DIY surge protection...

Surge protectors are not capacitors. They are made from material that will conduct electricity when the voltage exceeds some particular design value. That excess electric power is converted to heat in the surge protector. If the "surge" or spike is too long lasting or occurs so often that the surge protector does not have time to cool, it will eventually produce smoke and stop working. At that time any and all surges and spikes will continue on to the rest of your house.

Usually the surge protector will die without you knowing about the death. There is no way to test them without a spike generator and an oscilloscope.

The power spikes can come from anywhere. I personally experienced equipment destroying spikes that came from the telephone wires. A construction company was excavating very deeply for a sewer pumping station near my office. Somehow they connected 220 volts to the buried telephone cable. The power went through the local phone company junction box and into our phone system and fax machine. The surge protectors immediately absorbed all the power they could and produced smoke. Then the power continued on to burn out circuit boards in the equipment.

We only discovered the source of the problem when a few days later I discovered a telephone guy installing a new junction box near our office. He told me about the construction company problem and how they were paying for the damage. They also paid us.

So, bottom line is the protectors are probably a one-time only protection. There is no easy way to test. The surge may come from an unprotected source. This applies to all protectors, including all- house protectors. All lines coming to a house must be protected, Not just the "hot" lines.

The "iron rings" you refer to are ferrite RF supressors. They reduce the electronic noise generated by switching power supplies.

Paul

They are not capacitors. They are electronic-semiconductor devices that are open circuit until some voltage threshold is exceeded, then they act like a very low resistance to try to limit the voltage. The limiting factor is the amount of power the devices can withstand before exploding due to the heat they generate when acting as s short circuit. I don't know a lot more than that, except that they are usually rated in Joules of energy they can dissipate before blowing up. They certainly cannot handle a direct strike to the power line, but induced voltage spikes due to nearby lightning can be handled if the joule rating is high enough.

I have a few whole-house (well, whole branch circuit) UL-rated surge arrestors. They are big (4" long by 2.25" diameter) plastic cans that attach to the main breaker box, and are wired into the branch circuits that they protect. The cans cost something like $50 each from the local electrical supply house, and are made by an outfit in Texas. I have a pair of their Model 302 arrestors.

The website doesn't work right for Safari or Firefox in MacOS, so it's probably MSIE only. But you can make it work anyway, with fiddling and indirection.

They are not just capacitors, they are industrial-size metal-oxide varistors plus capacitors.

No.

Probably ferrite EMI-supression "beads", which have no effect on computer-smoking surges. Yes, it's hot+return.

Joe Gwinn

They are not capacitors. They are electronic-semiconductor devices that are open circuit until some voltage threshold is exceeded, then they act like a very low resistance to try to limit the voltage. The limiting factor is the amount of power the devices can withstand before exploding due to the heat they generate when acting as s short circuit. I don't know a lot more than that, except that they are usually rated in Joules of energy they can dissipate before blowing up. They certainly cannot handle a direct strike to the power line, but induced voltage spikes due to nearby lightning can be handled if the joule rating is high enough.

====================================================

So the surge protector is a kind of crude voltage regulator?

Well, how about this:

Why not put a 100 A relay in the service, with the coil connected to a fast-acting voltage-sensing amplifier. If the voltage goes up by more than, say, 10%, the relay is activated (or deactivated, if NO), all power to the house is broken, with the relay latching out, requiring a manual re-start. Proly a NO relay.

A little more dramatic/intrusive in its action than surge protector, in that power is removed, but it should do the job, protection-wise. AND this would have the advantage of being re-usable essentially forever, and also testable.

If you wanted to get fancy, you could have this coordinated with a UPS and generator, so that no perceptible power interruption occurs. Much more $$, of course.

Relays are mechanical devices and, as such, operate in slo-mo compared to electricity.

What do fast-acting UPS's use? Use the same thing? Mebbe solid state relays?

They're proly cheaper than mechanicals, by now, and I would assume pretty fast. There's an A/C supply outfit that sells, iirc, a 2 pole 50 A jobby for under $20. The neat thing is, the "coil" is good for, like, 100-300 V!!!

Hmmm, So you think coil driven mechanical relay is as fast as spikes or surge? Forget it. But there is such a thing call S.S. relay. and it is not a voltage regulator it is a limiter.

Paul sez::

""The power spikes can come from anywhere. I personally experienced equipment destroying spikes that came from the telephone wires. ASo, bottom line is the protectors are probably a one-time only protection. There is no easy way to test. The surge may come from an unprotected source. This applies to all protectors, including all- house protectors. All lines coming to a house must be protected, Not just the "hot" lines.""

Yep! Call them, "Fail dead and burned open" with usu. no visible way of determining when failure occurrs. Yeah, I know some have a pilot light but it is easy to ignore.

Bob Swinney

MOVs = Metal Oxide Varisters

Spikes can go WAY bigger than a paltry 300 v And the unsuppressed arc across a relay contact - 30KV?

You were probably still a baby when the whole thing about power protection got started. There was a HUGE argument between marketing people and engineers relating to "UPS". Marketing called them uninterruptable power sources (UPS) and engineers demanded they be called Stand-By power sources(SPS). All the things you buy today are really stand-by power sources. They have a real mechanical relay that switches from the power line to battery source. A real UPS will cost many hundreds to many thousands of dollars. They continually supply power from batteries and the AC just keeps the batteries charged. An electronic circuit keeps the internally generated AC synchronized to the external power frequency. The marketing people finally won the battle. Guess it was the money, not the truth.

All computers and associated equipment, today, used switching power supplies which can continue to operate during the 2-4 cycles it takes the mechanical relay to switch and the time to start the electronics to begin supplying AC power.

Paul

the microseconds. Relays take thousands of microseconds to operate.

Here's a traditional answer:

repairmen used them to make hot-chassis sets safer to work on. I have one but use a UPS instead.

jsw

I agree with you, but you can get a "true online" UPS for less than several thousand dollars (but more than the typical thing you'll find at your local computer store)

The keywords that differentiate the two are "line-interactive" (as you describe) and "true online" (which powers the UPS receptacles from the batteries through an inverter 100% of the time.)

Now when you talk about true online UPS units, you also have to consider the quality of the power that comes out of them... need to find one with a GOOD inverter that produces a nice sine wave...

nate

You talkin' to ME, kid???

Whoa! At last EA is showing his true intelligence exposing his level of ignorance. 'nuff said.

Paul sez:

"There was a HUGE argument between marketing people and engineers relating to "UPS". Marketing called them uninterruptable power sources (UPS) and engineers demanded they be called Stand-By power sources(SPS). All the things you buy today are really stand-by power sources. They have a real mechanical relay that switches from the power line to battery source. A real UPS will cost many hundreds to many thousands of dollars. They continually supply power from batteries and the AC just keeps the batteries charged. An electronic circuit keeps the internally generated AC synchronized to the external power frequency. The marketing people finally won the battle. Guess it was the money, not the truth."

One very knowledgeable cohort of mine used to speak of them as real UPS's and "chicken UPS's" During that time, I was project manager on a couple of large UPS installations, both on PBX plants. One was a 4000 line PBX and the other was 2000 line. They were Lorraine Electric units, uninterruptible in every sense of the word, with huge lead-acid batterys. As I recall the batts in the 4000 line unit were sized for nominally 24 hours.

Bob Swinney

"chicken UPS's"

installations, both on PBX plants.

I recall the batts in

Hi, In my working days in radio telcomm. UPS was composed of battery bank, motor-generator set and control(switching) unit. I don't recall we ever suffered radio link outage. This is true UPS.

Here's a very good document on home protection written for the non-electrically inclined-

Pay particular attention to the section on GPR, there's a lot of people in the industry who, while they may know about it "intellectually", don't really think about it enough. :(

Suppressors don't just protect against lightening but also against transient spikes on the power lines induced by heavy equipment etc.

How much protection you get depends on how much money you spend. The cheaper MOV "little black box" units that Mr. Holme's electrician is so in love with (he does do neat wiring, though :)) are good for the occasional spike, if you live in an area prone to lightening & you own a lot of $$$ electronics you might want fork out for an industrial strength unit-

but figure on ~$1000 for a top of the line one with SASD devices that will stand up to the abuse.

H..

No, it isn't. A "true" GPS, be it motor-gen or otherwise, is not the genuine animal unless it's output is being generated constantly whether commercial power is present or not. A motor-gen UPS would not be fit the true GPS definition unless the generator ran constantly from its source of power, battery or commercial. In the "true" GPS, switching is implemented only in the event of primary power failure; and then only to interrupt or restore primary power to the batteries charging component.

Bob Swinney

"chicken UPS's"

installations, both on PBX

I recall the batts

Hi, In my working days in radio telcomm. UPS was composed of battery bank, motor-generator set and control(switching) unit. I don't recall we ever suffered radio link outage. This is true UPS.

Either you buy a protector that will somehow absorb all that energy. Or you buy protectors based upon how it was done even 100 years ago. Protection is always about where energy dissipated. Either that energy remains outside the building. Or that energy is inside hunting for earth ground destructively via appliances. Adjacent protectors simply give surges even more potentially destructive paths through adjacent appliances.

An effective surge protector means even the protector remains functional. A minimal 'whole house' protector starts at 50,000 amps. Direct lightning strikes are typically 20,000 amps. Yes, the protector must be sized to even earth direct lightning strikes and remain functional. And that means the connection to earth must be additional requirements - short ('less than 10 feet) to earth, no sharp wire bends, no splices. all protectors meet at (again 'less than

10 feet to') the single point earth ground, ground wires separated from other non-ground wires, not inside metallic conduit, etc.

Protection is always about where energy dissipates. If those hundreds of thousands of joules dissipate in earth, then no damage. This is how it was done even 100 years ago.

But somehow a magic box next to the appliance will absorb all those joules? Always view the tech specs. Plug-in protectors rates at hundreds of joules will somehow make hundreds of thousands just disappear? That is what they claim. In analysis, we even traced surges earthed destructive through a network of powered off computers because the surge was permitted inside the building. And because a surge on the black (hot) wire was connected directly to the motherboard by the protector. The protector bypassed protection inside the computer's power supply.

Telcos do not waste money on protectors adjacent to electronics. That switching center must never suffer damage. A switching center, connected to overhead wires all over town, may suffer 100 surges with each thunderstorm - and no damage. Why? Each protector connects short to the single point earth ground. And the protector is up to 50 meters separated from electronics. That separation increases protection.

No protector is protection. None. The only effective protectors make that short connection to single point earth ground. Ineffective protectors (a $3 power strip with some ten cent protector parts selling for $25 or $150) are profit centers. The NIST (US government research agency) discusses those ineffective protectors by describing what every protector must do:

by diverting the

grounding is

The NIST describes plug-in protectors as "useless". Obviously. It does not even claim protection in its numeric specs. Find those spec numbers that list each type of surge and protection from that surge? No plug-in protector makes protection claims. They are a profit center.

Protection is always about where energy dissipates. IOW why facilities with effective protection both meet and exceed post 1990 National Electrical code. Where does energy dissipate? A protector is only as effective as its earth ground - which no plug-in protector has and therefore will not discuss. Effective 'whole house' protectors come from General Electric, Keison, Intermatic, Siemens, Square D, and Leviton. An effective Cutler-Hammer solution sells in Lowes and Home Depot for less than $50.