Margery ConnerTechnical Editor Margery Conner's PowerSource streams the latest developments in electronic power design and related technologies.


Profile

RSS Feed

  • Add this blog to your RSS newsreader!

Recent Posts

Recent Comments

Most Commented On

Archives

By Category

Power Management Articles

Blog

Wednesday, March 19, 2008

EMI, power surges become important when Ethernet leaves the building

Mar 19 2008 9:12PM | Permalink | Email this | Comments (1) |
Blog This! using:  Blogger.com | LiveJournal |
Digg This | Slashdot This | add to Del.icio.us


I was talking with Amit Gattani, director of marketing for Akros Silicon, about their recently-introduced AS1135 PD controller IC for 30W PoE (Power over Ethernet) – this is the  version of 802.3at that’s still in the Draft version, although Gattani says the PHY layer is wrapped up, and that’s the layer that this chip addresses.

My interest was in Akros’ continued emphasis on EMI filtering and surge protection for their PoE chips. When they introduced their AS1601/02 chip last year, they made a point of including filtering and protection on the chip, yet when I asked several PoE vendors about their need for surge protection I got blank stares in return – it just wasn’t a hot button for the folks I spoke with.

 Gattani explained that Ethernet has traditionally been an enterprise technology that’s all inside one building. “As the technology moves to PoE, we see more use of networking outside of buildings now, such as security cameras mounted outside, WiFi access points, or fiber-to-the-home with Ethernet running outside the home. Suddenly there’s a bigger scope where you’re more prone to lightning and surge-related issues.” Plus, in emerging countries like China, Russia, or India, power is not highly controlled compared to an American deployment, and you can get some serious power surges. He agrees that if your application is totally inside a building, which is a relatively benign environment, then you’re more concerned about issues lie cable discharge/disconnect events, but he argues that EMI filtering/surge protection is still valuable to have built-in to your PoE controller chip, and makes for one less headache to worry about in your design.

 


Related entries in: Communication functions | Communications & Networking | Power Sources/Controllers | Power Supplies | 


Reader Comments


at 3/20/2008 2:38:29 PM, Chas_nc said:
Well, as with most of the people who were part of the original PoE study groups, your contacts in the article are also wrong about building environments being benign. There are many scenarios that can cause a 100-200V ground potential from one end of a building to another, and most of the PoE implementations are almost oblivious to these problems due to the additional costs to protect and/or ignorance of the surge problem. One easy scenario to understand this problem: Most offices have 480V feeds. If a 480V line-to-ground fault occurs at the end of a building's feed, the 480V is divided between the building wiring feed inductance and the return path ground inductance until the feed breaker opens (up to 10-20 ms). That means the local ground at the fault can end up ~138Vrms (277/2) over the entrance source ground. The peak of 138Vrms would be about 195Vpk. Any cabling spanning that ground path will see this ground bounce. A typical 5-20ms ground fault is plenty of time to destroy the low voltage devices used in most cheap PoE solutions, and their loads as well. There are also issues with personnel ESD, building static discharge, and building lightning. Since many industrial buildings are made with metallic outer structures, they make great lightning attractors.

Post a comment


Display Name

Before submitting this form, please type the characters displayed above:


ADVERTISEMENT

©1997-2008 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

Please visit these other Reed Business sites