Lightning Protection

I've been reading all the great posts on this forum concerning lightning protection and have an idea of what I need but thought I would just make sure.
 
I'm trying to protect my equipment from wires coming from outside the house.
 
The sprinkler valve selenoids are 24VAC and I was planning on connecting to one of these prior to wiring to the M1XOR.
 
http://www.l-com.com/surge-protector-high-power-3-stage-surge-protector-for-24v-ac-control-lines
 
The low voltage lights are 12-22VAC and will also be connected to the above and then a transformer for low voltage lights.  This transformer has a simple on/off relay which will also be connected to the M1XOR with one of these inbetween:
 
http://www.l-com.com/surge-protector-high-power-3-stage-surge-protector-for-24v-dc-control-lines
 
The M1XOR will then be connected to the ELK M1G panel with another of the above between.
 
For the outdoor speakers I was planning on:
http://www.l-com.com/surge-protector-outdoor-4-8-ohm-audio-systems-lightning-protector
between the speakers and the amp
 
Although these are outdoor units, I was planning on installing them inside for greater WAF unless they would be safer outside.
 
I would appreciate any input!
 
THANKS
DJK

Ditek (or Edco for that matter)
 
Enough variety and product selection to fit the bill.

dkemme said:

I would appreciate any input!

Click to expand...

 
Good topic!  You may also want to look into irrigation valves which will fail into the closed position if affected by lightning.  My neighbor's house got hit by lightning (or maybe the ground near it did), and the lightning bolt somehow activated all his sprinkler valves and left them running in the open position--while he was on vacation.  Not sure what kind of valves he had.  Don't solenoid valves typically close unless held open by current?  I'm having trouble imagining exactly how lightning could have done what it did.  Could the initial surge have opened it, and the rest of the surge melted the coil wire, effectively welding the rod and holding it in the open position? Do you need a surge protector of some kind mounted on each and every solenoid to prevent it from happening?  What's the best defense?
 
Come to think of it, I have one of those touchH20 faucets on the kitchen sink, and I suppose the same could happen to it.  What about laundry machine and dishwasher solenoids?  Could the same happen to them?    That might be a worst case scenario if lightning welded their solenoids open too.  In such a scenario, those appliances  would presumably overflow, because the pumps that normally drain them as part of a normal cycle would probably be offline.  If it could happen to the solenoids on your ice cube maker too, you'd get another overflow and flood there as well.

Ok, anyone else glad to see someone spell lightning properly?  The internet is not kind to spelling pet peeves...

All bets are off when you're trying to predict the failure pattern post-lightning or surge.
 
I've had traces blown off circuit boards with cables/devices blown apart and also wiring charred, and I've also had the same where there was no outward sign except for other system components that took a hit.

Much more likely that the triacs (solid state switches) in the sprinkler controller failed short (and supplied voltage to the valves) than that something happened to the sprinkler valves to keep them stuck open.  That is a typical failure mode for solid state devices (triacs, transistors, MOSFETs, etc).  If there is enough energy they open after they short but that takes quite a bit more energy - the wire connections inside the part blow out.
 
Those protectors look like they should be pretty effective - note that they need a good (i.e. short, direct) ground.  Largely ineffective without that.  Single point ground for everything where it comes in to the building is best.  As has been pointed out, it isn't always possible to predict exactly what will happen when a strike hits nearby.  But you can put what you consider reasonable protection in place to lessen the risk.

JimS said:

Much more likely that the triacs (solid state switches) in the sprinkler controller failed short (and supplied voltage to the valves) than that something happened to the sprinkler valves to keep them stuck open.  That is a typical failure mode for solid state devices (triacs, transistors, MOSFETs, etc).  If there is enough energy they open after they short but that takes quite a bit more energy - the wire connections inside the part blow out.
 
Those protectors look like they should be pretty effective - note that they need a good (i.e. short, direct) ground.  Largely ineffective without that.  Single point ground for everything where it comes in to the building is best.  As has been pointed out, it isn't always possible to predict exactly what will happen when a strike hits nearby.  But you can put what you consider reasonable protection in place to lessen the risk.

Click to expand...

 
Thanks for your posting.  That sounds very plausible.  I think you probably nailed it.
 
So, as added protection--in addition to the surge protectors the OP put forth--are there any sprinkler controllers that  fail to OFF instead of fail to ON for lightning strike/surge scenarios?  Anybody know?

For the reason given (protection against low-side surges), active-disconnect seems more bullet-proof, at least on the face of it: http://archive.lib.msu.edu/tic/holen/article/2000jun23.pdf
 
Notionally, it isolates your equipment if there's a threat of possible lightning by "unplugging" your equipment before the lightning gets too close.  Most likely you don't need to be irrigating if a lightning storm is deemed imminent, so why not "unplug" automatically until after the threat has passed?
 
Comments?

Isolating things as the paper suggests is a good idea.  I've looked at some lightning detectors that are designed for this purpose and unfortunately, the ones I've come across aren't exactly cheap.  There are some relatively inexpensive "personal" detectors that are meant for use on sports fields, golf courses, etc and I wonder if they could be adapted for use in a home automation system.
 
One other thought is to keep the sprinkler system isolated by default, and un-isolate it only when you are going to operate it.  Of course, you would still need to worry about whether a thunderstorm is around at that time.   But if you ignored that part, it would still significantly reduce the chances of damage since the system would be isolated most of the time.  And if you used MOVs to protect it while it was operating, you'd still have some level of protection.

I use the ZAP products for my Rain8 system. http://tmanufacturing.com. Had an issue a couple years ago with that and my pool controller.  So far so good (fingers crossed). For my Jandy I use a BB Electronics opto-isolator. It just protects the indoor pool controller and the serial communicator.

RAL said:

Isolating things as the paper suggests is a good idea.  I've looked at some lightning detectors that are designed for this purpose and unfortunately, the ones I've come across aren't exactly cheap.  There are some relatively inexpensive "personal" detectors that are meant for use on sports fields, golf courses, etc and I wonder if they could be adapted for use in a home automation system.

Click to expand...

 
There's this one-wire lightning detector that's not too expensive ($32):  http://www.hobby-boards.com/store/products/Lightning-Detector.html
Pete uses one, though as far as I know, not for this purpose.  Maybe he could comment on how well it performs.
 
A ghetto approach (not that I'm recommending it) is to tune an AM radio to an unused band and "listen" for the lightning that way.  I would hope that the above detector is doing more and better than just that....
 
So, you could monitor the lightning sensor with an arduino, or with your home automation equipment if you prefer.  Or, you could activate/deactivate your isolator automatically based on internet weather reports.  In that case, maybe you wouldn't even need a lightning detector, or maybe it would supplement a lightning detector if your internet connection happens to be working when it counts.    
 
So, the big question is:  what's the best disconnector to link it with?  As pointed out in the article, you probably want something better than just a relay, or the spark might (probably?) jump the narrow relay gap, almost like a spark plug.  Now that it's Y2014 and not Y2000 (when the article was written), what are the best disconnector choices other than relays?
 
Edit: according to Wikipedia, a "disconnector" might still allow arcing to happen, whereas not so with a "load break switch."  So, whatever it's called, we should all want the kind that doesn't arc.
 
I can see a good use for such a device, and not just for irrigation controllers.  Automatically switching other devices over to battery power and running wirelessly that way during an electrical storm just seems prudent.  Here electrical storms are typically short lived, and I confess to unplugging the power adapters on my family's computers and similar if I'm around and otherwise awake.  It may be a primitive precaution, and old-school, but I can bank on it working if it's done.  Anyhow, it would be nice if it could be done automatically....

There is a bit on MOV protection on the opensprinkler FAQ page.  This is a good first step but not as good as some of the other suggestions here.  The MOVs are placed in parallel with the triac switches to clamp the peak voltage.
 
http://rayshobby.net/?page_id=207

NeverDie said:

So, the big question is:  what's the best disconnector to link it with?  

Click to expand...

I guess the lack of response  means no one here is using disconnectors?

I would look for a relay that has the widest contact gap you can find.  There are some relays meant for use with solar panels and/or power systems that have gaps of 2.5mm or more.  I recall seeing some that were on the order of 6mm.
 
A 6mm gap will withstand a surge of 5kV or so.   But nothing will withstand a direct lightning hit.  Fortunately, direct lightning hits are unusual, and mostly what happens is an indirect hit, which can still induce significant surges in your wiring.  If you have MOVs or other surge suppression devices downstream of the relay to divert as much of the surge as possible to ground, the best you can hope for is that the MOVs will reduce the voltage that reaches the relay to a level that is low enough so that it won't arc over the gap in the contacts.

I wonder whether this type of relay might be somewhat better?
 

 
The author states, "One of the best ways to protect receivers and transmitters, besides having a good ground system and cable entrance feedthrough that is bonded to the mains ground, is to disconnect the feedline from the equipment. This disconnect has to be done properly....If you use disconnect relays or switches, they should be a double-make double-break style. A double-make double-break looks like [the above].  Lightning cannot pass through from side-to-side because the shorting bar is grounded when not energized. Any arcs across contacts would go to ground."
http://www.w8ji.com/ground_systems.htm