I need some help
 

What is the best power supply protection I can get? I'm confused. We live in an apartment so re-wireing is out of the question. when I ask some of the geeks I know I get a dozen different answers. is there a single outlet surge protector I can get, or is the power strip protector better? Some people say that I need an external ground. I need some help I will send you an autographed picture of my self for your help. :help: :redface:

The first question is what kind of protection you need. Do you have frequent power outages? Do you suspect your power of fluctuating?

Most people get by on grounded outlets and a standard surge protector, but if you have power wierdness, then it's a whole other ball game.

You do need a grounded (three prong) outlet at minimum.

The best would be a power-conditioning UPS, APC make rather nice ones.

I concur with jaguar, but would also mention if your outlet is incorrectly grounded or not grounded it may cause issues with the UPS (or if the polarization is mis-wired though that is less likely to cause a faut in the UPS).

My main concern is close lighting strikes we had one last summer that took out my TV and cable box. THX 4 all your help.

Then what you want is a good surge protector. Expect to pay $25-$50 for a good one. What to look for.

You might enjoy this thread which has TMI. http://cellar.org/showthread.php?t=9268

A standard 'computer grade' UPS does nothing to condition or 'clean' electricity. UPS in battery backup mode connects a computer directly to AC mains. Any AC electric transient gets connected directly into the computer. But no problem. Computers already have effective protection that might be attached to its power cord. That protection assumes you have purchased a 'responsible' power supply. Many clone computers 'forget' to install that essential internal protection to sell their inferior product only on price. Grossly overpriced plug-in protectors forget to mention that effective protection is already inside that computer.

Plug-in protectors also forget to mention that an adjacent protector provides typically destructive transients with more paths into and through a computer. Yes, even demonstrated how a plug-in protector adjacent to a computer damaged that computer and others in the network by replacing every damaged IC. Traces the electrical transient by replacing each damaged IC in a path that passes right through the plug-in protector.

How good is that internal computer protection? Look at an output when UPS is in battery backup mode. That modified sine wave is a 270 volt spike surrounded by 200 volt square waves. IOW 'dirtiest' power is from this UPS when in battery backup mode (which is why UPS manufacturers recommend *not* plugging a power strip protector into a UPS output). But those UPS created transients are no problem because a computer already has effective internal protection.

Provided in the "Power Squid" discussion was another damning plug-in protector problem:
http://www.westwhitelandfire.com/Art...Protectors.pdf
This ignored problem was ignored is but another reason why a power cord protector is not effective.

The UPS is for data protection from blackouts and brownouts. It claims hardware transient protection. Then forgets to mention 1) hardware protection is from transients that don't typically exist, and 2) it does not claim to protect from transients that typically damage electronics. Again, a plug-in UPS is only for data protection.

Appliances have internal protection that assumes building earths destructive transients. So what do plug-in UPSes and power strip protectors completely avoid mentioning? Earth ground. Why? Ineffective protectors are sold mostly on myths. They avoid the most essential component of a protection 'system' - earth ground.

Unfortunately, a protector is nothing more than a connection to protection - earth ground. That means a protector must be part of a building's wiring AND the building's earth ground must both meet and exceed post 1990 code requirements. So what is an apartment dweller to do? A kludge solution starts by locating a three prong wall receptacle closest to incoming utility power. (If the apartment only has two prong plugs, then the only alternative is to buy an effective solution and get the landlord to install it.) Then buy a protector with highest joules rating. Cut its power cord as short as possible and plug it into that wall receptacle. Then verify computer is powered from same phase. Computer is best connected to a wall receptacle as far as possible from incoming power (increased wire length between protector and computer increases protection).

Hopefully the building has a minimally sufficient earth ground AND that AC electric is grounded to same earthing used by a telephone line protector (provided free by telephone provider). If using cable, then hope that cable is also earthed to AC electric before entering the building. Notice what defines electronics protection - earth ground. A protector is nothing more than a temporary connection to protection - earth ground. A fact avoided by protectors that also don't claim effective protection - ie plug-in UPSes and power strips. Why discuss a technical fact that will only destroy sales?

One final point. Effective protectors have manufacturer names such as Square D, GE, Leviton, Intermatic, Siemens, and Cutler-Hammer. APC, Tripplite, and Belkin are most definitely not on the 'responsible' list. But those latter and ineffective products are recommended without first asking a simple question, "What do they do?" It is irresponsible to recommend APC products for surge protection. They do one thing well. Battery backup power is only for data protection; not for hardware protection.

How to identify an ineffective protector: 1) Protector has no dedicated connection to earth ground, and 2) Manufacturer avoids discussing earth ground. Plug-in protectors from APC, Belkin, and Tripplite violate both points.

Principles used by effective protectors were demonstrated by Ben Franklin. Same reason why commerical radio stations are routinely struck by lightning and don't fail. Same reason why emergecy response center operators need not remove headsets and leave the room during every thunderstorm.

Two factors that determine protection quality. 1) Earthing determines how a 'system' protects during each transient. 2) Joules determines protector life expectancy over many transients. Effective protection means a human never even knows the transient existed. "How to Choose a Surge Protector" only mentions joules and does not discuss the most critical protection component - earthing.

I admit I have no real technical knowledge in this area. However, my house has wildly fluctuating power problems, and in the first six months of living here, we fried three motherboards--on three different computers, the second two with expensive surge protectors and top-of-the-line power supplies after we realized there was a problem--simply by having them on when the lights flickered.

We purchased an APC brand UPS, and have not had another problem in the two years since. Like I said, purely anecdotal. No proof that the UPS is truly what solved the problem. But you also clearly have an axe to grind about the quality of power supplies.

APC
 

Get yourself a small UPS unit. APC is a good brand but any will do. It will protect against any electrical surges and provide enough power for you to properly turnoff your computer...
;)

Well now, I can't understand why everyone dislikes APC so much? At work I have a Silcon that we are replacing with an infrastucture (much more scalable) . Granted, the Silcon is the size of a small shed, and the new infrastucture is 4 full size racks with an 80amp 440 wired directly into it (the building is 3 phase). Now sure this is way more than you need but APC makes very nice conditioning and battery backup in one units for just a single workstation. They range from 60-100$. I highly recommend them, and I wonder why folks are badmouthing them.

the Back-UPS ES 500 for example is 60$ and it *is* also a power conditioner.... :)

Thanks guys I think my boss and I can now make an informed desision. :thankyou:

Flickering lights must never cause motherboard damage. That voltage variation must be made irrelevant by a power supply. In another thread, I noted how so many will recommend power supplies and yet not know of those power supply functions. China has discovered a profitable market. Americans without sufficient technical knowledge. Asian manufacturers dump power supplies into the clone computer market because so many Americans only buy on price. A cheaper power supply means higher Asian profits.

Even when a power supply completely and spectacularly fails, still, it must never cause motherboard damage. This was standard even 30 years ago. If flickering lights caused a power supply to damage a motherboard, then that power supply is rubbish. If a power supply does not cost $65 full retail, then it is probably missing essential functions and therefore can contribute to motherboard damage.

So let's see. We save $15 on an inferior power supply. Then must buy a $100 UPS to correct the power supply defect? I see this reasoning routinely where people buy on price rather than learn simple technical concepts.

Meanwhile, another problem that connects destructive transients directly to a motherboard. Bypasses a power supply completely. This required wire is why plug-in surge protectors contribute to damage of an adjacent computer. This wire is why a building's equipment ground (not to be confused with earth ground) must be properly wired and should be common to all interconnected boxes. This wire is also why other incoming utilities (cable and telephone) also must be grounded to a common point when entering a building.

Oh grandma! What big eyes you have. Oh grandma! What big teeth you have. This is also proof that APC conditions power.

Maffick, did you read the technical response - with numbers - before posting? This is your $60-$100 power conditioner: It connects computer directly to AC mains when not in battery backup mode. Where is the conditioning? Is that internal relay a power conditioner? That is what you have claimed. Its called Rush Limbaugh logic. It does not really exist. But the naive will promote it as fact anyway.

A big rack UPS that is properly earthed provides effective transient protection as well as clean UPS power. A plug-in UPS actually creates more transients than it (claims to) suppress. So where is all this power conditioning? I noticed you never once posted numbers - as Rush Limbaugh also avoids. Shame on you for even implying a plug-in APC UPS "conditions" power. Its called promoting myths.

TW - What's the difference between the ground wire in your basic three prong plug and the earthing thing you talk about? If a house is wired correctly, the ground wire does eventually connect up to something stuck in the ground.

TW, I bow to your superior technical knowledge but sometimes I want wring your neck. Get off your high horse and just tell me in plain English.
When I shop for a PC, how do I know what I’m getting? No, I like most people, I’m not building one. So by buying Dell’s most expensive PC, I’m safe?
My APC 500 isn’t protecting my PC from bad shit? My house system is wired properly with the 10’ copper spike to earth. What else do I need? :confused:

Anecdotal:

I installed an APC 1400 and an APC 1000 in my rack and they held up power for 5 different systems for 6 years. This included two events when power had to be generated by a portable gas generator. After 6 years one of them had a battery die. None of the systems ever had a power-related failure.

An APC 350 held up power for my main desktop but the power supply, an Antec True Power 480, seemed to develop a fault on the 12v rail. The system became unstable until the supply was changed out.

Electricity is not same at both ends of a wire. Electricity appears somewhat same at both ends of a wire in some conditions. For human safety, ground on a wall receptacle is same as the safety ground inside a breaker box. But for transistor safety, both ends of this wire are not electrically same. The technical reason is wire impedance; not to be confused with wire resistance.

Destructive surges are not stopped, blocked, or absorbed - despite what plug-in protectors would appear to claim. As Franklin demonstrated, electrical transients must be shunted (redirected, diverted) to earth ground. For Franklin lightning rods, a connection to earth must be short, no splices, no sharp bends, not inside metallic conduit, etc.

Surge protector does same; shunts transients to earth. During a surge, a protector is but a wire. Connection from each AC electric wire to earth ground must be short, no splices, no sharp bends, etc. Protector makes that connection IF protector is properly earthed.

'Whole house' protectors are effective when connected less than 10 feet to earth ground. 'Whole house' protectors from responsible manufacturers such as Square D, Cutler-Hammer, Siemens, Intermatic, Leviton, and GE are installed at the service entrance. Earth ground wire from that mains breaker box to earth must be 'less than 10 feet', no spliced, no sharp bends, not inside conduit, and separated from non-earthing wires. Requirements that both meet and exceed post 1990 National Electrical Code (NEC) requirements.

Note manufacturer names specifically not mentioned: APC, Tripplite, Belkin. Products have no earth ground connection AND they avoid this discussion. Instead, they promote sound bytes and 'word association' as a replacement for good science and the all so critical earth ground.

Return to that equipment ground at a wall receptacle. That equipment ground wire is bundled with other non-earthing wires. It has numerous sharp bends. Too many splice. Far more than 10 feet from the earth ground rod. In short, a wall receptacle equipment ground, obviously, is not an earth ground. No earth ground means no effective protection.

Your APC does not even claim to protect from the typically destructive transient. They use classic propaganda techniques so that you will assume such protection. The APC UPS does one function - battery backup of data during blackouts and extreme brownouts.

Listed were protectors that can make that 'less than 10 foot' earthing connection. More responsible retailers such as Home Depot (Intermatic) and Lowes (GE and Cutler-Hammer) sell these solutions. Also most electrical supply houses. Never saw an effective protector sold in Radio Shack, Kmart, Sears, Staples, WalMart, Office Max, Best Buy, Circuit City, or Target.

Meanwhile, 'whole house' protectors are so effective and so inexpensive as to be provided, free, by the telephone company in your NID. This, too, will only be as effective as the earth ground 'you' have provided. CATV wire requires (and best has) no surge protector. Protection is provided by a hardwire connection from cable ground block to, again, the common point earth ground.

All incoming utilities (including satellite dish) must enter at the same location to make that 'less than 10 foot' connection. Otherwise a building's earth ground must be enhanced as demonstrated by Cinergy in this figure of wrong, right, and preferred earthing:
http://www.cinergy.com/surge/ttip08.htm

Notice what defines protection: earthing. Pictures demonstrate what Orange County did to correct frequent surge problems - it's all about shunting to earth as Franklin demonstrated in 1752:
http://www.psihq.com/AllCopper.htm

Thanks, tw. At least now I know.

I'm in an older home, though, and that sounds like a lot of electrical work to add after the fact.

How could one possibly know whether Comcast grounded their box correctly?

Holy Shit! In Florida they had to sink the ground rod 120 feet!!! :smack:

I once sat for coffee with two Comcast installers who had just returned from training on this subject. Previously, no cable installers (apparently) had any idea what earthing (a 1930 technology) was about. As I explained these various points to them, the one kept saying, "So that is what they meant" and "So that is why we are suppose to do that". Like electricians, these concepts were completely foreign to most all installers.

The ground block on a cable wire is shown below. From this ground block (that is sometimes inside a gray box that had nothing but this ground block), a 10 AWG wire connects to an earth ground that you (the homeowner) must provide. If you did not provide this common earthing point, then Comcast, et al may install something inferior.

That earth ground, at minimum, must be a 10 foot copper clad rod (sold in any Home Depot or Lowes for about $10). Too often electricians or installers will even cut an 8 foot rod in half. That four foot rod is woefully insufficient earthing. A rod that is not solidly in earth (can be removed) is also disconnected from earth. Earthing concepts well proven before WWII and now made essential for every home are too often compromised by those who never learned how and why.

(BTW, those three light ground testers report nothing about earth ground.)

One rod is a major earthing improvement. Anything additional provides exponentially decreasing improvement. That Orange County Emergency Response center must be so reliable as to spend $thousands more on a 120 foot ground rod just to make the last 0.1% of lightning totally irrelevant. But a residential owner must, at minimum, install that earth ground rod with a 6 AWG bare copper wire that connects less than 10 feet to breaker box.

We have now identified both ends of the cable earthing wire. Returning to those Comcast installers' questions: let's say a cable connects only to a second floor TV. So they attach cable at second floor level, and then run an earthing wire from that cable down to an earth ground rod. They have just made that TV a target for damage. Incoming cable wire must attach at second floor level, drop down close to the common earthing electrode, make a less than 10 foot earthing connection, then rise up (separated from the 'drop down' wire) to attach to that second floor TV. Cable must attach short to earth ground before entering a building AND must separate its 'before earthing' section from 'after earthing' section. The 'less than 10 foot' rule is essential.

That wire from ground block to earthing electrode must have no sharp bends, no splices, must remain separated from all other earthing wires until they all meet at the earth ground, AND must remain completely separated from non-earthing wires. Too many installers want to make things neat. They nylon ty-wrap (bundle) wires together and make neat sharp bends. An old rule: neat often means an inferior installation. Neat can also make induced transient damage likely.

Pre-1990 buildings use water pipe as earth ground. No longer acceptable and not sufficient. Quality of an earth ground is not just in better conductivity. Distance from each incoming utility wire to that common earth ground is most important. That every incoming utility connects to the same earthing electrode is essential. Yes, that means even a satellite dish cable or TV antenna must first connect to earth ground before entering.

This is a ground block that would be inside that Comcast gray box. Earthing wire (often gray or green but in this case black) connects to this ground block:

Isn't that about 95 feet into the water table down there? :)

Since my cable installer did nothing of the sort, my best bet is to throw Comcast out and tell the satellite installers how to do their business.

My cable comes down the pole and enters a watertight thermocouple head on a plastic pipe, 4 ft up the pole, then comes 40 ft under ground into the cellar. Then it travels 8 ft, makes a 90 and travels 3 ft to the grounding block before splitting into three legs. The wire from the ground block is a straight 18 ft run to the service entrance ground at the breaker box.
Doesn't sound like much lightning protection. :smack:
Oh well, maybe the lightning will fall off the drip loop at the thermocouple head.

18 ft wire is cable company just barely meeting NEC requirements for human safety. If concerned for household transistor safety, then utilize a solution demonstrated by Cinergy in their 'bad, good, and preferred' pictures cited earlier.

Even underground wires provide no transistor protection. Even underground wires must be 'surge protected' by connecting to a common earth ground point at the service entrance.

Above posts define secondary protection. Inspection of the primary protection system also may be necessary:
http://www.tvtower.com/fpl.html

Curious. How many actually know what a drip loop is?

That's nice, but you left out the part where for the second and third computers we bought high-end, non-Asian-market surge protectors, and they also didn't work. Or perhaps you were speaking hypothetically?

Dang this thread has been a great read.
props, TW!

First every shunt mode plug-in protector is or is same as the "Asian-market surge protector". There is no magic in these devices. They all use same circuits discussed in that 1970 GE Varistor Manual. Furthermore, most every MOV is now manufactured in Asia.

Shunt mode plug-in protectors - cheap or expensive - are same if they have same joules rating. However many love to be scammed. They may pay $100 for the Monster Cable model that is electrically same as a $20 model from others.

Why did a protector not work? From previous posts: One final point. How typically is a modem damaged. Remember, the phone line already has a 'whole house' protector provided by Verizon, SBC, Qwest, et al.

Lightning strikes an 'antenna like' device - AC electric lines above the street. Incoming through breaker box seeking earth ground. Incoming on computer's AC wires. Bypasses power supply to connect to motherboard ground plane. Through modem, out phone line, and to earth ground via the telco 'whole house' protector.

Part most often damaged on modem is its DAA section - section adjacent to phone line. In particular, the PNP transistor that drives an off-hook relay. A small transient will leave the PNP transistor shorted. When computer is powered, then phone line is always connected through modem. Phone line acts like a phone has been left off hook.

A large transient opens that PNP transistor. Therefore computer complains about "No Dialtone Detected" because off-hook relay never connects modem to phone line.

Some are quick to assume rather than first learn underlying theory. They assume this transient entered on phone line because only modem was damaged. But notice an essential requirement. To have an electrical transient, both an incoming and outgoing path must exist. Incoming on AC mains. Outgoing on phone line. Only after transient is passing through everything in that circuit does a part somewhere in that circuit fail. Besides, why would the transient enter on phone line when phone lines already have a 'whole house' protector.

The above example demonstrates why transients on AC electric so often damage modems, fax machines, answering machines, and portable phone base stations.

look at what these folks have , I have ALLWAYS had good luck with their products , not the cheapest but I can PERSONALY recomend them .

http://www.solaheviduty.com/products/index.htm

I work on scales ,a scale ( old school analog ) is like a volt meter that goes out 7-10 decimal places , now put that in a steel mill ( think of those Big towers with power lines , they transmit KVA ( KillaVoltAmps )) or at the end of a long transmittion line AFTER 2 steel mills or just WAY out in BumFuck Egypt ,
Steel mills use HUNDREDS of KVA power per electrode ( 3 generly , but ther are some that use 1 big electrode ( DC ) , this can SERIOUSLY Fuck with the power !!!!!
Sola Isolation trans formers clean up the power about 90+% , and YES TW I have used a power analizer to test this befor .
Sola also has some of the best REAL world surge proctecters and UPSs on the market , NOT cheap but if you are looking for 1 sorce for power conditioning you can do a WHOLE lot worse than Sola .

Just my 2 cents .

here ya go Clod , this should help ,
http://www.solaheviduty.com/products...cr_mpcport.htm