Battery swap for cyberpower UPS

[chrcoluk]

I just spoke to the tech in the van (he waiting for the drilling work).

I had to explain what a UPS is, after that he confirmed with me on my photo the 254v is out of spec but said he thinks the power cut issue is having no effect as its a different leg. (There was a fob off attempt that if my UPS is tripping its my problem, as politely as possible I told him the voltage is going out of spec and he backed down from it.)

So yeah these guys arent here for my issue, and the tech has the opinion its unrelated, so will have to wait and see if the engineer they assign to me to use voltage reader and see what happens from that, but I am still hopeful when they fix the power cut it will reduce my voltage, as its one big fat coincidence, this happened at the same time.

--

Its 255v now. Thinking about shutting things down, I might take your approach Bill and ring the local number they gave me to hint if any of my equipment gets damaged they will be on the hook for it.

They have done, and at end it dropped to 251v, and is now fluctuating between 250-251v for a while. Higher than before and would raise @Panther_Seraphin eyebrows but is at least in spec now.

-- After some sleep, Down to 248V now.

 

[Shrek]

Has the clicking stopped?

 

[Bill_Bright]

If he has been sent there on a different trouble call, then that is his priority. And if he doesn't know what an UPS is, then he clearly is not a technician trained and certified for residential electrics.

255V is clearly too high. You might want to enlist your neighbors. It sure would be good if you had a multimeter or a power meter - actual test equipment, something they would understand - so you could check their outlets to prove and verify it is not just you, but the whole neighborhood. More of your neighbors calling and complaining will get more attention.

The nice thing about those power meters is they just plug into the outlet and the display shows what is happening. No experience or training necessary and no sticking in meter probes.

 

[Shrek]

I have voltmeters running 24/7; there are units for your locality available on amazon.co.uk

 

[Bill_Bright]

Gee whiz. I recommend outlet testers so often so folks can verify their outlets are properly wired and grounded that I am embarrassed I didn't think of mentioning them in my last post. I blame that blatant omission on decaffeination and an acute case of GCF*.

Shrek is right. As seen here, this one for the UK is very affordable and IMO, every computer user should have access to one.


*GCF = geriatric cranial flatulence

 

[chrcoluk]

Its not resolved.

They started refilling the pavement and then the voltage went up, the technician was gone, but the digger guy kindly rang one for me, and he was then asked to quickly undig again, and I rechecked my voltage, and it was back down again. But this was only temporary as they had to put the pavement back to normal, so now the pavement has been restored and I have high voltages again.

My incident is still on the system, I got another SMS today saying they will contact me within the next few days.

Sadly we have a public holiday tomorrow so its happened with quite bad timing. Tuesday is the next working day, and its a 2 working day response.

I turned my NAS machine off yesterday. Its basically moving rapidly between 252v and 259v currently. Probably wont keep this machine powered up for long either.

Have read the above replies, will see if I can get something next day on Amazon.

Ordered the voltmeter. Delivery tomorrow.

Has the clicking stopped?

Yeah thats not came back on since I said it stopped I think,

Sadly I dont currently have plot data for the UPS history. But its not been clicking today whilst I have been awake and sitting in front of it.

 

[Bill_Bright]

They started refilling the pavement and then the voltage went up

undig again...and it was back down again

now the pavement has been restored and I have high voltages again.

That really makes no sense. First, they should not be digging anywhere near underground power cables or distribution node points - at least not without the necessary and required digging permits. Those permits are not just to have the proper permissions, but to allow time for the essential surveys to be done just to ensure any digging does not disturb or damage other utilities (gas, water, sewer, communications, power, etc.) that might be buried there.

Even with the proper permits, the cause and effect events you described should not happen UNLESS (1) they damaged something during their initial dig, or (2), there already was some damage or fault and moving Earth around is causing some existing faulty connection or connections to move and alter the electrical continuity through that bad connection or connections. Point being, the connection(s) needs to be repaired properly, and quickly - if for no other reason than to keep moisture, dirt, and other contaminates out of the connection.

I mean, you are in the UK. The UK is not exactly known for its dessert, arid climate!

Since this appears to be a new problem, I suspect they caused the damage.

Bank holidays should not matter. Again, I would enlist your neighbors and have the report the problem independently from your report (so it is a bunch of reports and not the same "1" report). It may not, probably won't get any faster response, but it should get more attention up their chain of command.

BTW, I bought a set of these neat little voltmeters - so I wouldn't tie up my voltage testers or multimeter. I keep one in the sister socket where my coffee pot plugs in. So every morning, I can quickly glance over and check the voltage in my house. It is sitting at 120VAC right now.

Note the description on Amazon is wrong. It is not battery powered. It gets its power from the outlet. And those of course, are for North American outlets but I am sure there are similar models for most other types used in different parts of the world.

 

[Shrek]

BTW, I bought a set of these neat little voltmeters - so I wouldn't tie up my voltage testers or multimeter. I keep one in the sister socket where my coffee pot plugs in. So every morning, I can quickly glance over and check the voltage in my house. It is sitting at 120VAC right now.

I had that model, but threw it away as it did not seem safe inside.

Went with these instead
Amazon.com: Prime Products 12-4059 Digital Ac Voltage Line Meter : Tools & Home Improvement
Much higher quality.

Time to open a new thread? as we are way off "Battery swap for cyberpower UPS"

 

[Bill_Bright]

I had that model, but threw it away as it did not seem safe inside.

You sure it was the exact same model?

Being so inexpensive (just $6.50 each), I admit, that was a concern for me too because my plan was to keep them plugged in 24/7/365. So as soon as I received them, I opened one up and was very pleasantly surprised to see nice, shiny soldering joints with no globs of solder or cold solder joints. The surface mounted components were neatly laid out and soldered-in, nicely snug to the PCB. The insulation on the wires was properly cropped (no exposed copper) with no signs of birdcaging or excessive heat. Even the case was solid, securely closed with 4 screws instead of being snapped together in a cheap, flimsy manner.

Again, pleasantly surprised at the quality.

Take a look at the Amazon page again and note the last image shows the interior. The fact they even show an image of the interior suggests to me they are proud of the construction and assembly (and rightfully so, IMO).

If you (and knowing a bit about your background) did not feel comfortable with what you saw when opened it up, I have to believe it was not the same product as the ones I have.

But, I still not 100% sure (again, because my intent was to keep them plugged in 24/7). So I checked the device several (many!) times over the next few days to see (feel) if it was heating up as that might suggest it was drawing too much current for my comfort. But I never felt any warmth at all! In fact, I just now checked it and the surrounding wall with my trusty laser guided IR thermometer gun. Right now, the outlet faceplate under the meter reads 71.1°F, the same as the surrounding wall. The underside of the meter reads 71.7°F. And the topside of the meter reads 72.6°F.

Considering it is a small closed case and the display is a constantly lit, backlit LCD display screen (black numbers with the entire background lit in blue) and it is powered on 24/7, IMO, a temperature difference of a mere 1.5° is just fine - great, even.

And considering your model (which I am sure works very well) costs $18 each, nearly 6 times more than one of these ($13 for two), and $4 more than a much more informative and useful GFCI Outlet Tester, for the purpose I bought mine, I'll stick with those I bought. And for sure, I have no reservations recommending them to any one, or buying them again should the need arise.

 

[Shrek]

It is exactly the interior that freaked me out... those mains pins sit close to the board and could easily be pushed in.

 

[Bill_Bright]

Nah!

But, you just gave me concern so I decided to see for myself with one physically in hand.

I opened the second one just now to see how close those pins come to other components, and how easy it might be for the mains pins to break loose and get pushed in. No way are they going to break out of the housing and become loose. I pushed hard trying to break them loose. Didn't budge. It would take some serious blows with a heavy hammer for them to come loose. Nothing even close to those forces would ever happen inserting this device into an outlet.

Also, you cannot tell from that image because it is looking straight down. But looking from the side, it is easy to see the back side of the housing is considerably deeper than the front. Those mains pins are recessed in that back side of the case housing, barely protruding just half its depth. And the PCB is recessed deep in the other side. Plus there are no raised components on the PCB opposite the pins.

They couldn't get further apart. Nothing comes close to anything else. Nothing about the design or construction of this device gives me, as a certified, master electronics technician for 50+ years, pause in any way - as long as the device is used for its intended purpose, and has not been hit repeatedly by a hammer!

But hey! If you want to spend 6 times as much, that's up to you. Like I said, I am sure yours works just fine too.

Now we need to get back on topic.

 

[chrcoluk]

I have tried the meter now, not yet in the PC socket.

So basically when the voltage moves up and down, that happens in the socket I tested as well (in the kitchen), however it is consistently about 2V below what my UPS reads so e.g. if UPS says 255V then meter says 253V.

The PC socket is an absolute pain to get to, never mind to see what the meter would be displaying, I am probably going to move my desks before I put the meter in there, so the access is easier in future.

 

[Bill_Bright]

Well, you didn't specify which meter you got so we don't know its quality. But, unless you got truly a professional model most decent meters have an AC voltage accuracy of ±1.5% or better. So with a 250VAC, the reading could easily show 255 if it reads high. And the same goes for the UPS display. It might read 245V with the same input.

Even high-end meters can be off by 0.5%. You would have to compare your meter to known "standard" - a precise, calibrated voltage. Some of the better meters let you calibrate the meter to that standard. But most do not as that level of accuracy is not normally needed out "in the field".

I would check the outlet used by the UPS. It is common for voltages throughout the house to be slightly different due slightly different resistances in the circuits due to different wire lengths, and resistances through junctions, breakers, switches, etc.

 

[chrcoluk]

Well, you didn't specify which meter you got so we don't know its quality. But, unless you got truly a professional model most decent meters have an AC voltage accuracy of ±1.5% or better. So with a 250VAC, the reading could easily show 255 if it reads high. And the same goes for the UPS display. It might read 245V with the same input.

Even high-end meters can be off by 0.5%. You would have to compare your meter to known "standard" - a precise, calibrated voltage. Some of the better meters let you calibrate the meter to that standard. But most do not as that level of accuracy is not normally needed out "in the field".

I would check the outlet used by the UPS. It is common for voltages throughout the house to be slightly different due slightly different resistances in the circuits due to different wire lengths, and resistances through junctions, breakers, switches, etc.

This I expect is why they sending out a meter reader. So they can verify with accurate equipment.

I have at least confirmed the fluctuations are not limited to the PC socket.

The one I ordered is the one you linked to.

 

[Wirko]

This I expect is why they sending out a meter reader. So they can verify with accurate equipment.

I have at least confirmed the fluctuations are not limited to the PC socket.

The one I ordered is the one you linked to.

Meter reader - is that a person or an instrument?

The utility company should connect a power quality analyser at your house, which would record voltage fluctuations over an extended period (several days, I guess). That's what they do here when there are complaints. Your mains voltage must be within limits at all times, not just when you're looking at the voltmeter. Sure, that doesn't mean 100% of the time, or even at "six nines", but at some retail quality-of-service level that is defined by ISO and other standards, and national regulators. See an example here (but read "tabs = 2 types" as "flicker = 2 hours").

Two examples of such analysers: this and this, both made by Slovenian companies {[not related to EKWB]}. We have quite a strong industry of these instruments here. The poor man's choice could be a datalogging multimeter like this one by UNI-T, those can capture slow changes in voltage but not the other types of bad power events.

 

[chrcoluk]

Meter reader - is that a person or an instrument?

The utility company should connect a power quality analyser at your house, which would record voltage fluctuations over an extended period (several days, I guess). That's what they do here when there are complaints. Your mains voltage must be within limits at all times, not just when you're looking at the voltmeter. Sure, that doesn't mean 100% of the time, or even at "six nines", but at some retail quality-of-service level that is defined by ISO and other standards, and national regulators. See an example here (but read "tabs = 2 types" as "flicker = 2 hours").

Two examples of such analysers: this and this, both made by Slovenian companies {[not related to EKWB]}. We have quite a strong industry of these instruments here. The poor man's choice could be a datalogging multimeter like this one by UNI-T, those can capture slow changes in voltage but not the other types of bad power events.

I dont know, voltage reader was the term used. I assumed at the time they meant a person. But I have since then considered it might be an instrument.

National grid are on the street again across the road (they have dug a much smaller hole other side of road).

(I see I said meter reader on post you replied to, thats a typo, they said voltage reader).

 

[Panther_Seraphin]

I dont know, voltage reader was the term used. I assumed at the time they meant a person. But I have since then considered it might be an instrument.

National grid are on the street again across the road (they have dug a much smaller hole other side of road).

(I see I said meter reader on post you replied to, thats a typo, they said voltage reader).

Knowing Grid it will be an engineer with the ability to take specific readings from multiple points most likely. I suspect he will test from the street back to the local substation looking for leakage/a fault.

 

[chrcoluk]

Knowing Grid it will be an engineer with the ability to take specific readings from multiple points most likely. I suspect he will test from the street back to the local substation looking for leakage/a fault.

I think that is actually whats happening, there is currently 10 holes, they seem to be digging a new one every so often.

Meanwhile locally, I have finally tested more sockets including the PC socket.

So this is what I can summarise.

The power to my entire home is moving up and down, when previously prior to the neighbouring power cut it appeared to be stable, the range seems to be around 5-11V between min and max.
In addition 'occasionally' the socket for my PC, has around 2V above the other measurements. But not all the time, sometimes its measurements match.
I have finally I believe diagnosed this occasional extra 2V to be a smart plug I had put there, I put it in as I wanted to take direct watt measurements at the socket (the UPS seems just reports 0W when power usage goes below about 90W, as well as to validate the UPS figures), the other reason was because the socket is so hard to access I wanted an easy way to kill the power. For now I have removed the smart plug, and since then the UPS measurements are the same as the volt meter.

I havent been contacted by National Grid yet but will give them a call today.

 

[Waldorf]

sine all proper UPS allow for adjusting the voltage range where AVR is used/batt kicks in,
i would just make sure its set to be to sensitive/min range possible.

 

[Bill_Bright]

It should be noted that it is common AND normal for the mains voltage to vary a few volts throughout the day. For example, while mine typically sits most of the time at 120VAC, I often see it drop to 117VAC and climb to 121VAC. This is perfectly normal as the load on the local transformers vary throughout the neighborhoods and across the regional power grid as users (and businesses) wake and go to sleep, the sun rises and sets, high wattage machines and appliances like ACs, refrigerators, furnaces, clothes driers, ovens, etc. turn or cycle on and off.

Where there are several power hungry businesses in the area, or on abnormally hot or cold days, those variances might even be wider and still be normal, and perfectly safe for our electronics.

And don't forget, IT IS NORMAL for different voltages to appear at different outlets throughout the house. Fact: The longer the conductor (wire) the greater the resistance. In fact the spec here says for 120VAC mains, the voltage "at the furthest" outlet should be no less than 114V (a drop of 5%). Note an increase of 5% would be 126V and still within allowed tolerances.

With 240VAC mains the variances would be similar and still perfectly normal and safe for ALL our electronics with many devices being significantly more robust and tolerant.

So do NOT expect your mains voltage to be rock steady at exactly 120VAC or 240VAC or whatever is standard for your part of the world.

What you can and should expect, however, at least in most advanced, industrialized countries, is steady frequencies. That is, 50 or 60Hz should remain at 50Hz or 60Hz at all times. In the US, the allowed tolerance is ±0.5% or from 59.7Hz to 60.3Hz. I personally have never seen it vary more than .1Hz, and typically it is exactly 60Hz.

 

[chrcoluk]

The voltage variance is fairly rapid (many cycles an hour, sometimes movements seconds apart).

It is also still going out of the legal limits.

 

[Shrek]

Variations are normal as someone uses an elevator, etc

Lift, I meant lift... I'm originally from England.


Might be nice to move this 'mains voltage variations' to a new thread as it has nothing to do with 'battery swap'

 

[Wirko]

Might be nice to move this 'mains voltage variations' to a new thread as it has nothing to do with 'battery swap'

Half of the thread is about this already, so it should be renamed to "UPS battery swap, followed by mains voltage variations".

 

[Bill_Bright]

I think the subject - both of them - have played out and there is nothing useful more to add until the power company is done doing their thing.

 

[Wirko]

It is also still going out of the legal limits.

The change from 220V (continental Europe) or 240V (UK) to 230V was an administrative one, without technical changes. It was made possible because regulation [I mean voltage regulation, not laws] had improved over time, and narrower tolerance ranges could be set.

Just guessing now - maybe some parts of the grid were within the allowed range before (240V +x% / -y%) but aren't now (230V +z% / - w%) because that would require new transformers or something else. Also note that the upper limit (in %) is often a different percentage than the lower limit (in %).