AC Surge Protection

[jmed999]

What is a good way to protect the Elk M1G from AC power surges? Regular plug in type surge protector with the Elk transformer then plugged into the surge protector, Elk 950, Leviton 47605, etc.???


What's the best way to go here? Thanks!


PS. I do not use a phone line so protecting a phone line isn't important to me.

 

[wuench]

The elk transformers have a self-resetting fuse. But you can plug it into an additional surge strip if you like. I have a Leviton 47605 in my cabinet, but due to the size of the transformer you lose an outlet.

The Elk 950 is for low voltage surge protection, like phone lines, which you said you don't need.

 

[Del91574]

The "fuse" in the Elk 1640 is only a PTC, protecting from shorts and overcurrents, not surges. It'll still pass a surge through to the panel, as it has no MOV protection.

Elk 950 is cheap enough, just don't use the phone portion.

Otherwise, you can pick up an AC surge by Ditek or Edco pretty cheaply.

 

[westom]

PS. I do not use a phone line so protecting a phone line isn't important to me.

Phone lines already had surge protection for longer than anyone here has existed. Most common source of destructive transients is AC mains. As another noted, fuses do not do that protection. Nothing stops a surge.

A lightning strike far down the street may be a direct strike to every household appliance. Are all appliances damaged? Of course not. For numerous reasons including protection already inside every appliance. And because both an incoming and an outgoing path (to earth) must exist to have damage.

Explains why telephone appliances are so easily damaged. Telephone lines already have superior and earthed protection. So current incoming on AC mains finds an outgoing path to earth destructively via a phone line. Damage is often not on an incoming path (AC mains). Damage is often on the outgoing path (telephone, cable TV)..

Best way to go is to connect that current to earth BEFORE it enters the building. Otherwise everything inside is at risk. Earth ground (necessary for AC electric, cable, and that telco 'installed for free' protector) does all protection. Either the incoming transient is connected to earth via a wire (ie cable TV). Or it is connected to earth via a 'whole house' protector (to protect everything).

'Whole house' protectors are, beyond any doubt, the best possible solution. And a least expensive solution. Were even sold in Lowes and Home Depot for less than $50. But - and this is most important - protection is only defined by what absorbs those hundreds of thousands of joules. No protector does that. Earth ground absorbs that energy. Protection is always about where energy dissipates. Always.

How does a transient connects to earth? Either harmlessly BEFORE entering the building (ie 'whole house protector or a short wire). Or destructively via appliances. Best protection for an Elk is also best protection for everything else inside the house. So that protection already inside every appliance is not overwhelmed by a rare and destructive surge.

 

[Del91574]

The above post is true, however the protectors on a phone line are only to help protect the telco equipment and many times, I've seen strikes feed into the house via the phone lines and kill alarm panels.

The reason why is because the alarm typically has the telephone and AC connection so it's very easy for a voltage potential to exist between those two items when transients start happening.

The whole house arrestors are a good idea, but I still would not forgo the installation of a supressor/arrestor on the alarm panel and components as well. The larger question looms of how fast that protection cuts over to ground. Many of the cheapie whole house units simply aren't fast enough to protect sensitive electronics, but compared to nothing, it's still an improvement.

In the case of an Elk 950, the "telephone" connection can be used for any other cable pair, so you could actually surge the 485 data on the M1 and isolate other sections of the panel.

 

[westom]

The whole house arrestors are a good idea, but I still would not forgo the installation of a supressor/arrestor on the alarm panel and components as well. The larger question looms of how fast that protection cuts over to ground. Many of the cheapie whole house units simply aren't fast enough to protect sensitive electronics, but compared to nothing, it's still an improvement.

If true, then a subjective claim was not posted. Instead, a number defines a 'too slow' protector.

Meanwhile, why would the telco install a 'whole house' protector on your house to protect their computer located miles away? A protector is required by code and laws to protect your telco appliances. A similar device is also on their side of the lightning strike. So that overhead wires also do not harm their $multi-million computer. Why do telcos operate equipment during every thunderstorm? Because about 100 surges per thunderstorm must not cause any damage. The technology installed on your end and their end is that well proven . A protector must exist on every subscriber interface. Where their wires connect to yours.

What number for a 'whole house' protector is too slow?

A number for a typically destructive surge is hundreds of thousands of joules. What number for a completely different device (adjacent to a panel) claims effective protection from that?

 

[Neurorad]

Welcome back, Westom.

 

[Neurorad]

+1, start with a whole house surge protector. I'm partial to Eaton.

Getting a whole house SPD is actually on my week's to-do list, as well as a few other items for the electrician.

 

[Del91574]

If true, then a subjective claim was not posted. Instead, a number defines a 'too slow' protector.

Meanwhile, why would the telco install a 'whole house' protector on your house to protect their computer located miles away? A protector is required by code and laws to protect your telco appliances. A similar device is also on their side of the lightning strike. So that overhead wires also do not harm their $multi-million computer. Why do telcos operate equipment during every thunderstorm? Because about 100 surges per thunderstorm must not cause any damage. The technology installed on your end and their end is that well proven . A protector must exist on every subscriber interface. Where their wires connect to yours.

What number for a 'whole house' protector is too slow?

A number for a typically destructive surge is hundreds of thousands of joules. What number for a completely different device (adjacent to a panel) claims effective protection from that?

I am not reading a question in regards to your post, so I'm unsure if there is actually one.
 
In the case of IDF and the construction of telco plants, there's generally a cheap carbon supressor installed at the Dmarc for the premises. It's designed to help prevent backfeeding from the subscriber's equipment to the telco, really not the opposite way, and honestly, it's effectiveness can be argued significantly. Telco's also typically have gas tube arrestors installed on their end, not carbon fuses or in the case of alarm equipment, transorbs. Different application and installation.

 

[westom]

I am not reading a question in regards to your post, so I'm unsure if there is actually one.
 
In the case of IDF and the construction of telco plants, there's generally a cheap carbon supressor installed at the Dmarc for the premises. It's designed to help prevent backfeeding from the subscriber's equipment to the telco, really not the opposite way, and honestly, it's effectiveness can be argued significantly.

 
No questions were asked.  A well proven solution was defined.
 
  In the 1950s, new technology (transistors) meant telcos reviewed existing protection.  So that transistors would be unharmed even by direct lightning strikes.  With minor corrections, 'carbons' (properly earthed) protected transistors from direct lightning strikes and other lesser transients.  A typical CO might suffer about 100 surges with each thunderstorm.  That $multi-million computer was protected from lightning and lesser transients (such as backfeeding from subscriber equipment).
 
 Today, telcos use better, semiconductor based protectors. (Transorbs are insufficient; telcos use better technology.) 
 
  A protector does not do protection. Carbon, gas discharge, or semiconductor is irrelevant.  In every case (as was true in the 1950s), protection was about where hundreds of thousands of joules harmlessly dissipate. Protection is defined by how energy connects to single point earth ground.
 
  A protector, adjacent to an appliance, does not claim to protect from destructive transients.  Protector, adjacent to what absorbs energy, protects even from lightning.  As carbons did before the 1950s.  Effective protectors connect energy to earth. And do not fail.  Effective protection means hundreds of thousands of joules (ie lightning) dissipate harmlessly in earth.
 
  This well proven and less expensive solution is provided by responsible manufacturers including Eaton (Cutler-Hammer), Siemens, Intermatic, Ditek, Polyphaser, ABB, Square D, Leviton, and General Electric - to name but a few. Similar solution is found in any facility that must suffer direct lightning strikes without damage even to a protector.
 
  In every case, a well proven protector connects hundreds of thousands of joules (ie a direct lightning strike) harmlessly to earth.  A protector is only as effective as its earth ground.
 
   A  telco 'installed for free' protector and an AC 'whole house' protector must both connect low impedance (ie 'less than 10 feet') to the same earth ground.  Otherwise, a voltage potential can exist between those two items when transients start happening. Causing alarm damage.  Effective protectors make a short connection to earth.  Protection, even in 1950 CO, was always about how a protector connected the transient to earth.
 

 

[Del91574]

Agreed, none of that is new technology, and yes, a surge is only meant to react to break over to ground, hopefully dissapating the transient to ground before damage. I threw transorbs as an example of component level protection, they are not what is used in high risk and value surge suppression installations.
 
The 10' rule isn't exactly true, the more important item is a bonded ground directly to a EG rod,  not a CWG, paying specific attention to the overall impedance of the cable and the ground rod >10 ohms. The other part of the equation is to make sure the path to ground is shorter than the protected circuit(s). The facts, however, is that "free" telco supplied protector usually bends the rules for being bonded to the same EG as other hardware, usually just to the closest CWG, or at least around my area, that's the pattern.

 

[westom]

The 10' rule isn't exactly true, the more important item is a bonded ground directly to a EG rod,  not a CWG, paying specific attention to the overall impedance of the cable and the ground rod >10 ohms. The other part of the equation is to make sure the path to ground is shorter than the protected circuit(s). The facts, however, is that "free" telco supplied protector usually bends the rules for being bonded to the same EG as other hardware,

 
  The OP greatest concern is a the protector.  A protector is simple science.  Most attention should focus on what actually does that protection.  Where hundreds of thousands of joules dissipate.  A utility in "Preventing Damage Due to Ground Potential Difference" demonstrates good, bad, and ugly (preferred, wrong, and right) solutions at:
    http://www.duke-energy.com/indiana-business/products/power-quality/tech-tip-08.asp