Advertisement

Zibb

Margery Conner Technical Editor Margery Conner's PowerSource streams the latest developments in electronic power design and related technologies. Follow Margery on Twitter at: http://twitter.com/margeryc.



   Advertisement

Profile

RSS Feed

  • Add this blog to your RSS newsreader!

Recent Posts

Recent Comments

Most Commented On

Archives

By Category

Power Management Articles

Blog

Monday, November 23, 2009

Ceramic caps promise long lifetimes for smaller LED lights

Nov 23 2009 10:11AM | Permalink |Comments (4) |


After seeing the role played by electrolytic capacitors in the lifetime of solar inverters and the importance of accurately derating them,, I asked the technical folks at AVX to talk about ceramic capacitors appropriate for another application that requires long lifetimes: LED lighting. Ron Demcko, application engineering manager at AVX, got back to me. Here’s Ron’s reply.

The perceived disparity between the life of LEDs (60,000 hours) and that of capacitors (perceived as 8000 hours) is a hot topic within the Solid State Lighting (SSL) world.

Many different types of capacitor technologies already exist that yield very reliable capacitors. Power film capacitors for example, have internal fuses within the electrodes. It’s not uncommon to talk about 100,000 hour reliability at rated voltage and 70°C. These devices are physically large and may be most applicable to stadium and large light schemes.

Existing ceramic capacitors are capable of meeting the reliability requirements of smaller-size LED-based lighting. For example, a ceramic capacitor operating at half-rated voltage and 50ºC can offer an operational life > 800,000 hours. [You can read the AVX research paper (pdf) here.]

Conservative voltage and temperature derating methods in conjunction with careful capacitor selection are critical to a long life design.

Next-generation ceramic capacitors for use in solid state lighting with temperature ratings of 150ºC in the X8R classification offer over a million hours of life even at temperatures of up to 85ºC. These capacitors are now being used in under-the-hood applications in automotive designs and can withstand harsh environments. Higher-temperature dielectric systems for operation up to 200ºC are expected to come to the market soon offering a wide range of operating temperatures.

With additional features like thermal cycle events becoming important for performance, special terminations have been developed for various dielectric materials. To address these issues, ceramic capacitors utilize a conductive epoxy material system between the first metallization of a MLCC and the Ni barrier layer. This conductive epoxy allows a MLCC to withstand > 1000 thermal cycles without failure and to perform reliably in applications where high vibration/board flexure is present. This termination can be utilized with all dielectric classes and operate up to 150ºC. The Flexi termination is also capable of high-voltage operation.

Next-generation low distortion ceramic capacitors can also help improve the noise emitted in PWM drivers associated with SSL circuitry.

With time expanded data bases and new products will dispel the concerns of ceramic capacitor applicability to solid state lighting circuits.


Related entries in: Components, Hardware, Interconnect | HBLED | 


Reader Comments



at 11/23/2009 11:01:16 AM, Andy T said:
Dear AVX,

Great "story".

Now give me a non-stacked MLCC in 470uF at 400V (minimum, 600V preferred) for <<$0.50, that's not the size of a San Jose home (which is tiny), that's made in the USA (not China with its quality problems that cause ceramic layer delaminations), that's rated for AT LEAST 100,000 hours of life at 125C, and that doesn't act as a big piezo speaker for my PWM's switching frequencies and you might be applicable to a REAL SSL design.

One more thing. Failure after a year of switching a $100 LED-based light on three times a day (the source of temp cycling) is not acceptable, nor is the fantasy of running it at less than 70C in a "100W" LED bulb in Desert Springs or Phoenix.

Until the marketers and business unit managers (the poor apps guy has to make lemonade from the lemons given to him by these people) face up to the real world needs and requirements, all the marketers and division VPs will try to position existing products and capability to pander to naive SSL designers (like 12V automotive for a 170V or 270V SSL application, or derated temperature to preach SSL-acceptable lifetimes) versus targeting R&D and product definition for the rigorous needs of SSL itself.

The SSL market will be huge - did I say "will be"? Maybe "may be". Unfortunately, most component vendors are treating SSL as a sideline that's not deserving of targeted, specifically-designed, product lines or any R&D money.

"Something for nothing" doesn't work within the potential of SSL's success, and in fact will doom it to mediocrity at best.



at 11/24/2009 2:22:29 PM, ElmasnoJ said:
A curious lack of publication date, with references dating from 1969 to 1982, causes one to ponder the current applicability & relevancy of the AVX Research study.



at 11/30/2009 2:06:59 PM, AAM said:
AVX apparently DO have some high capacitance, high voltage devices. BUT, quoting prices of $86 a pop isn't going to see many of these in an SSL anytime soon! Get them donw to compare with cheap long life electrolytics and they have a real market to open up...!



at 12/1/2009 2:44:47 PM, MikeS said:
What is the publication date of the AVX paper? This is pretty sloppy to not have any hint of what year it was put out.

There is no copyright notice with a date, so has AVX given up their copyright and it is OK for me to copy it any way I want?

Post a comment



Display Name

Change Image
Before submitting this form, please type the characters displayed above.
Note the letters are NOT case sensitive.


ADVERTISEMENT

©1997-2010 Reed Business Information, a division of Reed Elsevier Inc. All rights reserved.
Use of this Web site is subject to its Terms of Use | Privacy Policy