[Top] [All Lists]

Re: [TowerTalk] whole house surge protectors

Subject: Re: [TowerTalk] whole house surge protectors
From: Jim Lux <>
Date: Tue, 15 Apr 2008 16:37:51 -0700
List-post: <">>
Tom Anderson wrote:
> Jim and fellow tower talk members:
> I'm in the process of having a 220 line run into the shack for a new 
> Alpha 9500 amp my XYL and I picked up at the plant in Boulder a month 
> ago.  Still waiting on the electrician showing up though.
> While I had him on the phone I asked him (and yes I've known him for 
> years (I was his son's Eagle Scout advisor) and he's done work for me 
> before and does a lot of commercial work near us) about a whole house 
> protector and in his opinion he felt a good individual unit for the 
> plugs and equipment you wanted to protect was better than a whole house 
> protector. He said many strikes, surges, etc. come into the house via 
> phone lines, TV cable, satellite dishes, etc. and often bypass the whole 
> house protector before getting into the home electrical system, which is 
> why he prefers good single unit protectors when needed.

Not to be argumentative, but it would seem that the peer-reviewed 
transient protection literature doesn't agree with the single unit 
protection philosophy.
(i.e. things that aren't  sales material or field application guidelines 
from one company or another.. but stuff that has been independently 
produced, or at least reviewed by folks without a financial or other 
interest.  Sure, the guys from Erico publish papers on transient 
suppression, and Erico sells products that do this, but the journal's 
pre-publication review process generally makes sure that 2 or 3 other 
folks take a look at it and make sure it's not all bushwa.)

Your friend is right that lots of transients come in other ways, which 
is why you need protection on *all* of them, and why everyone is so 
obsessed with all those "bonding" rules.  However, there's a whole raft 
of reasons why protection at the entrance is needed, and preferred, and 
which should be properly coordinated with the downstream protection.

The guidelines for things like safety critical installations (e.g. the 
FAA guidelines and DoD guidelines) all talk about transient protection 
at the service entrance, but make little or no mention of "point of use" 

Also, a lot of protection devices get a lot of their protection ability 
by absorbing the transient (and getting spectacularly destroyed in the 
process).  I think I'd rather have the spectacular failure out by the 
meter, rather than on some plugstrip behind the desk.

Really, the only hazard that point of use protection protects against is 
a transient being induced in your house wiring closer to the equipment 
than to the service entrance, because of the time delays involved. 
(transients from outside, which are MUCH more common, get suppressed at 
the service entrance) Or, if you have something in your house that puts 
big transients on the power wiring (like that 1kW tesla coil you fire up 
in the garage.. been there, done that, cooked my garage door opener and 
a UPS.)

  Of course, a decent series mode suppressor (i.e. a LC low pass filter) 
will solve that problem.  Most modern electronic equipment already has 
fairly decent low pass input filters on the AC line connection (to 
prevent signals inside the box from coming out and creating EMI havoc), 
so the sub microsecond transient resulting from that spurious induced 
voltage, and reflecting back from the clamp will get smoothed out by the 
filter to a level that is lower than the damage threshold for your device.

For folks wanting to spend some time and about $20, get the book by 
Standler on overvoltage protection. (google standler overvoltage and 
you'll probably find it) He's got all the data, the impulse waveforms, 
the theory of what works and what doesn't.  His website also has some 
info.  I can't recall which off hand, but in one place or the other, he 
talks about the UL1449 standards, and the problem with "joule ratings" 
and "clipping voltage".

TowerTalk mailing list

<Prev in Thread] Current Thread [Next in Thread>