Driver to drive?

On Tue, 15 Nov 2016 10:14:08 +0200, upsidedown@downunder.com wrote:

On Mon, 14 Nov 2016 20:40:32 -0800 (PST), sideley@yahoo.com wrote:

Ok, i see that there are people in here.

I am running a led fixture made out of a fluorescent fixture and a MC PCB: 8 1W leds are soldered to a MCPPCB which is glued to the aluminium shade by means of a glue and thermal paste. See photos for your reference.

The 8 leds are mounted into 2 series of 4 leds.

http://www.plantedtank.net/forums/attachment.php?attachmentid=690081&d=1479100005

When the leds are supplied with a 175mA current, the temperature on the aluminum shade (right at the back of the PCB) is about 35°C while the ambiant temp is about 30°C. But when supplied with 350mA, the nominal current, the temperature would raise up to roughly 60°C, which is way to hot.

In my opinion, the heat is partly because those leds are cheap and ineffcient. My question is if i replaced them with better quality chip leds with proven efficiency of about say 130lm/W, and run them at 350mA Would i still need to add over radiators to this homemade fixture ?

Look carefully at the LED specifications. The life time for those "1
W" LEDs are usually claimed at 150 or 175 mA, dropping rapidly at 300
or 350 mA.

The 130 lm/W is claimed at 150 mA or similar, but may drop to 100 lm/W
at 350 mA.

One other thing to watch for when reading data sheets, is the junction
temperature Tj used for the measurements (forward drop, light output)
Still most manufacturer specify Tj=25 C, i.e. during the first
millisecond when the LED is turned on but before heating up, giving
favorable characteristics.

Some reputable manufacturers specify characteristics at Tj=85 C, which
is closer to realistic steady state conditions. These data sheets also
contain parameter correction coefficients for other temperatures
around Tj=85 C.

That 130 lm/W seems a bit optimistic, probably obtained at Imax/10,
Tj=25C (startup only) and with a horrible spectral distribution.

Take a look at the Cree web site, it contains some good white papers
about thermal design.

Run the LEDs at 1/2 maximum power and compensate the loss of absolute
brightness by doubling the number of LEDs. This will also simplify
thermal design, since the internal junction temperature rise relative
to heatsink is also halved.

 

[Jon Elson]

sideley@yahoo.com wrote:

Ok, i see that there are people in here.

I am running a led fixture made out of a fluorescent fixture and a MC PCB:
8 1W leds are soldered to a MCPPCB which is glued to the aluminium shade
by means of a glue and thermal paste. See photos for your reference.

The 8 leds are mounted into 2 series of 4 leds.


http://www.plantedtank.net/forums/attachment.php?attachmentid=690081&d=1479100005

When the leds are supplied with a 175mA current, the temperature on the
aluminum shade (right at the back of the PCB) is about 35°C while the
ambiant temp is about 30°C. But when supplied with 350mA, the nominal
current, the temperature would raise up to roughly 60°C, which is way to
hot.

In my opinion, the heat is partly because those leds are cheap and
ineffcient. My question is if i replaced them with better quality chip
leds with proven efficiency of about say 130lm/W, and run them at 350mA
Would i still need to add over radiators to this homemade fixture ?

Running white LEDs at 350 mA, they will be absorbing 1 Watt, each. Now, a
fair amount of that power turns into light. But, some of it stays behind as
heat. Chip LEDs can deliver this heat to some substrate much better than
leaded parts. I built an LED retrofit that gives about 4 square inches of
PC board material to each chip LED for dissipation. These have a
temperature rise of maybe 10-15 C (You can be greatly fooled by touching the
front of the LEDs while they are on!) So, you have to turn them off and
then quickly check the temperature.

Jon

 

[Joerg Niggemeyer]

In message <yaednenouZSkFbbFnZ2dnUU7-L_NnZ2d@giganews.com>
Jon Elson <jmelson@wustl.edu> wrote:

current, the temperature would raise up to roughly 60°C, which is way to
hot.

Think about automotive head lights, that temp is not very high as long
as the cooling chain is good. What matters is the LED junction temp.
When the final mounting position will not raise the case hotter than
80° I think can still operate them with no problems assuming your temp
interface was done correct, which I can not see in detail at the
moment.

In my opinion, the heat is partly because those leds are cheap and
ineffcient. My question is if i replaced them with better quality chip
leds with proven efficiency of about say 130lm/W, and run them at 350mA
Would i still need to add over radiators to this homemade fixture ?

The need for a good cooling contact is still there and the best lumen
performance you get only when the junction is kept at 25°c operated in
a fridge ;-)

leaded parts. I built an LED retrofit that gives about 4 square inches of
PC board material to each chip LED for dissipation. These have a
temperature rise of maybe 10-15 C (You can be greatly fooled by touching the
front of the LEDs while they are on!) So, you have to turn them off and
then quickly check the temperature.

This is not a good idea, because there is not much heat capacity of
such small emitter present. You can solder a K-thermal element as
close as possible to the LED solder point and keep it running.

If you want to know precisely how hot your LED junction is driven
against the ambient and what lifetime you are targeting I have
developed a MCU based principle on that
http://www.led-temperature-protection.com


Joerg


--

 

[George Herold]

On Wednesday, November 16, 2016 at 3:46:41 AM UTC-5, Joerg Niggemeyer wrote:

In message <yaednenouZSkFbbFnZ2dnUU7-L_NnZ2d@giganews.com
Jon Elson <jmelson@wustl.edu> wrote:


current, the temperature would raise up to roughly 60°C, which is way to
hot.

Think about automotive head lights, that temp is not very high as long
as the cooling chain is good. What matters is the LED junction temp.
When the final mounting position will not raise the case hotter than
80° I think can still operate them with no problems assuming your temp
interface was done correct, which I can not see in detail at the
moment.


In my opinion, the heat is partly because those leds are cheap and
ineffcient. My question is if i replaced them with better quality chip
leds with proven efficiency of about say 130lm/W, and run them at 350mA
Would i still need to add over radiators to this homemade fixture ?

The need for a good cooling contact is still there and the best lumen
performance you get only when the junction is kept at 25°c operated in
a fridge ;-)


leaded parts. I built an LED retrofit that gives about 4 square inches of
PC board material to each chip LED for dissipation. These have a
temperature rise of maybe 10-15 C (You can be greatly fooled by touching the
front of the LEDs while they are on!) So, you have to turn them off and
then quickly check the temperature.

This is not a good idea, because there is not much heat capacity of
such small emitter present. You can solder a K-thermal element as
close as possible to the LED solder point and keep it running.

If you want to know precisely how hot your LED junction is driven
against the ambient and what lifetime you are targeting I have
developed a MCU based principle on that
http://www.led-temperature-protection.com

That's interesting.. you are using the led as it's own temperature sensor?
(measuring forward voltage drop at some set current.)

George H.

Joerg


--

 

[Martin Brown]

On 15/11/2016 04:40, sideley@yahoo.com wrote:

Ok, i see that there are people in here.

I am running a led fixture made out of a fluorescent fixture and a MC PCB: 8 1W leds
are soldered to a MCPPCB which is glued to the aluminium shade by
means of a glue and thermal paste.
See photos for your reference.

The 8 leds are mounted into 2 series of 4 leds.

http://www.plantedtank.net/forums/attachment.php?attachmentid=690081&d=1479100005

When the leds are supplied with a 175mA current, the temperature on the aluminum shade
(right at the back of the PCB) is about 35°C while the ambiant temp
is about 30°C.
But when supplied with 350mA, the nominal current, the temperature
would raise up to
roughly 60°C, which is way to hot.

In my opinion, the heat is partly because those leds are cheap and ineffcient.
My question is if i replaced them with better quality chip leds with
proven
efficiency of about say 130lm/W, and run them at 350mA Would i still
need to add
over radiators to this homemade fixture ?

You would be better off with batwing LEDs screwed down onto on something
like 1/8" thick aluminium 2" wide strip. That will get more heat away
from the LEDs although you still have to live with the fact that a fair
amount of low grade thermal energy has to be disposed of.

Heatsinks with fins or tube designed to encourage convective heat loss
will work better but be bulkier. More efficient LEDs will only help a
little bit. Having a much better heatsink could help a lot.

Since it is getting above 55C and is bare metal at present you could
improve things a bit by adding a thin coat of black(*) paint. A metallic
surface is a terrible thermal radiator and radiation losses start to be
important at 55C and above.

(*) The black paint can be almost any colour you like only not metallic.
It only needs to be black in the thermal IR band.

--
Regards,
Martin Brown

 

[George Herold]

On Wednesday, November 16, 2016 at 1:52:51 PM UTC-5, Joerg Niggemeyer wrote:

In message <efc84c89-30ad-4138-9691-388f8c6ecb7e@googlegroups.com
George Herold <gherold@teachspin.com> wrote:


http://www.led-temperature-protection.com

That's interesting.. you are using the led as it's own temperature sensor?
(measuring forward voltage drop at some set current.)

The LED itself is the sensor. The PWM dimming pauses are used to
sense the voltage drop of a cap in parallel to the LED at a pre set
time delay to get an LED forward voltage at low current situation for
being evaluated.

The prior art of NXP is switching in between two current levels,
I simply switch off ;-)

Prior art? Do you mean NXP got a patent for measuring a diode voltage drop
at two different currents to determine the diode temperature.
(That seems pretty "old school". But perhaps I'm not understanding.)

I'm not sure how you get the led to measure it's own temperature
if you are not putting any current through it.
(But you don't have to share your secrets.)

George H.

It so happened, that it is working

very well with their new automotive ASL3415SHN. In combination
with this driver no extra cost - just software ;-)


--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsfßhrer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[Lasse Langwadt Christense]

Den onsdag den 16. november 2016 kl. 19.52.51 UTC+1 skrev Joerg Niggemeyer:

In message <efc84c89-30ad-4138-9691-388f8c6ecb7e@googlegroups.com
George Herold <gherold@teachspin.com> wrote:


http://www.led-temperature-protection.com

That's interesting.. you are using the led as it's own temperature sensor?
(measuring forward voltage drop at some set current.)

The LED itself is the sensor. The PWM dimming pauses are used to
sense the voltage drop of a cap in parallel to the LED at a pre set
time delay to get an LED forward voltage at low current situation for
being evaluated.

The prior art of NXP is switching in between two current levels,
I simply switch off ;-) It so happened, that it is working
very well with their new automotive ASL3415SHN. In combination
with this driver no extra cost - just software ;-)

some 20 years ago in school I build a temperature controlled
soldering iron from a kit
The controller measured temperature by measuring the resistance
of the heater in the zero-crossing of the rectified AC used to
drive the heater

-Lasse

 

[Joerg Niggemeyer]

In message <efc84c89-30ad-4138-9691-388f8c6ecb7e@googlegroups.com>
George Herold <gherold@teachspin.com> wrote:

http://www.led-temperature-protection.com

That's interesting.. you are using the led as it's own temperature sensor?
(measuring forward voltage drop at some set current.)

The LED itself is the sensor. The PWM dimming pauses are used to
sense the voltage drop of a cap in parallel to the LED at a pre set
time delay to get an LED forward voltage at low current situation for
being evaluated.

The prior art of NXP is switching in between two current levels,
I simply switch off ;-) It so happened, that it is working
very well with their new automotive ASL3415SHN. In combination
with this driver no extra cost - just software ;-)


--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsführer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[George Herold]

On Wednesday, November 16, 2016 at 4:40:14 PM UTC-5, Lasse Langwadt Christensen wrote:

Den onsdag den 16. november 2016 kl. 19.52.51 UTC+1 skrev Joerg Niggemeyer:
In message <efc84c89-30ad-4138-9691-388f8c6ecb7e@googlegroups.com
George Herold <gherold@teachspin.com> wrote:


http://www.led-temperature-protection.com

That's interesting.. you are using the led as it's own temperature sensor?
(measuring forward voltage drop at some set current.)

The LED itself is the sensor. The PWM dimming pauses are used to
sense the voltage drop of a cap in parallel to the LED at a pre set
time delay to get an LED forward voltage at low current situation for
being evaluated.

The prior art of NXP is switching in between two current levels,
I simply switch off ;-) It so happened, that it is working
very well with their new automotive ASL3415SHN. In combination
with this driver no extra cost - just software ;-)

some 20 years ago in school I build a temperature controlled
soldering iron from a kit
The controller measured temperature by measuring the resistance
of the heater in the zero-crossing of the rectified AC used to
drive the heater

-Lasse

What was the "source" for the measurement?
Oh a cap on one side to hold some voltage..
(I think Joerg N. mentioned a cap.)

George H.

 

[whit3rd]

On Monday, November 14, 2016 at 8:40:38 PM UTC-8, sid...@yahoo.com wrote:

I am running a led fixture made out of a fluorescent fixture and a MC PCB: 8 1W leds are soldered to a MCPPCB which is glued to the aluminium ...

In my opinion, the heat is partly because those leds are cheap and ineffcient. My question is if i replaced them with better quality chip leds with proven efficiency of about say 130lm/W

A good red LED is 7% efficient. An excellent lighting-standard one might be 6%, and the
best might be 12%. That's because you lose light in driving the phosphors when you
ask for white-light output.

8W input, with a 6% lamp, will make you dissipate 7.52W as heat.
Double the efficiency, and 8W input with a 12% lamp means you dissipate 7.04W as heat.

You don't get much better heat performance by playing with different lamps, so much as
by using fewer such lamps you can get the same light and use less input power. To get
better heat performance, consider mounting a cooling fan, and/or a radiator-like heatsink.

 

[Jon Elson]

Joerg Niggemeyer wrote:

In message <yaednenouZSkFbbFnZ2dnUU7-L_NnZ2d@giganews.com
Jon Elson <jmelson@wustl.edu> wrote:


I built an LED retrofit that gives about 4 square inches
of
PC board material to each chip LED for dissipation. These have a
temperature rise of maybe 10-15 C (You can be greatly fooled by touching
the
front of the LEDs while they are on!) So, you have to turn them off and
then quickly check the temperature.

This is not a good idea, because there is not much heat capacity of
such small emitter present. You can solder a K-thermal element as
close as possible to the LED solder point and keep it running.

Oh, I'm not measuring junction temp, I'm measuring the PC board temperature.
I figure if I can keep the PC board to 40 C or so, the LEDs will do fine.
The LED chips are on a leadless ceramic substrate that is soldered directly
to copper planes. Each "plane" is 4 square inches, and has an LED on each
end. So, each LED has 4 square inches of radiator attached.

If you want to know precisely how hot your LED junction is driven
against the ambient and what lifetime you are targeting I have
developed a MCU based principle on that
http://www.led-temperature-protection.com

That's cool, but more work than I wanted to deal with. My LEDs have been
running fine for over 2 years, now. I need to make more and convert more of
our light fixtures.

Jon

 

[krw]

On Wed, 16 Nov 2016 09:46:27 +0100, Joerg Niggemeyer
<joerg.niggemeyer@nucon.de> wrote:

In message <yaednenouZSkFbbFnZ2dnUU7-L_NnZ2d@giganews.com
Jon Elson <jmelson@wustl.edu> wrote:


current, the temperature would raise up to roughly 60°C, which is way to
hot.

Think about automotive head lights, that temp is not very high as long
as the cooling chain is good. What matters is the LED junction temp.
When the final mounting position will not raise the case hotter than
80° I think can still operate them with no problems assuming your temp
interface was done correct, which I can not see in detail at the
moment.

In the Arizona sun, the case won't get hotter than 80C? I think
you're wrong. There is a reason that most automotive electronics is
specified to 150C.

In my opinion, the heat is partly because those leds are cheap and
ineffcient. My question is if i replaced them with better quality chip
leds with proven efficiency of about say 130lm/W, and run them at 350mA
Would i still need to add over radiators to this homemade fixture ?

The need for a good cooling contact is still there and the best lumen
performance you get only when the junction is kept at 25°c operated in
a fridge ;-)

With an infinite heat sink.

leaded parts. I built an LED retrofit that gives about 4 square inches of
PC board material to each chip LED for dissipation. These have a
temperature rise of maybe 10-15 C (You can be greatly fooled by touching the
front of the LEDs while they are on!) So, you have to turn them off and
then quickly check the temperature.

This is not a good idea, because there is not much heat capacity of
such small emitter present. You can solder a K-thermal element as
close as possible to the LED solder point and keep it running.

If you want to know precisely how hot your LED junction is driven
against the ambient and what lifetime you are targeting I have
developed a MCU based principle on that
http://www.led-temperature-protection.com


Joerg

 

[Joerg Niggemeyer]

In message <3eb89701-03f4-4813-9f42-4393886c4440@googlegroups.com>
George Herold <gherold@teachspin.com> wrote:

The prior art of NXP is switching in between two current levels,
I simply switch off ;-)

Prior art? Do you mean NXP got a patent for measuring a diode voltage drop
at two different currents to determine the diode temperature.
(That seems pretty "old school". But perhaps I'm not understanding.)

http://www.nxp.com/wcm_documents/technologies/documents/nxp_sensorless_led_temp_sensing.pdf

They do have 20+ ip rights on it and so I was indeed a little bit
impressed when I got from German patent office the result of their
first research referencing NXP.

I'm not sure how you get the led to measure it's own temperature
if you are not putting any current through it.
(But you don't have to share your secrets.)
The low supply current is taken out of the discharge of a cap.

It is not a secret since I have granted US Patend and the German
patent went through last week - so everything is public available,
except the software, which is not a big problem to have it run on a
8bit 16F low cost PIC.

I don't want to design the LED gears myself, for instance I would like
to get others using this idea for instance inside a smartphone
protecting the hi pow LED from overheat - just in cases like the Lio
starts burning in some hot pants ;-)))))

Joerg


--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsführer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[George Herold]

On Thursday, November 17, 2016 at 3:49:03 AM UTC-5, Joerg Niggemeyer wrote:

In message <3eb89701-03f4-4813-9f42-4393886c4440@googlegroups.com
George Herold <gherold@teachspin.com> wrote:



The prior art of NXP is switching in between two current levels,
I simply switch off ;-)

Prior art? Do you mean NXP got a patent for measuring a diode voltage drop
at two different currents to determine the diode temperature.
(That seems pretty "old school". But perhaps I'm not understanding.)

http://www.nxp.com/wcm_documents/technologies/documents/nxp_sensorless_led_temp_sensing.pdf

They do have 20+ ip rights on it and so I was indeed a little bit
impressed when I got from German patent office the result of their
first research referencing NXP.


I'm not sure how you get the led to measure it's own temperature
if you are not putting any current through it.
(But you don't have to share your secrets.)
The low supply current is taken out of the discharge of a cap.

It is not a secret since I have granted US Patend and the German
patent went through last week - so everything is public available,
except the software, which is not a big problem to have it run on a
8bit 16F low cost PIC.

I don't want to design the LED gears myself, for instance I would like
to get others using this idea for instance inside a smartphone
protecting the hi pow LED from overheat - just in cases like the Lio
starts burning in some hot pants ;-)))))

Joerg

OK thanks for sharing. Good luck with the invention!
(I wonder if people do the same thing with mosfets..
reverse bias 'em and measure the body diode.)

George H.

--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsfßhrer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[Joerg Niggemeyer]

In message <gq2q2cdt1tjabhvotorscgucq7tpkh8lbt@4ax.com>
krw <krw@somewhere.com> wrote:

In the Arizona sun, the case won't get hotter than 80C? I think
you're wrong. There is a reason that most automotive electronics is
specified to 150C.

Sorry for the misunderstanding my English is not the best ;-)
I ment it as a comparisson. The requirement for automotive LED is
working up to 125°C ambient or temp of the cooling block.

Inside home applications doing some retrofits under a wooden ceiling
I would try to keep the outer lamp housing temp << 100°C.

J Ni

--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsführer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[Phil Hobbs]

On 11/17/2016 03:30 AM, Joerg Niggemeyer wrote:

In message <3eb89701-03f4-4813-9f42-4393886c4440@googlegroups.com
George Herold <gherold@teachspin.com> wrote:



The prior art of NXP is switching in between two current levels,
I simply switch off ;-)

Prior art? Do you mean NXP got a patent for measuring a diode
voltage drop at two different currents to determine the diode
temperature. (That seems pretty "old school". But perhaps I'm not
understanding.)

http://www.nxp.com/wcm_documents/technologies/documents/nxp_sensorless_led_temp_sensing.pdf

They do have 20+ ip rights on it and so I was indeed a little bit
impressed when I got from German patent office the result of their
first research referencing NXP.


I'm not sure how you get the led to measure it's own temperature if
you are not putting any current through it. (But you don't have to
share your secrets.)
The low supply current is taken out of the discharge of a cap.

It is not a secret since I have granted US Patend and the German
patent went through last week - so everything is public available,
except the software, which is not a big problem to have it run on a
8bit 16F low cost PIC.

I don't want to design the LED gears myself, for instance I would
like to get others using this idea for instance inside a smartphone
protecting the hi pow LED from overheat - just in cases like the Lio
starts burning in some hot pants ;-)))))

Joerg

Do you have a new wrinkle on the technique? Folks have been doing that
sort of thing approximately since dirt was invented, or at least the
last dozen years.

From BEOS 3:
"LED wavelengths drift at about +100~ppm/K, which is not particularly
small. Efficiencies also drop with temperature, at about -0.5\%/K, so
to maintain calibration you'll probably need to use closed-loop control
with a photodiode. Alternatively, you can use the diode forward voltage
as a thermometer and use closed-loop temperature control, which will get
rid of the spectral shift as well. For a good discussion of all these
effects, see \eg\ Chhajeda \etal{}.{}\footnote{S. Chhajeda, Y. Xi, Th.
Gessmann, J.-Q. Xi, J. M. Shah, J.
K. Kim, and E. F. Schubert, "Junction temperature in light-emitting
diodes assessed by different methods", \emph{Proc. SPIE} {\b 5739},
16-24 (2005)}

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[Joerg Niggemeyer]

In message <n9CdnS86ULkeV7DFnZ2dnUU7-Q_NnZ2d@supernews.com>
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

Do you have a new wrinkle on the technique? Folks have been doing that
sort of thing approximately since dirt was invented, or at least the
last dozen years.

The physical background of it is not new. Now we do have MCU power
available which can be used to save a lot of parts and costs. It's not
ment to be the most precise like Mentor Graphics it shall be the most
simple and effective method.

Regards

Joerg



--

WEB: http://www.nucon.de http://www.led-temperature-protection.com
Geschäftsführer: Joerg.Niggemeyer@nucon.de
Steinbecker Muehlenweg 95, 21244 Buchholz idN, Germany
UST-IDNR.: DE 231373311, phone: +49 4181 290913, fax: +49 4181 350504

 

[Joerg Niggemeyer]

In message <c0ce4bcb-7761-4afa-9eb6-c897ffa7d8e8@googlegroups.com>
George Herold <gherold@teachspin.com> wrote:

OK thanks for sharing. Good luck with the invention!
(I wonder if people do the same thing with mosfets..
reverse bias 'em and measure the body diode.)

Thank you !

Regarding the MOSFETS ;-)) : there might be some arround who like to
use them as a heater, I normaly prefer to keep them running cool....

rgds Joerg

--

 

[Phil Hobbs]

On 11/17/2016 12:52 PM, Joerg Niggemeyer wrote:

In message <n9CdnS86ULkeV7DFnZ2dnUU7-Q_NnZ2d@supernews.com
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:



Do you have a new wrinkle on the technique? Folks have been doing that
sort of thing approximately since dirt was invented, or at least the
last dozen years.

The physical background of it is not new. Now we do have MCU power
available which can be used to save a lot of parts and costs. It's not
ment to be the most precise like Mentor Graphics it shall be the most
simple and effective method.

Regards

Joerg

Hmm, maybe. I find it a bit hard to believe that at the time of the
Chhajeda et al. reference (2005) MCUs didn't have enough power to
subtract two voltages and multiply by the reciprocal of a tempco, on a
timescale faster than the heating up of a LED. (In 2000 I was running a
96-pixel thermal imaging camera with a PIC17. It needed a bunch more
normalization than that.)

I looked for your new patent 9,326,349, but Google Patents doesn't have
it. I found the application at
<http://www.freepatentsonline.com/20150245429.pdf>.

I didn't read it at all carefully, but from Fig. 1 and the claims, it
looks like the novelty you're claiming is basically the series RC to
provide a small amount of bias to the LED while the PWM is turned off,
so that you can measure the forward voltage without the ohmic
contribution. That's a reasonable way to do it, for sure, but seems a
bit thin for a patent. However, the USPTO is the USPTO, and if you can
make some dough out of it, terrific. I wouldn't go suing anybody over
it though--the defendant's first response would probably be to file an
inter partes review (IPR) which IMO you'd be very likely to lose.

There are two big advantages to their doing that. First, IPRs are cheap
compared with district course cases (about $25k plus lawyers' bills),
and if they lose they can still fight in court. Second, and more
important, in court an issued patent carries a strong presumption of
validity, whereas in an IPR the standards are the same as in the
original examination. That makes an IPR far more dangerous to a patent
holder than a district court case. (I'm not a lawyer, but so I've been
told by folks who are, and I've seen it happen.)

Cheers

Phil Hobbs

(Not a lawyer, but involved in two inter partes reviews as we speak, as
well as four or five previous ones.)

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[Jasen Betts]

On 2016-11-17, Joerg Niggemeyer <joerg.niggemeyer@nucon.de> wrote:

In message <gq2q2cdt1tjabhvotorscgucq7tpkh8lbt@4ax.com
krw <krw@somewhere.com> wrote:


In the Arizona sun, the case won't get hotter than 80C? I think
you're wrong. There is a reason that most automotive electronics is
specified to 150C.

Sorry for the misunderstanding my English is not the best ;-)
I ment it as a comparisson. The requirement for automotive LED is
working up to 125°C ambient or temp of the cooling block.

Inside home applications doing some retrofits under a wooden ceiling
I would try to keep the outer lamp housing temp << 100°C.

Why? Wood is safe to over 200°C, and incandescent lamps are typically
hotter than that.

--
This email has not been checked by half-arsed antivirus software