This is a place to sort of continue a lively discussion on a range of topics. The original question was kind of simple. Many helpful users offered the poor guy too technical of answers for him. Some of the posters did offer different points of view. As suggested, the topic is hardware even if not fully a linux issue.

http://www.linuxquestions.org/questi...ectors-945914/

Not sure we ever helped the OP in that but anyone who wishes to add to that is welcome here.

Thanks.

So what is a surge protector good for? Why would I use it? Any cost issues? Code issues? NFPA or NEC quotes? Whole house versus point of use? Joules, I think bad.

I would like to kind of keep lightning as a static component out. I'd like to keep this more of an effect of lightning on copper in volts, current. Any other cause of surge, spike, transient, freq offset, square waves, modified sine waves, harmonics, and such is welcome.

These rules do not describe "point of entry" or "whole house" protection. That kind of protection is the best protection, but is also very expensive, often requiring moving where wires and pipes enter the building. But in this post I will describe "point of use" protection.

All sensitive devices that can be damaged or destroyed by a surge, which will have wired connections between them in ANY form whatsever, need to be considered together as a group for the purpose of surge protection. ALL wires (no exceptions) going between any of these devices and outside this group need to all run together through ONE SINGLE protector device. The first obvious wiring is power (see below for battery powered devices), so we are talking about power strips. Additional wiring from outside, such as coax cable or twist pair phone wiring, must ALSO be connected through the (power strip) protector. Thus you must get a protector with connections for the type of outside wiring you have.

You can cascade additional power strips after the one providing primary protection. But the other wires must be connected to the FIRST protector. DO NOT use two parallel protectors plugged into separate outlets.

The group wired together must not exceed the capacity of the ONE OUTLET they are plugged into. If the group is so large it exceeds this capacity, consider rewiring to raise the capacity up to the practical limits according to electrical code or regulations. In certain countries using 110 to 120 volts, if all devices can be operated at 220 to 240 volts, and if protectors can be obtained that are designed for this voltage, wiring a special higher voltage circuit may be a viable option for increasing capacity.

If you have a CAT5 ethernet LAN through your house, this wiring must be treated as "outside of the group" wiring for the purpose of "point of use" protection. The fact that it does not exit the building does not matter. You will need a power strip surge protector with ethernet connections, which is rare (but I do have two of them myself, so some exist). I got mine from HERE but they are listed as sold out while I am posting this

Since posting, I have found a couple surge protectors that may provide a good range of protection:

100-127 volts: http://www.tripplite.com/en/products...xtModelID=4134

220-240 volts: http://www.tripplite.com/en/products...xtModelID=4180

In a typical home, the entertainment system is one group (VCR and DVD and Bluray connected to a TV, along with a stereo and speaker). Cable TV or Satellite TV is there and involves a feed from outside the building. Connect the cable or satellite coax and ALL the power connections through ONE SINGLE surge protector. If more power outlets are needed, connect additional power strips or power splitters into this ONE SINGLE surge protector, or get a larger single surge protector.

Additionally, the coax entering the home for either cable TV or satellite MUST come in as part of the "point of entry". Even if you don't currently have a full "point of entry" protection system, at least run the coax in right next to your electrical panel. Use a coaxial grounding device and connect it to the grounding wire (the one that goes to the grounding rods) where it comes out of the panel. From there run the coax inside the building to where it needs to go AND connect it to the "point of use" surge protector at the destination (before it goes to the set top box or satellite receiver, which is plugged into that same ONE SINGLE surge protector).

In a typical home, a computer desk might have a computer, monitor, and external backup drive that need power. Connect everything through one single surge protector because these things are going to be connected to each other. If you have home CAT5 wiring, connect it through this same ONE SINGLE surge protector. Anything else that comes into here and gets connected to the computer must get its AC power from that same surge protector (unless it uses its own batter).

Laptops, netbooks, and other mobile devices with battery power, are mostly isolated from surges. When they are plugged in to a charger or AC adapter, or connected to anything like a CAT5 cable, they must be treated like any other AC powered device, and protected with a surge protector. If ONLY CAT5 is connected, it still needs to be protected even if the protector is not plugged into power.

Even if the device is not connected to anything, very close lightning strikes can still damage the extra sensitive radio circuitry in the wireless parts, though this is very rare (it has, however, happened to me). There is little you can do about this but keep them under your tin foil hat when not in use

Getting into theory.

Electricity does not flow very much when going out to the end of a wire that is not connected to anything on that end. Since the flowing current is what damages devices, keeping equipment on the "end of one wire" is what is protecting it. You want to avoid any voltage DIFFERENCE between any two points within the group of connected devices, to avoid the current that would flow if there is a voltage difference.

Virtually every wired interconnection has 2 or more wires. You must also avoid a voltage difference between these wires. Fortunately, the voltage levels that generally start to cause damage for short term surges is somewhere between 350 and 700 volts. This allows us to connect power wires and signal wires together though MOV parts that do not conduct electricity until the voltage reaches a near damaging level. At the level where damage may be possible, these MOV parts conduct electricity and keep all the wires at the same voltage. Any current that would flow between these wires now flows through the MOV parts. But distances must be short for this to work, so this is why it is necessary to do all the protection within a single device like a surge protection power strip.

Never connect any point of use protection system to ground, other than through the ground wire that comes with the power. The purpose of connecting to this ground wire is ONLY for including it in the voltage equalization because it is part of the wiring. Grounding to earth at "point of use" will only increase the risk of surge damage because it will provide a means for high current flow during a severe surge. The "point of entry" protection system is where ground is to be connected.

A surge MAY also cause a circuit breaker to trip. Surge protection does NOT use a circuit breaker tripping to provide the protection. Tripping a circuit breaker means a failure of some kind. If this happens, the surge protector is likely to be damaged and should be checked. If it has no indicator lamps to show if it still works, dispose of it after this kind of event, or mark it as "NO SURGE PROTECTION" and only use it for power strip purposes, if it is not shorted out.

Many users are in some industry where a home and a business conditions are very different. For example we had a 50 HP motor VFD single phase that affected systems back to the power station. It left the building with only 2 phases. In the process of trying to stop the motor it sent a surge into almost every machine. No point of service could have prevented the VFD's feedback into the other systems. A point of use ups was on many computers and protected them. Not all systems had them one.

A typical UPS connects the computer directly to AC mains. It takes milliseconds to disconnect from AC mains. Meanwhile a surge is done in microseconds. Most computers have best protection inside their power supply. Rated at thousands of volts. Necessary because a UPS cannot possibly disconnect before a surge overwhelms that superior protection.

Hi,

I think a little clarity on the type of UPS available is due(white paper was dated but still applicable since it has been updated);Each type is defined in the white paper and along with simple block design for each.
HTH!

I guess I should get an idea of the exact type usp we tend to have. Generally they are spec'd by the original install contractor and we tend to try to match it when available.

It gets rather easy. If not on the order of $1000, then its output in battery backup mode is as demonstrated by waveforms provided by this utility:
http://www.duke-energy.com/indiana-b...ech-tip-03.asp

Does not matter how 'clean' that output. Because electronics first converts that to much higher voltages. And then to high voltage, radio frequency spikes. Far 'dirtier' than anything out of any UPS. No matter how 'dirty' a UPS makes power, the applinace's supply makes it even 'dirtier'. And then turns that much 'dirtier' power into rock solid and stable DC voltages.

A UPS has only one function. To provide temporary and 'dirty' power during a blackout.

Then you missed the link posted by onebuck. You should read that first before saying there is only one use for a ups.

The many forms and functions provided by a UPS is a function of the use and intended protection.

We have whole room UPS systems in every plant, serviced by factory techs generally as they are the only ones that can get parts. They provide stable power, filtered power, surge protection and backup. I have to generally shut down less important systems to reduce loads when city power is lost. These systems were installed while the building was designed. They are not a trivial product by any means. The batteries for one room cost almost $90K last time we changed them.

We have thousands of true sine wave systems that were spec'd by the contractors. Could they run on modified sine wave? They may run on modified sine but when it is a warranty issue worth hundred of thousands of dollars I am not going to test it.

We also have some of the lower modified sine wave devices for not too essential stuff. I prefer the true sine wave devices in all cases even though the price is 4 or 5 times the cheap stuff.

onebuck's link pointed out the many versions of usp's. That information is not always clear on the sales info.

Hi,

I had to maintain a 110 unit battery rack along with a 'Double Conversion On-Line' system. 'Delta' was not cheaply available at the time. Those batteries were not cheap and very heavy construction. Even though the batteries were lead-calcium the racks were isolated and vented for safety sake (human). The interconnection is important so you had no drop at any of the terminals.

UPS & rack filled two rooms with conditioned isolated air flow. All this for a continuous source for 6-8 hours run time.

This system was built in the early 80s' and cost around $50K. We built the system and racks. I do not remember the costs for each battery but I'm sure they were not cheap. Laboratory grade batteries and hard to get your hands on at the time. We had a HVAC company build and install the vent/Air Conditioning system for each room. We provided the control for that subsystem. Fun but a difficult system to put together and maintain safely.

I agree that there is too much FUD when it comes to UPS and line conditioning.

View those waveforms cited previously. Those square waves are also a sine wave output. No true sine wave exists. Even high school math explains why square waves are only a sum of pure sine waves. If a UPS has a ‘true sine wave’, then that wave is spec’d by numbers that any layman can post. Such as %THD. True sine wave is a subjective term so that a salesman can sell a myth. Define 'true sine wave' with specification numbers.

Many UPS technologies come down to one purpose for an adjacent computer. To provide temporary and 'dirty' power during a blackout. Spending $1000 to improve power from a $100 UPS accomplishes nothing. But again, no matter how 'clean' that power, the computer simply converts power into something 'dirtier'. Then makes that even ‘dirtier’ power into ideal DC voltages. Square waves from a cheapest UPS is often more than sufficient - if learning from spec numbers.

I have a UPS called a sine wave output. Its sine wave output looks similar to waveforms in "Tech Tip 03 Standby Power Supplies Can Lock Up Your PC". Its poor %THD is also ideal power for a typical PC.

We are not discussing a building wide UPS system that cost many $thousands and that performs other functions. We are discussing the typical UPS. Each UPS technology still boils down to what that typical UPS does: provide temporary and 'dirty' computer power.

Now, if your computer needs other electrical anomalies solved, then define that anomaly. Explain why any typical computer users needs that solution. And again - with a perspective only possible with numbers.

It might be a better term to call this 'dirty' power 'digital power' since a switching mode power supply just eats it up.

A UPS I did have many years ago did have a much cleaner waveform unloaded. It got a little worse well loaded, but not as bad as the one shown in the link. It was an online dual conversion type, but that just means whatever it does it does it all the time.

I don't know what the two UPSes I have now do. They are cheaper units. I wish I had that previous one back to use on the entertainment system (it's battery had died just before I was moving and a strategic decision at the time was to let it go).

If the UPS spec numbers do not state how 'clean', then assume the worst. But the point remains. AC mains routinely provide 'cleaner' power. 'Dirtier' power happens when a UPS disconnects from AC mains and powers directly from battery. That 'dirty' power is still sufficient for any properly designed electronics.

A typical UPS has only one purpose. To provide temporary and 'dirty' power during a blackout.

Questions were also asked about other anomalies. Each is completely different. Must be addressed separately. No device solves all anomalies. Most anomalies are made irrelevant by what already exists inside electronic appliances. Blackout is addressed by a UPS. Some other anomalies can only be addressed elsewhere.

A building wide UPS can address more anomalies due to its location.

Hi,

That too will depend on the type of circuit design for the UPS and cost for same. Most consumer grade UPS are standby design thus minimal filtering or just poor line filtering while live. Hopefully on run the filtering and power specs are provided by the UPS if not the source will indeed be 'dirty' with losses thus failing.

Any time someone who reaches for a UPS that same user should be aware of the needs for his/her circuit. Meet the criteria specifications for the system to be serviced. Simple line filtering can provide cleaner 'source' from the utilities at a lower cost. But that too requires one to know how and what to do.
You are not specifying but making a general misleading statement. Something you earlier condemned. What is a 'Typical' UPS too you? How do you apply that statement to a general user that is unaware of circuit needs or specs?
The design criteria and specifications for the circuit will direct a user to the proper UPS design. Or it should. Most users fall into the trap of spin for the marketeers. UN-knowingly accept the spin without looking at the numbers(which most cannot interpret) and not making the correct decision for the type of UPS necessary for the site or circuit in need of protection.

Broad brush here! Not everyone would be in need of building wide system nor can most afford one. As I have said before, line conditioning/filtering can help most users at a lower cost by isolating the circuits that are in need of service.
What happened here? Your statement and stance seem to be varied. I read through that post and tend to agree but there are some irregularities with your points here. Not wanting to argue but to get clarifications here.

For members;
Please read the rest of the wiki to get a good picture of most UPS designs and usage. Good diagrams and UPS definitions.

Measuring efficiency
Spin can be deceiving to get your $$. Buyer be aware! Please notice the underlined above.

HTH!

A 'dirtiest' UPS (which has no filtering) is also called ideal power for electronic appliances. Even 1970 design standards required circuits to be that robust. Design criteria for electronic appliances says a 'dirty' UPS is sufficient power. Why would anyone have to learn circuits?

A more expensive UPS for cleaner power does nothing for or to protect electronics. Most who recommend a $1000 double conversion UPS do not even know 'cleaning' is undone inside electronic appliances. Best and superior filtering already exists in the appliance. Filtering that means any UPS is 'clean' enough.

That UPS has one purpose. To provide temporary power during a blackout. Each UPS is sufficient. Why would anyone spend $1000 for a UPS when the $100 version is sufficient? Because hearsay recommends it? A UPS provides temporary power during a blackout. Due to circuits already inside electronics, any UPS must be sufficient. Even 'dirtiest' power from a UPS is made irrelevant by circuits required by design standards. No consumer need design criteria and specifications for the circuit. Electronic appliances must operate properly. Because design standards demand electronics be that robust.

Hi,

No one said you must learn circuits. But at least compare specifications to make a wise selection. If you/member is not able to discern then ask here. Enough members that do have 'circuit' experiences to aid instead of FUD/hearsay.

Sad design criteria on your part. I have worked in some of the 'dirtiest' environments from a power stand point. UPS were specified and used to prevent damage/disruption to expensive commercial grade equipment. I would never have specified a UPS without including/supplementing proper line conditioning/filtering. If I had then my term of employment would have been short.

Stand your ground and make that type of recommendations in a commercial/professional grade installment and you will soon be looking in from the outside.
You may want to read some of the linked information. Sure by definition a UPS is and can provide source for a piece of equipment for blackout/brownouts but not always the main reason for someone to design or spec a UPS design. You have tunnel vision when it comes to UPS. UPS should be specified for the system criteria and environment. Most common consumer usage will be for small system support. Yet that doesn't mean that unit will suffice for a commercial grade system usage or environment. If a consumer wishes to use a lesser grade UPS then so be it. But if that users environment is 'dirty' then I for one would be looking at line conditioning and filtering along with quality proper design type UPS for circuit needs. Sure some manufactures do spec line conditioning but what type. One size does not fit all!

Again, buyer be aware!