Is UPS putting out simulated sine wave when AVR is active ?

1

1noob

n00b
Joined
May 31, 2015
Messages
5
Hi,

I have a question for you guys.

I own this UPS http://www.njoy.ro/UPS/isis1500l (can select english in top right corner of page) for some time but recently i've moved to another apartment witch doesn't seem to have very stable power, the UPS AVR seems to kick in quite often and stays on for hours at a time, this never happened in my old apartment.

The LCD display shows ~243 incoming V and ~207 output so i am guessing it considers 243V to high and lowers it down to 207.

My question now is, when this is happening, when the AVR is active, does the UPS put out the same simulated sine wave that it does when running on battery ? or does it simply lower the voltage and provide the normal sine wave (pure?) that's coming out of the power socket ?.

Your thoughts on this are welcome, thank you.
 
Wow, whatever it is doing exactly for AVR, you need to disconnect it because the typical line interactive unit will fry itself after that much time on AVR!

Now, the datasheet of the unit states a simulated sine wave, so, sorry, but that's probably an ugly square that is not good for efficiency.

Again, check if the enclosue is hot, sides, top, I bet it's very hot. If so, disconnect your stuff and have someone check the wiring.
 
BTW are there jumpers on the back of the unit for setting AVR thresholds?
 
It's not hot at all, it has a fan that turns on when AVR kicks in, keeps it very cool.

The UPS is considerably hotter under normal use vs when running on AVR with fan turned on.

So does it output the simulated sine wave while the AVR is running ?

michalrz said:
BTW are there jumpers on the back of the unit for setting AVR thresholds?
Click to expand...

Nope.
 
In your case it looks like a 'buck' AVR mode and not 'boost', so I guess the battery might not be outputting at all. Probably some resistor network being cooled by the fan, but that's just a guess.

Edit - was wrong - it might be a transformer with or without the battery.

Problem is, it might still be stressing the components too much whilst doing that constantly.
 
Last edited:
Does it help if i am pulling ~max 450w of its rated 900w capacity, are the AVR components under less stress ?
 
You know I really shouldn't be pretending that hard. I don't know that much about UPS devices.

But the fact remains - for me all the cases like yours have always ended up with the unit dying in a matter of weeks/months.
 
Okay, thanks for the input.

More answers are always welcome so feel free to throw in your 2 cents guys.
 
My guess is - it's a transformer or a set of transformers that regulate the change the voltage. I'm not sure if the battery gets used in 'buck'. If it is, you could tell by measuring the voltage across the battery during load and idle - or - if you dare - an amperometer in series between the UPS PCB and the positive terminal of the battery.

If there's additional conversion like this going on then the transformers are probably wasting a lot of power for heat. I'd bet that if you disconnected the fan or covered it somehow it would get VERY hot during AVR.
 
That would be 2 advanced for me, i don't want to take the UPS apart and lose warranty, i am not good with electronics anyway.
 
It definitely helps you're quite close to the middle of the load spectrum and not nearing it's upper end.

Less wasted heat.
You could cover the fan intake for fun but I don't wasnt to be responsible if somethnig like a fire happens so pretend you didn't read it :D

As an example of the opposite - I had 1000VA units loaded up to just ~160 watts and because of low voltage in our building the batteries in the more sensitive UPS were dead from constant AVR boosts.
Thankfully I only had two of those.
 
keep in mind he is located in Europe ( Romania most likely) where they use 220v not 120v like in the states. The fluctuation is not as high as some might believe. you need a power conditioner to protect your electronics from getting damaged
This is what you need http://www.ebay.com/itm/APC-AV-Black-1-5kVA-H-Type-Power-Conditioner-120V-/151628146229

What you have does nothing but provides you 4 hours of electricity when the power runs out
 
As an eBay Associate, HardForum may earn from qualifying purchases.
I'm located in 230V Poland. It's just as bad.

A year ago, I kid you not, for a full hour on a busy Monday the voltage in the grid in my whole 30 thousand or so town just randomly became 160V.

Everything read 160 volts. the light in the toilet was dimmer, The women couldn't boil their f***k coffee, it was like the end of days.

it was also pretty amazing because only the older cheap units from a local brand Active Power operated properly. 3/3 APCs did well, however every other UPS was unusable.

This proud beast:

zasilacz_awaryjny_230v_activepower_easy600_wisla.jpg


was just like '160, yeah, whatever'.
The fancy LCD ones literally ripped the positive terminal from the battery casing, oozing gel out, some went discotheque, or got red hot.
The hospital had working battery backup thankfully.

Whenever it gets windy, you get a building wide beep once every 20 minutes. The more aggresive AVR units (another local brand Fideltronik is notorious for this) got crazy hot. Batteries die after like 2 years.
 
I think AVR is typically a 5-tap transformer, though there are many ways to design it and technically it doesn't have to use a transformer at all.

If a UPS uses battery in boost mode, it's either below the threshold where its AVR works or it is attempting to lower the amount of drain on mains.

As load impedance drops, current increases, which lowers voltage. Two main reasons this can happen. First, devices in the distribution grid such as transformers have power limits and when you reach those limits, voltage has to drop as current increases. Your load then sees both reduced voltage and also does not receive as much current as it demands. The second reason is that as load impedance decreases, the ratio of load impedance to supply grid (cable) impedance gets worse, making the distribution cabling sink more power.

If the grid can't supply enough power, the battery would be absolutely required to maintain the output. This isn't always the case, but I'm sure there is lots of equipment out there that will use the battery when it doesn't absolutely have to, even if it is within AVR range. Of course, if the built-in AVR can't get the voltage close enough to target, it would also need to use battery. For example if a transformer-based AVR ideally delivers 120V but has taps to handle 120V output at 100V, 110V, 120V, 130V or 140V input, and the input is 90V, AVR alone can't fix that. Some units will also flip to battery if there is instability in the line just because the AVR might not react fast enough to fluctuating voltage.

Transformer-based AVR shouldn't get particularly hot when in use, so if something is getting hot, you've got more circuitry at work than just a transformer.

Sorry for the lack of specific product info. Just figured I'd give a little basic electronics info.
 
Some use the batteries and some don't for AVR.
If it does use the batteries and its not a true sine wave ups then it will be stepped.
Technically AVR should not use the battery and still be called AVR.
 
So, guys, what would you say I could have done better to prepare for such a weird situation I ran into a year ago, where everything read 160. That's right at the low end of most UPSs input range.
Is that why they went haywire?
Or perhaps it was using both the multi-tap transformer AND started to output from the battery thus 'feeding' the city grid?
I'm curious, so if anyone could explain it'd be awesome.
 
Last edited:
michalrz said:
So, guys, what would you say I could have done better to prepare for such a weird situation I ran into a year ago, where everything read 160. That's right at the low end of most UPSs input range.
Is that why they went haywire?
Or perhaps it was using both the multi-tap transformer AND started to output from the battery thus 'feeding' the city grid?
I'm curious, so if anyone could explain it'd be awesome.
Click to expand...

For that amount of difference, you can try a separate AVR/autotransformer between the mains and UPS. As long as the grid can still supply enough power (would need more current to counteract the decreased voltage), it should work fine.

It is not necessarily the case that the grid would be able to supply additional current because drops in voltage tend to occur specifically because it can't supply the current demand. If this is true, the only thing you can do is what you are doing - use a UPS and let it go to battery.
 
dandragonrage said:
For that amount of difference, you can try a separate AVR/autotransformer between the mains and UPS. As long as the grid can still supply enough power (would need more current to counteract the decreased voltage), it should work fine.
Click to expand...

Would you be so kind as to point me to a link featuring such a device? That 160 drop was a one time thing, the most typical fail modes are simple 'blinks' which the UPSs handle just fine. Would I need one for every workstation? (I have one UPS per workstation).
I'm afraid online units and/or generators are out of my league financially...

It is not necessarily the case that the grid would be able to supply additional current because drops in voltage tend to occur specifically because it can't supply the current demand. If this is true, the only thing you can do is what you are doing - use a UPS and let it go to battery.
Click to expand...
Sorry if this sounds ignorant but are you referring to the additional work needed to counteract 'reactance' losses?
 
michalrz said:
Would you be so kind as to point me to a link featuring such a device? That 160 drop was a one time thing, the most typical fail modes are simple 'blinks' which the UPSs handle just fine. Would I need one for every workstation? (I have one UPS per workstation).
I'm afraid online units and/or generators are out of my league financially...
Click to expand...
The least expensive way is more ups that worked last time.
The ideal solution is a step down avr power conditioning transformer setup and for the whole office or a separate computing power network.
What they do in your case would take 440/460/480(whatever it is there) in and output 230v.
However you are are talking some serious cash.
 
Sadly, the one line that worked flawlessly through that ordeal is considered vintage :D

I did have the power people install a second grid and a second set of sockets in each room especially for computers so that coffee time doesn't automatically bring the servers down :D

So, I do have a single point in the distribution board to 'snap' things in solely for PCs...

I have not considered using the other 'phases' though. Not sure code allows it at my job.
It is a starting point though, thanks stormy, thanks guys.
 
michalrz said:
Would you be so kind as to point me to a link featuring such a device? That 160 drop was a one time thing, the most typical fail modes are simple 'blinks' which the UPSs handle just fine. Would I need one for every workstation? (I have one UPS per workstation).
I'm afraid online units and/or generators are out of my league financially...
Click to expand...

This might also be out of your price range, but if you can find something like this used inexpensively, it could be a good addition:

http://www.tripplite.com/line-condi...-power-conditioner-ac-surge-protector~LC1800/

As the product shows, these are often called "line conditioners" and are basically surge protectors + AVR. These things don't have batteries to fall back on, so they are beefier than a low-end consumer UPS. The surge protection will generally be better than a cheap UPS, also. You can plug a UPS into one, though honestly it would be best if you had a decent UPS with a transformer as the cheap consumer transformerless varieties of UPS offer little to no isolation.

But I'd sooner advise you to look for a good UPS used.

Note: I see results for cheap line conditioners from a brand called "Opti-UPS" - never used any of their stuff but I see very bad customer reviews on it, so I'd avoid it.
 
Last edited:
dandragonrage said:
This might also be out of your price range, but if you can find something like this used inexpensively, it could be a good addition:

http://www.tripplite.com/line-condi...-power-conditioner-ac-surge-protector~LC1800/

As the product shows, these are often called "line conditioners" and are basically surge protectors + AVR. These things don't have batteries to fall back on, so they are beefier than a low-end consumer UPS. The surge protection will generally be better than a cheap UPS, also. You can plug a UPS into one, though honestly it would be best if you had a decent UPS with a transformer as the cheap consumer transformerless varieties of UPS offer little to no isolation.

But I'd sooner advise you to look for a good UPS used.

Note: I see results for cheap line conditioners from a brand called "Opti-UPS" - never used any of their stuff but I see very bad customer reviews on it, so I'd avoid it.
Click to expand...

The tripplite is only rated for a drop to 180V for the 230V model as are many others I found.
That's why a step down model is likely needed.
To be honest hiring a electrical engineer to design the system is what is really needed.
 
I gave up on believing there exists a guy in my area who actually knows what he's doing.

When I asked for some additional runs to three rooms, to check and/or put better surge and leakage breakers for the server room runs and have the batteries in the emergency lights changed, I got:
- a bill for what I earn in 2 months
- smirks from the electricians who thought a UPS is overkill
- half of the emergency lighting gas tubes were disconnected because they couldn't find the right batteries

Another time, someone was pulling some cat5 and drilled into a neutral. That was corridor lights power. They just 'excised' that part of the circuit and used the protection (green-yellow) as the neutral :)

I also keep seeing varistor based surge protectors connected at the outputs of UPSs in large organisations :D

I have heard a testimony from a guy who, as a prank, wanted to turn his dorm room neighbour's loud music by sticking a thick steel nail right in the socket.
The nail flashed and off it was enroute to another dimension. Yes. The radio kept playing just fine.

My first IT job as an intern for a 60+ plus government org. Every single day I had to get up onto the roof of the tallest building in town, perform a tightrope walk on a super old cornice with nothing but the lighting protection wire for support.
There was no money in the budget for a fucking wifi bridge, but they still managed to change the carpets every single year :)

Or, imagine this:
- there's the main distribution breaker board inside near the building entrance and a second board for the computer equipment in my office. You can disable the latter completely by flipping the switch on the main one.
- we called an electrician in to fix some random thing
- he disabled the 'computer' board completely at the entrance board - to be safe and proceeded.
- after 5 minutes, as we're poking around with the wiring, my boss walks in all smug and proudly declares 'hey guys I fixed it, the breakers were turned off at the main board'.
That was the day I finally realized in my heart that deep down I really long for a job without human contact.

I do know the UPSs don't provide conditioning. On my absurdly small IT budget (about 2500 dollars a year) I am still in the process of buying varistor-based power strips to replace regular ones.
Best I could find? 600 joules between each line.

I really like the idea of putting a bunch of quality power conditioners in the computer breakers room.
I also see the merit of stormy's step-down idea, but while I see the merit, there's no way I'd find the right people to implement this.

I think I'll slowly start moving towards on-line units as a compromise.
 
you have to really look for buck/boost on UPS's as many only stay with a 20V range while others go to 40 and a couple even 60v.. They use the same transformer for everything, just activate or deactivate the windings as needed or when on battery power etc.. Normally the power goes straight through and the transformer is not used. But during AVR mode, battery mode is enabled but I noticed that unless the voltage drops really low, the battery is not used since the battery charging light is also on. But the battery comes on until the volatge goes back up. I had to look a lot to find a UPS that worked upto 160v.. I only found a cheap powercom one, all the others only worked upto 190v or so.. APC's were over 200V and even the expensive ones did not seem to have multi boost and buck modes, ie 20V steps in 2 stages.. They were all single stage +/- 40V.. Since it is still line power I think the waves would be the same as the power line, with just extra windings added or deleted to adjust the voltage. Line conditioners which are similar cost as much as the UPS.
 
Should have mentioned - I was talking about a drop TO 160V from 230. However I can hear mechanical relays clicking on sensitive UPS' when someone sends a print job to a laser workhorse, etc.

Hearing relays would suggest battery mode, right?

Are there any 'big' line conditioners that I could install in one place, like the breakers room and have them support a few PCs each?

Last one - would you say on-line UPSs are a good in-between for this?
 
michalrz said:
Should have mentioned - I was talking about a drop TO 160V from 230. However I can hear mechanical relays clicking on sensitive UPS' when someone sends a print job to a laser workhorse, etc.

Hearing relays would suggest battery mode, right?

Are there any 'big' line conditioners that I could install in one place, like the breakers room and have them support a few PCs each?

Last one - would you say on-line UPSs are a good in-between for this?
Click to expand...

I do think online UPS are very good and I always recommend it unless people can't stand any fan noise.

A relay could also be used to switch taps on a transformer for AVR (though there are other ways to accomplish this than relays as well) but most likely for a cheaper UPS it is indeed just a switch to battery. Transformerless UPS would have no reason to switch taps on a transformer that didn't exist.
 
Sorry for the confusion, regarding a super sensitive model I meant another, it's a local brand. It's well built, stable, has a transformer and big lcd. And it wastes the batteries after like 4 years so maybe it's doing both.
Only the government and aliens know
 
You must log in or register to reply here.
Back
Top