60 HZ fault variation

pcguy2u · February 2009

Has anyone ever run into a fault on a system where the inverter thinks it saw an out of range frequency?

My UPS's for my wife's and my PC, my Fluke 115 and the Xantrex 6048XW all find the same thing usually at about 8:05AM on some mornings - but not all and not always on the same days of the week and not always at that time. But frequently on Mon, Tues. and Wed. at 8:05AM, but not always.

I have my ideas and I will share them, but I don't want to "lead the witness" yet.;)

Many thanks, Nick

halfcrazy · February 2009

Re: 60 HZ fault variation

Grid tied or off grid? if it is grid tied i have to wonder if there is something the utility co is doing to read something and the equipment sees it as out of freq?

pcguy2u · February 2009

Re: 60 HZ fault variation

Sorry, yes it's grid tied..............

The utility (PG&E in Calif) swears up and down that there is no possible way for the frequency to vary by more than a couple of HZ without the grid being turned of and a lot of alarms going off in a lot of places.

I believe this has something to do with something local?????

Thanks,

Nick

BB. · February 2009

Re: 60 HZ fault variation

How far off are you getting are you reading?

Another possibility is that you are getting a distorted wave that is so far from sine that the frequency counter is counting extra/missing cycles. (somebody clipping peaks when starting the shop air compressor, somebody switching huge currents distorting wave forms, etc.).

-Bill

pcguy2u · February 2009

Re: 60 HZ fault variation

Hey Bill, you're the winner. The high reading from my fluke is 67.5Hz and is the same everytime I have it poised to capture the incident.

What I suspect is that the only other service on the transformer is causing the problem. It's a pressure pump for an irrigation system. I am trying to find out what times the battery operated timers for the drip systems (there are 10 or more) are set for. They should all be off because no irrigation is needed at this time of year, but, if the guy who maintains it has forgotten to turn off just one and it's set for 8:00AM, it would be likely that around 8:06 or so, the pump would kick in????

I'll post the results of the process.

Many thanks again,
Nick

mike95490 · February 2009

Re: 60 HZ fault variation

Grid CANT vary Fq more than 0.25 Hz, before all the other spinning generators drag it back. Somewhere - I saw a chart of the nationwide Hz sag, at the time of the Great NE Blackout. As soon as Fq starts to shift, they shed load, to keep the grid sync'd.

pcguy2u · February 2009

Re: 60 HZ fault variation

Tested my hypothesis this morning by running that pump with my Fluke ready to capture HZ variations - nothing??????????

Also though, the HZ at my home changed briefly to 75HZ at approximately 8AM this morning????

Don't have anything on a timer that's set for 8AM and the only heavy duty motor items I have are a compressor which is always turned off and the refer which only comes on at will???

Still searching,
Nick

BB. · February 2009

Re: 60 HZ fault variation

I will tell you that trying to find/capture transiet events in a power line is really difficult.

Do you have any power line networking/signaling equipment? Bad florescent fixture? Arc Welding?

Open up your breaker panel(s) and look for signs of overheating/corrosion. (browned insulation, dark stain on nearby vertical surfaces (hot objects cause air currents which may leave stains on the surfaces). Water leaks, condensation on electrical connections (including the neutral connection blocks) can all cause problems.

Turn on a portable AM radio to a "quiet spot on dial" (off station--FM will not work--so no Ipods) and listen for interference. Walk the radio around to various equipment and see if the sound on the radio from the "event" can be found when you walk near any other devices...

If you can get a digital oscilloscope, sometimes you can set a trigger and capture a strange wave form... Be really careful making the connections... Most scopes have a ground clip--that is connected to ground and if you attach them to AC--Instant destruction--only do this if you know what you are doing.

I am a little confused--are you the only person connected to your local utility distribution transformer? Or are there others too?

My guess is that your interference would probably be limited to those sharing the distribution transformer with you.

But, if you have a nearby friend not sharing your transformer--can you use your Fluke to monitor his power and see if you find the same issue or not...

-Bill

Ecnerwal · February 2009

Re: 60 HZ fault variation

Certainly sounds like interference from something ugly, relatively local.

Welders, PWM speed controls, things which dump garbage back on the line. Need not be on the same transformer as you.

Ideally you want something like a digital storage scope with a huge storage, though having a known/usual time window helps.

Really, the utility should be all over this - they have the overpriced tools to record it accurately, and they should have an interest in tracking it down and making the originator of the noise either disconnect, or filter the offending device. Drop a note to your Public Utilities Commission if they don't get helpful.

Cariboocoot · February 2009

Re: 60 HZ fault variation

How friendly are you with the neighbors? In order to hunt this problem down you have to determine if it's all over the neighborhood or just at your house.
The fact that it appears only at certain times points to something being turned on which is causing a situation which 'looks like' high frequency. The usual suspects are, as others have mentioned, induction motors. However, transformers can also do this. If there's a motor driven off a step-up transformer somewhere that would certainly be suspect.
Any pre-existing condition such as a faulty neutral on a line transformer would cause the frequency to be off always or at random times. A 240 volt motor on a system with faulty neutral would do it on start-up, but there would probably be variations as long as the motor ran.

You have to start taking things off-line and checking frequency as each item is added on. Not easy if the problem is outside your house system.

If nothing else, it's a very interesting puzzle (and we all want to know the answer!):D

niel · February 2009

Re: 60 HZ fault variation

cariboocoot is onto something, but he didn't follow through with it. ask your neighbor if you could be at his place to see if 1> that same thing happens there and 2> if you see something there that may account for it. you may try the radio idea, but the quick burst may not register with your ears that would otherwise show up on more time sensitive equipment such as an oscilloscope. 8am may also be the time that the utility kicks in another generator to help handle higher daytime loads and it just cannot sync fast enough to suit your stuff.

pcguy2u · February 2009

Re: 60 HZ fault variation

Ok, first, the Fluke 115 is time sensitve and when set to capture based on HI/LO/AVG, it will maintain a record of those as long as it is turned on, and it can be set to not turn itself off. So capturing a record the event is relatively simple - as everyone has pointed out, it's the cause that is hard to find.

I should have been more clear on my last post - yesterday (Friday) I tested the neighbor's pump (the only other service on this transformer and the only device on the service - no house) with negative results at about 9:30AM and the 75.00HZ anomaly that I captured occurred at 8:05. Could it still be the pump and that having run once an hour or so earlier, could the problem have been non-existent at the later time????

PG&E (the utility company) has set up a monitoring of the grid by inserting a device between my panel and the meter. The bad news is that they called back after installing it and said that the equipment only monitors voltage. I haven't been testing the voltage on my end, but if the monitor saw a spike of some sort at the same time the 75.00HZ occurred on Friday morning, I may have a case with them. The engineer that called was of a mindset that the problem can not be originating on the grid - he really didn't address the idea that the grid begins at the transformer and even went so far as to suggest I should have my meter checked, even though I had told him that the UPS's, the Fluke and the inverter all saw the same anomalies at the same time?

Then he tried to blame the inverter, except, on Friday the invert was off-line and not doing anything with the grid. This has also been true at other times in the past.

More to follow,
Nick

pcguy2u · February 2009

Re: 60 HZ fault variation

And this morning at 7:45AM while waiting for that magic hour, the Fluke captured something I haven't seen before - 33.75HZ for a minimum. I do recall now that I have received both hi and lo freq faults from the inverter, but have never recorded the event?????

Nick

BB. · February 2009

Re: 60 HZ fault variation

Completely out of left field... Try connecting one end of your DVM to your ground rod of your home (or breaker panel) and the other end as far away as you can to a metal fence post or something metal driven into the ground (long nail, etc.). Possibly in the direction of your utility pole or transformer (maybe 10-100 feet away).

I am wondering if you have some sort of grounding problem in the area (hot wire energizing your earth). I have seen 60+ volts (and have gotten shocked) from two grounds ~100' apart by a saltwater pumping system (old Marine World in Redwood City, CA)

Even if you do--it may not be causing your problem... But something strange is going on.

I agree that you are most likely not seeing a real frequency shift... Changes that large would cause the power grid to fault--or damage to the electrical generators/grid in the area.

You are most likely measuring the effects of some sort of distorted wave form caused by somebody local to you.

Are you seeing any flickering lights, TV issues, radio noise (most likely AM or short wave)? Digital alarm clocks losing settings/time (those without battery backup)?

Has a desktop computer, DVD Player, etc. (not on UPS's) ever lost their "marbles"--especially during the time of day your inverter/ups's are faulting?

Any visible/audio effects at all (other than the ups/gt inverter)?

Any chance/results of plugging your meter into a neighbor's home?

Any "heavy" industry/loads in your immediate area? Use Google Earth/Map to look for possible sites?

Ask your neighbors if they have any problems with electronics/computers at the same time?

You may end up needing to setup an O'scope of some sort to see/capture these events to find anything useful... And it may take you a few days of fiddling around to get a useful trigger (low freq reject, peak voltage just outside normal range, etc.). The standard o'scopes don't have freqency based triggers and such... You might be able to use your UPS alarm as a post trigger (digital capture and stop on trigger)...

-Bill

pcguy2u · February 2009

Re: 60 HZ fault variation

Hey Bill, that's a mouthful - where does one start? First, this is a remote hillside of ag property in rural Sonoma County at elevation 400' with the nearest house more than 1/2 mile down the road and on another UG transformer. Except for my nearest neighbor, the one with the irrigation pressure pump and no house here, there is noone nearby.

I do notice that since I installed the demand water heater (Takagi JR.) all the AM stations recieve a pulse of static, all the time - if I unplug it, the pulse disappears. No timer there and I doubt any association. No other visible signs of the problem. I would have never persued this except that the inverter fault provided some detail and has been ongoing for some time now.

The timing of this symptom certainly indicates a timed device, but what. There's is nothing nearby. I suppose it could be a hot tub - the nearest neighbor has one? I'll email him - he live in Napa and isn't here frequently.

If the PG&E monitoring device doesn't pick up an anomaly from yeterday morning or this morning, I'll have to assume the problem is in the wiring in the house - I have no idea what it might be.

While writing this, I noticed that the low reading on the Fluke has now dipped to 33.73 from 33.75 earlier, an indication of an ongoing incident.

Nick

BB. · February 2009

Re: 60 HZ fault variation

How long do the incidents last? I cannot imagine that they last for more than a few cycles without you noticing flickering of an old (magnetic) ballast florescent lamps or filament bulbs (modern CFL's are high frequency anyway--so who knows how they ride through the event).

Anyone running a linear particle accelerator "up the road" from you? :roll:

I got noth'n else...

-Bill

PS: It is either something local to your home, or relatively large on your distribution network--I doubt a home hot tub would do it... Even then, you would probably see a voltage droop instead.

If you turn off the GT inverter--do you still get the events? -- just guessing...

Cariboocoot · February 2009

Re: 60 HZ fault variation

Let's see if we can help solve this thing before we all go crazy!

pcguy2u - is your house on it's own, separate transformer? Or does that transformer feed more than one house? If you're on your own transformer the problem isn't likely to be in the grid (never say never). Second question: this frequency variation is being measured across the poles (i.e. 240 volts) or reference to neutral/ground?

I've seen several instances of failed neutral line causing an apparent increase in frequency on one pole when a 240 motor kicks on or shuts off - the residual magnetic field 'back feeds' one side and the waveform multiplies. This problem nearly always exists with a step-up transformer (the reason for forced phase synchronization).

The other 'likely' I can think of is something inducing a field on top of the 60Hz. Any two relatively strong electromagnetic fields close to each other can do that. Seen it with two heavy motors sitting next to each other.

pcguy2u · February 2009

Re: 60 HZ fault variation

The duration of the incidents are alway very short and never produce anything visual that I have seen. The reading is taken with a ground reference on one side of the 240, but the inverter is looking at the whole picture, not just one side.

No other homes on the transformer (all new home/transformer/12KV 2.5 years ago). The other service on the transformer only has the pressure pump on it which, when I tested it (see earlier post) failed to show any involvement.

I'll start trapping voltage variations instead of HZ and see if that shows anything.

And, I'll report the results of PG&E's monitor test next week.

niel · February 2009

Re: 60 HZ fault variation

the neighbor's pump at 9:30, or any hot tubs i would imagine, have nothing to do with the problem you are experiencing at about 8 and i was only asking if the neighbor's electric service might show the same anomaly. your neighbor just doesn't have any equipment sensitive to such an event as you are seeing at your place. if the fluke can detect it over your neighbor's place around those times that you experience it, then i would suspect a grid originated problem. the frequency quark may or may not show a voltage variation from the pole and is like checking your oil for antifreeze levels. i also have my doubts that the problem originates with any neutral or grounding problems, although i can't say for sure. if you were to detect any frequency variations at the specified times at your neighbor's place that this may rule out actual problems originating at your place.
don't forget that the long stretches of wire from the poles can act as an antenna too and pick up stray stuff, but i've got my doubts that would ever be strong enough to disrupt your stuff. this is reaching for straws, but are you or your neighbor into radio transmitting such as cb or ham by chance? that would show more of a randomness to the problem's time occurance i would think anyway.
in finding the problem, you may have to start to find it by ruling things out. if this turns out to be the utulity kicking in another generator at a specific time that they will be reluctant to say what the problem realy is or do anything to rectify it.
nothing difficult is ever easy.

pcguy2u · February 2009

Re: 60 HZ fault variation

Here is some more info - there are no other subscribers on the same transformer. Every other service is at least one transformer plus 1/2 mile from here on the 12KV buss.

There is a water treatment plant about 3/4 mile away and they have a computer system that appears to have some degree of trouble staying online and many of their anomalies occur at 4AM and 8AM as well as other times. When I asked the plant owner if he was suggesting that his equipment was potentially at fault, he said, "on the contrary, our equipment also seems to be subject to the problem".

After discussing the other equipment he has at the plant (other than the computer), it came out that he has a variable frequency pump that may be acting up??????????? It's 5HP. I think the conversation contracted quite a bit after the question about cause or effect.

Comments and thoughts?

Nick

mike95490 · February 2009

Re: 60 HZ fault variation

AFIK, variable speed motors, have a VERY expensive control circuit, and would be unlikely to cause noise on the AC line.

That said, 5HP is enough power, it could mess things up.

BB. · February 2009

Re: 60 HZ fault variation

Actually, a variable speed motor can very much be the source of a fault...

Actually, my bet would be the controller. Depends on the actual design and construction of the motor--but many variable speed motors are just poly-phase AC motors (saves on brush/commutator/slip ring maintenance).

What the controller can do--in a failure condition, is overlap the turning on of FETs (transistor switches) where both on the same leg, can be on at the same time (as one example of a failure that leaves the motor running, but still causes electrical interference).

The problem can be very difficult to diagnose.

I had a similar problem on a much smaller scale... A 10 watt DC fan that used an "H" bridge type DC to three phase A\C circuit on a piece of telephony network adapter gear...

First day on the job, I was given a pile of "dog" systems that nobody could fix... And I had never seen the gear before.

And one of them kept randomly resetting/dropping data and nobody could figure out why. I put a scope on the board and saw what appeared to be some real nasty high frequency noise spikes--but I was not sure--first time working on their equipment and a using their O'scope. Eventually, I made a little antenna out of a loop of wire and moved it around the equipment until I saw a similar noise spike--was right behind one of the DC cooling fans.

The two FETs turned on at the same time for only a few 10's of nano-seconds (10 x 10^-9 seconds)--but the 12 volt + to - short through the two FET's was more than enough to disrupt the rest of the internal digital components.

Anyway--back to your issue... Many times, sewage plants will experience peak flows first thing in the morning (people getting up) and/or high flows during rainy seasons (if they have points where rain/ground water can enter the sanitary sewers).

In the end--You may have found your source of problem--but proving it to them (and others) may be difficult without good test equipment and access to the plant. And somebody that knows how to use it.

They may be open to you putting your frequency logger in their wall outlet and see if you find the same (or perhaps more often) occurrences.

-Bill

Ecnerwal · February 2009

Re: 60 HZ fault variation

Even controllers that are not "faulted" can put out some nasty garbage, because they are frequently operating by by drawing no power for a part of a cycle, then switching on for 3/4 - 1/2 - 1/4 of a cycle. In the absence of filters on the line feeding them, the sudden transition from no current to many amps can send all sorts of noise up the line.

The first page of this PDF is for parts associated with variable speed AC drives under 20 hp. While the "Line Reactor" is "optional, but recommended", note that one of its features is to "reduce the harmonics associated with AC drives" - basically noise associated with abrupt switching inside the variable speed drive. More detail on page 3. It's basically an inductor, and will tend to squelch high-frequency noise in either direction. There are two additional components that would also help to filter noise, though they would be more likely to help the computer in the sewage plant than your house - the EMI and RFI filter. If the 5 hp variable speed pump was installed "on the cheap" they might have skipped these components, and be dumping noise onto the power line.

http://web5.automationdirect.com/static/specs/gsaclr.pdf

System2 · February 2009

Re: 60 HZ fault variation

If you want to learn about power quality and other problems like this check out Fluke's website and their application notes, they have a lot of information on these type of problems. Speaking from experiance I was called by a friend to fix his irrigation system that stopped working. The pump was run off a drive that was not installed correctly and started to blow up motor starter coils that were connected acoss the line with it. All in all, motors and drives would be the first thing I would suspect if I were you. Try to get a consulting firm that does power quality to come out and hook up their own test gear, they might be more help then the power co.

60 HZ fault variation

Comments

Categories