Prev: User forum design
Next: jfet as current source, temp dependence in-theory question
From: Richard Rasker on 12 Apr 2010 07:23
Paul Keinanen wrote:

> On Sun, 11 Apr 2010 21:53:37 +0200, Richard Rasker
> <spamtrap(a)linetec.nl> wrote:
>
>>Hello,
>>
>>I've been asked to design a simple motor-driven pump for medical use, and
>>one of the primary requirements is that the device can be sterilized using
>>an autoclave, which works with steam at 135 degrees Centigrade.
>
> Have you looked at some power LEDs designed for 0,7-1,5 A operation,
> which are claimed to work at very high temperatures and running such
> devices at much lower currents for indicator purposes ?
>
> For instance the Luxeon K2 the claimed storage temperatures range of
> -40 to +185 C.
>
> The data sheet also claim:
>
>>Autoclave compliant— JESD22 A-102
>
>>Autoclave Conditions 121°C at 2 ATM, 100% RH for 72 hours max
>
> Is this sufficient for your application ?

Yes, these temperature specifications would definitely suffice, and I'll
certainly take these devices into consideration -- although the relatively
high cost (some $5 each) may put the customer off. Ah well, that's not my
problem :-)

Thanks for your suggestion, best regards,

Richard Rasker
--
http://www.linetec.nl
From: osr on 12 Apr 2010 07:57
K2 is out of production, take a look at Luxeon Rebel.. Then look at
Seoul Semiconductor to get the costs down!

Steve
From: osr on 12 Apr 2010 08:05

> >Steve
>
> Now that is nice, a 3055 with build in headlights :-)

Rolling on the floor laughing my posterior off... Thanks Jan, that
was exactly what I was thinking when I posted that.

Steve

From: RST Engineering on 12 Apr 2010 17:49
On Sun, 11 Apr 2010 21:53:37 +0200, Richard Rasker
<spamtrap(a)linetec.nl> wrote:

>Hello,
>
From what I can find, LED's are specified for a maximum of
>80~85 degrees C storage temperature, whereas most semiconductors have a
>range up to 150 degrees C (bipolar) or even 175 degrees C (FET's). So it
>would appear that using LED's may pose a problem in this particular case.
>
>Does anyone have any more information about this?
>
>Thanks in advance, best regards,
>
>Richard Rasker


What's wrong with cobbling up a board that has maybe 25-30 of them and
running them through a hundred autoclave cycles to see how many of
them fail?

Jim
From: Don Klipstein on 12 Apr 2010 21:56
In <4bc228c1$0$736$7ade8c0d(a)textreader.nntp.internl.net>, R. Rasker wrote:
>
>I've been asked to design a simple motor-driven pump for medical use, and
>one of the primary requirements is that the device can be sterilized using
>an autoclave, which works with steam at 135 degrees Centigrade.
>
>Basically, there's no need to use any semiconductors or electrolytic
>capacitors in the device, but my client asked if I could include LED's in
>the design. From what I can find, LED's are specified for a maximum of
>80~85 degrees C storage temperature, whereas most semiconductors have a
>range up to 150 degrees C (bipolar) or even 175 degrees C (FET's). So it
>would appear that using LED's may pose a problem in this particular case.
>
>Does anyone have any more information about this?

There was a time when LEDs generally had a maximum junction temperature
of 85 degrees C, for both storage and operation.

This is no longer true. I can think of 3 examples, with one of them
sadly no longer being manufactured (IIRC):

1. Silicon carbide LEDs - the original blue LEDs developed by Cree. As
far as I can tell, nobody makes these anymore, and Cree now uses silicon
carbide in LEDs only as a substrate for gallium nitride and indium gallium
nitride.

Silicon carbide blue LEDs appear to me to be close to bombproof, with
the die having no problem with any temperature that epoxy encapsulant can.
If someone would encapsulate them in silicone in a TO-18 package or a
TO-18 package with no encapsulant (less efficient optically), they could
probably easily take 200 degrees C, likely much more.

One SiC LED that still glowed after a lightning surge scorched the epoxy
body encapsulating the die is shown in:

http://ledmuseum.candlepower.us/led/ledsic.htm

2. There are modern high power LEDs able to take junction temperature
as high as 150 degrees C for operation and 135 degrees C for storage for
white, green and blue, 135 degrees C (for both storage and operation) for
red, orange and amber-yellow. One example is Philips-Lumileds "Luxeon
Rebel". Maximum external temperature when operating at 350 mA is 135
degrees C for white, green and blue, and 120 degrees C for red, orange and
amber-yellow.

A datasheet supporting tolerance of such temperatures is at:

http://www.philipslumileds.com/pdfs/DS65.pdf

3. Ordinary low power LEDs can now tolerate temperatures well above 85
degrees C. One example is Avago HLMP-ED31-SVOOO, and at least many of
their other HLMP-ED*, HLMP-EH*, and HLMP-EL* LEDs. The maximum junction
temperature is 130 degrees C for operating, and maximum case temperature
is 100 degrees C for both operating and storage. The datasheet is at:

http://www.avagotech.com/docs/AV02-0342EN

- Don Klipstein (don(a)misty.com)
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: User forum design
Next: jfet as current source, temp dependence in-theory question