usage, or about every
2.8 years. For example, if a room has four each two bulb
fixtures, every 2.8 / 4 = 0.7 years = 8
months, something would have to fixed!!!,
(assuming ergodic failure mechanisms.) Note that at 16
months, the first alternative would have a total cost of
ownership advantage over the replaced components of this,
the second alternative.
Again, be advised that this alternative has significant
risks, the reader should understand them, and find
competent suppliers and distributors.
Both alternatives have about the same light output of the
original fluorescents, and both have similar cost of operation
savings, with the later having the lowest initial cost but the
highest maintenance costs, meaning the first has the lowest
total cost of ownership.
Figure I. PVC Tubing Implementation
Figure I is a photograph of the PVC tubing Implantation's
with two SMD5630 LED strips of 60 Leds, each, attached at 120
degrees inclusive angle.
In an attempt to increase the reliability of the SMD5630
LED strips by lowering the operating temperature, the SMD5630
LED strips were removed from the PVC tubing and placed on an 1
1/2" X 1 1/2" X 39.37" aluminum angle, (about US $1 to $2 per
foot at the local home improvements store,) which acts as a
heat sink, (the included angle is 90 degrees due to the
unavailability of desired 60 degree included angle.) The
SMD5630 LED strips were eclectically insulated from the
aluminum angle with common electrical tape. Assuming both
sides of each angle are subject to convectional air cooling,
the area would be 100
* 1.5 * 2.54 * 2 = 762 cm^2 and the
temperature rise would be 50 / sqrt (762) =
1.81 degrees C per Watt, for a one meter
length of SMD5630 LED strips. The calculated temperature rise
of the aluminum angle in the middle is 1.81 * 12
* 1.1 = 23.9C = 43.0F. The measurements,
after several hours of operation:
- From a digital thermometer:
- The ambient temperature was 84.4F
- From an IR thermometer:
- The aluminum angle was 113F
- An LED in the middle was 123F, a temperature rise of
38.6F = 21.4C, (this
probably accounts for the 50K hour MTBF quoted in the
manufacturer's data sheet, and is typical for
semiconductor devices, e.g., about 200K hours at a
junction temperature of 25C.) Note that the back side of
the SMD5630 LED strip is at heat sink temperature of
113F, and the front side is convection cooled at an
ambient air temperature of 84.4F; the actual temperature
rise of the LED was between 123F - 113F =
10F = 5.6C and 38.6F =
21.4C, (ignoring the thermal insulation
of the electrical tape)
- From a digital thermometer probe, (digital food
thermometer):
- The aluminum angle, (measured with heat sink
compound) was 116.3F
- The air temperature rise inside the S-60-12 power
supply was
18F = 10C. The
maximum temperature rise for the S-60-12 power supply
with an output current of 5A would be about 20C, and the
MTBF is specified at an ambient temperature of 25C, or
the FET output drivers, (probably the dominant failure
mode,) would be operating at an ambient temperature of
about 45C. This probably accounts for the 300K hour MTBF
at an ambient temperature of 25C, (excluding the MTBF
effects of any electrolytic and/or tantalum capacitors
in the power supply)
The measured temperature rise of the aluminum angle was
116.3F - 84.4F = 31.9F = 17.7C,
which agrees, reasonably well with the calculate value of
23.9C, (the discrepancy probably due to the front side heat
dissipation of the SMD5630 LED strip itself.)
Figure II. Aluminum Angle Implementation
Figure II is a photograph of the aluminum angle
implementation with two SMD5630 LED strips of 60 Leds, each,
attached at 90 degrees inclusive angle. The SMD5630 LED strips
were eclectically insulated from the aluminum angle with
common electrical tape.
With the heat sink, the SMD5630 LED strip reliability
probably is around 50K hours, for an ambient temperature less
than 44C = 111F, (human skin
necrosis.)
Suitable power supplies, (12V, 5A,) are available from
U.S. manufacturers/distributors that specify an MTBF of around
300K hours, minimum, MIL-HDBK-217F
(25C), albeit at about a 2 X price premium. At an ambient
temperature of 44C = 111F, (human
skin necrosis,) 44C - 25C = 19C,
or the reliability would be reduced to about
300K / 4 = 75K hours, (e.g.,
each temperature rise of 10C increases the failure rate by 2
X, from the Arrhenius
equation; 20C would be 4 X, assuming the failure
mechanisms are Ergodic.)
However, the lower MTBF can be raised by de-rating the power
supply maximum output current specification by a factor of 2,
(a common engineering practice,) thus increasing the MTBF by a
factor of 2^2 = 4, or
75K * 4 = 300K hours, meaning
the failure rate would be dominated by the failure rate of the
SMD5630 LED strips.
Since the power supply output current has to be de-rated by
a factor of 2, and 1.1A is required for each of the two
SMD5630 LED strips, (per meter,) in a fixture replacement, for
a total of 2.2 A, a 12V 4.4A power supply, minimum, would be
required, or 12V @ 5A power supply would be specified, per
meter of dual SMD5630 LED strips. Of course a less expensive
alternative is workable if expedient replacement is
considered. A S-60-12 power supply, 12V @ 5A was used in the
measurements, available on the Internet stores for US $10 to
$15. If many dual fluorescent lights are being replaced, it is
advisable that each dual SMD5630 LED replacement have an
individual power supply, (if the many replacements have a
common power supply, it is a single point of failure for all
the lights connected, albeit at a cost advantage.)
The light output of 1 meter of two SMD5630 LED strips,
(12V, 60 Led/m, IP20, daylight white, 6000-6500K,) per 90
degree angle aluminum, with a 12V 5A power supply, was
compared with two 4 foot Philips F40T12/DX/ALTO, T12 bulb,
20000 hours, CRI 79 Ra8, 6500K, 2600 lumens, in a shop light
fixture, (measured in the middle, at 1m with a digital LUX
meter, parallel to the floor, aimed at the light under
test):
- 1m dual SMD5630 LED strip: 330lm
- 4' dual Philips F40T12/DX/ALTO: 359lm
Which is about the same, despite 1m =
3.28' of SMD5630 LED strips being compared
with 4' of Philips F40T12/DX/ALTO fluorescent tubes.
License
A license is hereby granted to reproduce this design for
personal, non-commercial use.
THIS DESIGN IS PROVIDED "AS IS". THE AUTHOR PROVIDES NO
WARRANTIES WHATSOEVER, EXPRESSED OR IMPLIED, INCLUDING
WARRANTIES OF MERCHANTABILITY, TITLE, OR FITNESS FOR ANY
PARTICULAR PURPOSE. THE AUTHOR DOES NOT WARRANT THAT USE OF
THIS DESIGN DOES NOT INFRINGE THE INTELLECTUAL PROPERTY RIGHTS
OF ANY THIRD PARTY IN ANY COUNTRY.
So there.
Copyright © 1992-2017, John Conover, All Rights
Reserved.
Comments and/or problem reports should be addressed to:
- john@email.johncon.com
- http://www.johncon.com/john/
- http://www.johncon.com/ntropix/
- http://www.johncon.com/ndustrix/
- http://www.johncon.com/nformatix/
- http://www.johncon.com/ndex/
|
