We're having to reset our office network at least once a day. it locks up and only a power cycle will get things going again.
We've replaced the router and the DSL modem. No joy. Firmware is up-to-date. The neighboring office manager says that they are experiencing similar issues. The building is old and will be bulldozed within a year (60's/70's vintage wiring).
How can we confirm or eliminate the power as the culprit? Would like to take steps to filter the mains power rather than measure it (ie, strip chart recorder) because I'm certain that it's the culprit, and spending a few $$ on cleaning up the power is $$ well spent, IMHO.
The current power strip powering the network devices -- DSL modem, Ethernet network router/switch (no wireless) -- has integral surge protection but nothing more. All suggestions are welcome.
IMO, If you think you have power problems, you need to spend a few bucks on professional advice before you spend bucks on hardware. Call an electrician.
That being said, if all the core equipment, (DSL router, broadband modem, WiFi access point, etc), is on a small UPS, you've eliminated lots of possible power problems. If you do this and you still have problems, you have big problems or they are unrelated to power.
Don't waste your time. Call your electric company and ask them to loan you a test set to monitor and log your power line.
The last place I worked was having excessive failures on the production floor, and erratic computer problems all over the building. I looked into one of the breaker boxes and told them what was wrong. They told me I didn't know what I was talking about, that the wiring was done to code. So, they called in the industrial electricians who wired the building. They walked around removing breaker box covers, scratching their asses, then sticking the covers back on.
Their final report was, 'We don't see nothing wrong.' Our problem was found by the utility company, and proven by using the monitor. The building was 25 years old, and used the then allowed two gauges smaller neutral three phase system. The harmonics caused by lots of switching power supplies cause high currents in the neutral, which causes heating and erratic equipment shutdowns. The neutrals were supplemented with larger neutrals, to make it equivalent to two gauges larger. 99 % of our problems went away.
If you are using the system in an office and the network is critical it would be advisable to use a U.P.S. to make sure everything is ok, even in case of a blackout U could lose any unsaved data. The U.P.S's I have seen here in aus come with an adaptor cable to automate a shutdown and save data in the case of a power failure. There is also a feature in the software supplied (or downloadable) to monitor the power supplied, I have heard of a friend that did this out in the country to prove that his power was dropping well below allowable limits and forced the company to install another substation to keep him going. Good luck with it. Ken
There seems to be 2 kinds of UPSs: those that switches to battery backup if power drops below a minimum voltage; and those that separate the load from the supply ("double conversion" type). This latter type is, basically, a battery charger running an inverter it's *always" in baclup mode (c: .
It seems that the double-conversion type is truly the only kind that can isolate equipment from "bad" power. All else relies on filtering and fast switching from grid-supplied power to inverter power. (We pray in fast relays!)
Am I right in thinking that a double-conversion model is the best for our needs (our problem being dirty power, not power-fail)?
There is a company called Best Power Technology that sells what they call FERUPS which uses a Ferroresonant transformer to regulate the output voltage and remove anything that is not 60HZ from the power. It will still switch to battery when the power fails completely and produces a sine-wave output.
Somewhere close to your facility is using a lot of power intermittently. Whenever they turn on or off a big piece of equipment, they send the surge to every other users nearby. Since your building will be torn down soon, a good UPS may be the best bet.
Mine does that now and then but it's caused by a connected computer that went off line in the middle of doing some networking that didn't complete. the Router never times out on it and hogs the BW to a point where the router may just stop. First indication is, things starting to move slowly through the router with out any apparent activity giving reason for slow up to the point where things may start timing out on you. etc..
So many who recommend a UPS niether know how electricity works nor realize that most all UPSes connect directly to AC mains via a relay. These UPSes maintain 'dirty' electricity during extreme brownouts and blackouts, do not clean electricity despite so many rumors otherwise, and create power so dirty in battery backup mode as to even harm some small electric motors. Many only assume a $100 UPS contain double conversion. Then convert that speculation into fact. You are correct. The only UPS that would 'clean' electricity is the $500+ version.
That $500 UPS may or may not solve the problem. Far easier is to first identify the problem. Simplest, easiest, and a most informative first step involves measuring voltages on AC receptacles with a digital meter. Measure each receptacle for router and for other networked equipment. Three voltages between each of three plug holes should be measured while all items are powered. Yes, there must be an AC voltage even between neutral (largest) hole and safety ground (round) hole contacts. What are those numbers?
You don't define what is connected where. Apparently router and DSL modem share same power. What else is on that power strip AND also on other receptacles of same circuit? If other network equipment is powered from separate circuits, then voltage measurements between those receptacle ground pins and receptacle neutrals should be collected. Easier by using a three wire extension cord. Those numbers will also provide significantly useful information. For example, numbers may expose a voltage problem that even the $500 UPS will not solve. Or it may expose a problem that will only create other problems later.
One of few who offered useful suggests is Michael Terrel. His building was wired with what is called a shared neutral. Shared neutrals are code legal but will create problems if neutral wire only meets code when powering large or many reactive loads (fluorescents, computers, etc) 'Shared neutral' wiring would be obvious by inspecting how those feed wires enter the breaker box and from voltage numbers. Most circuits are not wired 'shared neutral'. But then most circuits don't create your problem.
Nothing is wrong with properly installed 1960/1970 wiring. As long as the circuits are three wire and not compromised by an ignorant human, then that wiring is just as good as 2000 wiring.
However some have wired circuits using push-in or stab lock connections. That is sufficient for electric light circuits and very bad for electronics. By removing each cover plate, all wires should be fully connected to screws on receptacle side. If not, wires would be connected in the back - that is bad. Every wall receptacle from router receptacle daisy chained to breaker box must have wires firmly attached to those side screws. Anyone connecting to or disconnecting from a receptacle not using screws may cause electronics power problems.
Not defined is even the router. Some router power supply 'bricks' have poor line filters. To eliminate what a $500 UPS might also accomplish, we put this line filter into a receptacle electric box with a duplex receptacle, flexible power cord and plug:
the $500 UPS is going to accomplish anything, then these filters will also create a noticeable reliability improvement.
Another useful test is temporarily install a 50 foot three wire extension cord so that router and DSL modem get powered from same receptacle as other computers or from a receptacle mounted on breaker box..
Reasons for power problems or symptoms associated are easily to located. Long before spending big bucks on a UPS, far more sense to first identify the problem. If building wire is creating the problem, well, it may also be the 'canary in a coal mine'. Curing simple symptoms with a $500 UPS may result in more serious problems later as well as be uselessly expensive. No, neither router nor DSL modem need be on a UPS for reasons also electrically simple.
Power strip protector does zero to solve your problem as made obvious by its numeric specs. Even numbers on its packaging demonstrated that it ignores any voltage below 330 volts. Votlages that would create your lockup would be far lower. The protector acts like it does not exist except when the rare much greater than 330 volt transient happens maybe once every seven years. But then your failing router powered from that protector strip also demonstrates same.
Unless the OP owns the copmany he is working for or it's a home computer, it's not his call if "$500", for the sake of discussion, is too much. As a technician, his job is to work with management people to calculate the business costs of a failure, calculate the costs of a couple ways for recover from the failure ithing the timeframes set by management and them let the Boss make the call to spend the money, or not.
Back in the 80s with low end mini-computers we saw these issues a lot. Many 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.
Wired to code is a safety issue. Clean power isn't really addressed by the code except where too small a wire would cause a voltage drop.
And yes the double conversion UPS is best. I treat the others as something to 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.
Every time we've had "strange" computer issues and I've gotten the office to 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.
But yes we do first check for obvious stuff in the breaker box, shared breakers with the offset press, old wiring that's failing, etc...
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.
A lifetime ago, I was hired to run a shiny new datacenter with the biggest new computer made by Digital Equipment Corp (R.I.P.). That was in the winter and it was in a first-class large building in Manhattan. We didn't have a UPS and we knew there were other large computers in the building in the same state. (I wasn't part of teh planning process).
Everything was fine until May 1st. At about 7:15 the system crashed and rebooted with a "power fail" message. OK. Sh*t happens but it proceeded to happen every day at about the same time.
We escalated the problem, step by step until I finally had a meeting with the "account rep" for Con Edison, (the utility company for non-New Yorkers.) Who knew that the power company had sales engineers?
He checked a few things and then declared that we were getting normal commercial-grade power. The deal was that in the summer they switched in capacitors to compensate for the power factor for the air conditioners that people would begin to use. This had been standard practive for decades and the solution was our problem.
We wound up getting a ferroresonant constant voltage transformer so large it had to be moved in with a forklift.
Why did the "other mainframes" not have problems?
Becuase they were IBM systems and all of them had motor-generator units included in the price. they were even better than CVTs and not a customer third-party option like all the other computer vedors.
If the $500 UPS is going to accomplish anything, then these filters
strange, we just replaced a battery in a UPS used to power the PLC so to prevent unstable voltage. It has a 120 watt inverter that runs all the time and the 120 volts does nothing but keep the battery charged. It's programmed to drop out if power does not recover with in 10 mins and makes sure power remains on for at least 2 mins before it turns on output to PLC. It has corrected many problems since we experience service problems in our location of operation commonly, more so lately than ever.
The last time we bought one of these it was only around $85 and it's a din rail mount with a side din rail battery mounted connected via a module cable. The battery isn't inside the unit. If the battery goes bad, it will shut down. the inverter does not use source supply to power its output.