How to clean up mains power?

Why would AC power cause that failure? One - those routers are defective as designed. Two - building AC electric has major problems that require immediate attention. The design of routers, computers, etc should be so robust that building power should never cause such failures.

For example, Michael Terrell's post demonstrates why electronics were failing. Neutral wires were undersized. Fixing the reason for electronics problem rather than curing symptoms with a $500 UPS also eliminated what was potentially a building fire. Electronics design should be so robust that no building AC power should cause problems. Curing the problem rather than curing symptoms with a $500 UPS obviously is a smarter solution.

Scary are those who only want to cure symptoms.

A story from Datamation magazine. One computer would crash constantly. The tech would run diagnostics all night - no failure. But when the tech went for coffee, then diagnostics would crash every time. Got so bad that the tech would loudly announce he was going for coffee, stomp out of the room, then sneak back to peek around a door and watch the machine. No failures.

Problem was traced to how the computer was grounded. It was grounded to the elevator because a human did not learn his job. Everytime the tech took an elevator down for coffee, that elevator would crash the computer.

There are no ghosts. But there are many humans who cannot bother to first learn even basic grounding; then want to cure problems with magic boxes because they never learned even simple concepts of grounding.

Con Ed capacitors should never cause problems for computers. But some early computers had marginal designs. For example, specs for PCs are quite blunt. Power must be completely lost for 17 milliseconds and computer must maintain output voltages. Computers that put fewer loads on their power supply can operate even longer without power. UPSes that switch over to battery backup may interrupt power for up to

Described were some simple procedures to even identify bad grounding. Of course, if the only failing device is the router, well, is that router design too marginal? So we buy a $500 UPS to fix a $50 defective router?

code except where too small a wire would cause a voltage drop.

use and hope the computer doesn't turn off when the power drops for a bit. Other than that the surge protection they have in them is about the same as the power strip you get at Staples.

go with said "good" UPS setups, the strange has vanished. And usually the money spent was way less than the total diagnostic route. Around here and in many parts of the country getting the power company's help in such issues is an almost hopeless task. They put on their DVM that the linemen carry on their belt and if it reads between 110 and 125 (or whatever the allowed range is) they declare victory and leave.

breakers with the offset press, old wiring that's failing, etc...

We were in a huge industrial park, with a lot of heavy industrial work going on. Even when almost nothing was on in the offices or on the factory floor, the power was dirty. It was cheaper, and better to upgrade the 15 year old neutrals than add a couple thousand UPS to the complex. Also, prior the the upgrade, several critical systems had 25 A

The sad thing was that the newer method was already in use when the complex was built, but the aged EE who designed the system thought harmonics on neutrals were pure BS.

Could be voltage fluctuations either high or low, possibly there may be spikes. Try using a battery backup system like a UPS. Most of those will clean up your power, and if there are mini outages, the UPS if sized correctly will keep power to your equipment.

The fire was later, and caused by the non Y2K 'Merlin' phone system that went up in flames two weeks before it was to be replaced. The power supply died, and stared a major fire. If the fire department had been a few minutes later, the entire office wing would have been lost, along with the design department. The place still smelled of burnt plastic, two years later. To think that a small system on a 20 A 240 VAC dedicated circuit could do so much damage.

Wiring to code is more than just safety, it also prevents ground loops and improper grounding and neutrals, which can lead to all kinds of sporatic problems.

Yes. As for the 500, would it be cheaper than losing all your data.

So far I know (SFIK -- hey! A new acronym!) this: The computers in this office utilize a medical records system whereby one of the computers hosts a server application (which contains the data base), and

Several times a day the client computers suddenly can no longer log into the data base (their first indication that something is wrong) and also cannot access any web sites.

As yet I have not been there when this happens, so cannot confirm ping success or failure. They've learned to turn off the switch to the power strip that supplies the routers and modem, and after a few seconds, all is well with the world again. It's a bit much to ask them to unplug a particular router's "brick" from the power strip, plug in again, and troubleshoot from there. I'll have to spend a day there, sometime soon, to experience it myself.

The next step, I think, is to ask the power utility (PG&E) if they will lend a power quality recorder.

Thanks, Sparky

In this particular situation, the building will be bulldozed in less than a year. Building management is not about to have a major repair done on the electric supply for the building in this time period.

So if, upon closer examination, it is determined that the problem is caused by loads elsewhere in the building generating spikes, confounded by inadequate wiring and/or imbalanced phases and overloaded neutrals, a $500 fix may indeed be our only choice.

Agree with the poster that pointed out that components such as routers should be able to withstand larger power quality issues than they are currently designed to withstand, if profit weren't the primary goal of the manufacturer.

Sparky

Please provide name & model of this stellar, inexpensive UPS. After determining the cause of our power problems it might come in handy.

Sparky

"Jamie" implied that it is a "dual-conversion" type:

"It has a 120 watt inverter that runs all the time and the 120 volts [input] does nothing but keep the battery charged."

Sparky

In comp.dcom.lans.ethernet w_tom wrote in part:

You overstate your case. Both "Standby" and "power conditioning" UPSes are readily abailable in the smaller sizes:

-- Robert

But isn't that the nature of USENET? (c:

Great resource. Thanks!

But it is! A proper neutral doesn't even have the row of little holes, much less the reeds.

BZZZZZT!!! Thanks for trying. NEXT!!!!!!!!!!!

times we'd find things like coffee makers on timers on the same circuit breaker. Or a welder next door but sharing the same transformer. Other times if we could establish the WHEN we'd have the folks spread out for a few days to see what might be going on. The most interesting one was a trash compactor across the street that was run at about 3:45 PM each day which is when the systems would act up. Turned out they were pulling down too much power and affecting the entire block.

Actually they were optional if the customer had a 400Hz source. I worked for a company where the entire datacenter was on a _large_ UPS that provided 400Hz power; none of our IBM mainframes ever had a MG set.

Jerry

But I was suggesting minor repair such as a wire pushed into the back of a receptacle instead of wrapped around the screw, a loose screw inside the breaker box, a miswire that somehow disconnected the safety ground causing significant voltage where voltage should not be, etc. A problem quickly identified with the meter or isolated by temporarily powering router/modem from the 50 foot extension cord. Obviously no expensive solution was proposed. But spending $500 for fix what may even be a defective $50 router design AND not even knowing that AC electric is a problem - that is a major and expensive repair.

One does not throw money at problems like it were a grenade. One first identified a simple problem. Especially when one does not even have to be there when router locks up to see why. Posted elsewhere is another better idea.

Write a simple script that is also an icon on the desktop. When router locks, then each user clicks on that icon to collect and store relevant information in a temporary file.

Some of the tasks that script must perform include ping 192.168.1.1 >>tmpfile ping 192.168.1.100 >>tmpfile ipconfig /all >>tmpfile

When you arrive, view what computers talk to router, what computers talk to each other, and even what computers still have IP leases. There is still no reason to believe power created this problem. If so, then why is the router's power supply so deficient as to not eliminate things that other power supplies (ie computers, modem, etc) easily make irrelevant?

Like the temporary 50 foot power cord, meter measurement (numbers then posted here so that other information you did not see in those numbers can be identified), $10 line filter, etc. So many simple things to first learn where a problem is rather than just automatically assume (for no reason) that it must be 'dirty' power and that a $500 UPS must cure it.

You don't even have to be there when router locks up. Voltages from the meter could report defects even when router is working OK. Again, we first collect data before automatically speculating blame. If power is the problem, then why is only one appliance failing? A question that PG&E will ask, be discouraged by an answer based only in speculation, and then tend to be less cooperative. To get useful assistance from PG&E, wild speculation (it must be a power problem) with is not sufficient. Why do you suspect power? No post here suggests that power is reason for the failure.

How do you know the problem is not expired IP leases? Again, not posted to be a solution. Posted as but another example of the so many existing reasons ignored because speculation blames AC electricity.

So where is the double conversion UPS selling for $85? The app note is simply a summary of names for each type of UPS. It says nothing about a double conversion UPS costing only $85. And it forgets to mention other paths of noise and interference that even completely bypass the UPS. And it does not discuss functions that all electronics contain so that every very dirty electricity from $85 UPSes is completely irrelevant.

What must that router do? Its power supply must be so robust that even an $85 UPS in battery backup mode outputing dirt - 200 volt square waves and a spike of less than 270 volts - is completely irrelevant to the router. Why did an APC app note forget to mention those numbers; forget to mention what electronics power supplies must make irrelevant?

Simpler solutions first identify a problem rather than throwing a $500 UPS to fix what may even be, instead, a defective $50 router. Nothing else crashes and locks? Somehow that also is not relevant? Somehow we know it must be an AC electric problem? A conclusion based in speculation. Somehow an $85 UPS does what a $500 UPS does? Where does the APC app note say that? Why does it not say so much that is relevant to this router problem?