EDAboard.com | EDAboard.eu | EDAboard.de | EDAboard.co.uk | RTV forum PL | NewsGroups PL

UPS Upgrade

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics Equipment - UPS Upgrade

Goto page 1, 2  Next

AncientOfyears
Guest

Tue Jan 19, 2016 7:48 am   



I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.

Steve
Guest

Tue Jan 19, 2016 8:21 am   



On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com> put
together some random words that came up with:

Quote:
I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.


Did the same thing using a APC SmartUPS. Used the biggest marine batteries I
could find. Used it to power a phone system and alarm system. The thing
supplied power for around 4 days before the batteries went dead.

Jeff Liebermann
Guest

Tue Jan 19, 2016 8:30 am   



On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com>
wrote:

>I need to fixup a 1500VA USP to run longer.

How long? If more than a few hours, I suggest a Honda inverter
generator.

>This is a pseudo sine wave unit.

On the bottom of my APC BE550G UPS is the following inscription:
The output of this device is not sinusoidal. It has a
total harmonic distortion of 45% and a maximum single
harmonic of 35%.
Assuming APC is talking about voltage distortion, a THD of 45% means
that
0.45^2 = 0.20 = 20%
of the total output power is wasted in harmonics of 60 Hz. If you're
driving a motor, transformer, or have a switcher with a low pass input
filter, or running a device that really hates harmonics, you're going
to have problem. Resistive loads are ok. Switching power supplies
are usually ok. Choose your loads wisely.[1]

>First I plan to derate it by 50% so I will draw no more that ~700VA.

Duz your proposed and unspecified model CyberPower UPS have a fan?
You'll probably need it. UPS's are also rated for some number of
hours of operation or some percentage of duty cycle. I suggest you
check the specifications.

Quote:
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.


They're called "deep cycle". The problem is that the original
batteries are probably AGM (Absorbent Glass Mat) batteries, which are
charged to some specified maximum voltage at a given temperature. If
you replace these with a different chemistry battery, such as
lead-acid flooded cells commonly found in automotive and marine
batteries, the charging characteristics and maximum charge voltage
will be different. I'm too lazy to dig out a table for you, but I
think you'll find that they're different. Also, removing the
batteries from the case also removes them from being thermally
connected to whatever temperature sensor is built into the UPS.
Fortunately, you can extend the leads on those and attach them to the
battery box. That's the sensor on the right:
<http://www.11junk.com/jeffl/pics/repair/APC%20XS-1300/slides/XS-1300-05.html>

To make matters worse, the various UPS manufacturers compete with each
other for the longest runtimes. To achieve high numbers, they charge
the batteries with an EOC (end of charge) voltage to be as close to
maximum as possible. This is not good for the life of the battery,
especially when changing battery chemistry. They also charge as
quickly as possible, which is yet another reason for relatively short
(5-8 year) battery life in a UPS. Fortunately, when going from the
stock battery, to a much larger battery, the lower relative charge
rate will help extend the battery life.

Quote:
So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.


Among consumer UPS's, APC is about the best I've seen. The others,
including CyberPower, are worse. However, comparing the pseudo sine
wave APC products (BackUPS) with the sine wave products (SmartUPS),
the SmartUPS products are much better. Among the various UPS's I own
and those I manage, 5 years battery life is about the average. If I
reduce the charge profile to charge slower, and terminate the EOC at
less than 100%, it won't run as long or recover as quickly, but the
batteries will last much longer.

I recently scribbled this rant about repairing APC BackUPS XS1300/1500
UPS's. Might be of some interest:
<https://groups.google.com/forum/?hl=en#!topic/sci.electronics.repair/hbJjN0a3Mis>
<http://www.11junk.com/jeffl/pics/repair/APC%20XS-1300/>


[1] One of the fun tests are running two UPS's in series. With a
pseudo sine wave, the 2nd UPS will often complain that the power from
the first UPS is too dirty to be useable.

--
Jeff Liebermann jeffl_at_cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Robert Baer
Guest

Tue Jan 19, 2016 8:30 am   



AncientOfyears wrote:
Quote:
I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.

NOT a good idea if it originally used 12V!!
I got a USED APC 330 XT+ (maybe 2-5 yrs old then) and it has been
providing excellent service for about 25 years; replaced battery about 4
years ago.

Rob
Guest

Tue Jan 19, 2016 5:10 pm   



Steve <nospam_at_nowhere.org> wrote:
Quote:
On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com> put
together some random words that came up with:

I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.

Did the same thing using a APC SmartUPS. Used the biggest marine batteries I
could find. Used it to power a phone system and alarm system. The thing
supplied power for around 4 days before the batteries went dead.


I did this years ago using an APC SmartUPS 1000 and two used sealed lead
batteries the size of a car battery (don't remember the capacity, was like
60Ah or so).

It did not work well. The charge current output by the APC (designed
for the standard 17Ah batteries) was not correct for those big batteries,
and they quite quickly became very unbalanced (one fully charged the
other one still mostly empty) and the UPS became confused.

I don't know if it was caused by the batteries already going dead (they
had been in a large PBX for some years) or by the mismatch between the
UPS designed capacity and the actual capacity. Maybe both were a factor.

Anyway, whatever you do make sure the connections are safe. The voltage
of the batteries is safe but the short-circuit current isn't!

Joe Hey
Guest

Tue Jan 19, 2016 8:20 pm   



On Mon, 18 Jan 2016 18:36:39 -0800
Jeff Liebermann <jeffl_at_cruzio.com> wrote:

Quote:
On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com
wrote:

I need to fixup a 1500VA USP to run longer.

How long? If more than a few hours, I suggest a Honda inverter
generator.

This is a pseudo sine wave unit.

On the bottom of my APC BE550G UPS is the following inscription:
The output of this device is not sinusoidal. It has a
total harmonic distortion of 45% and a maximum single
harmonic of 35%.
Assuming APC is talking about voltage distortion, a THD of 45% means
that
0.45^2 = 0.20 = 20%
of the total output power is wasted in harmonics of 60 Hz.


Sorry, but I disagree with that statement.
The voltage distortion is just that, and how much of those higher
harmonics are converted into heat is totally up to the load's
characteristics.

Loads with an inductive component (i.e.: R + jX) tend to have a much
higher impedance for the higher harmonics than for the base harmonic
(60 Hz), hence the power lost in the higher harmonics is lower than the
percentage of voltage distortion for that harmonic.
An additional factor that limits the Watt-losses (real power, or active
power) is the fact that higher harmonic currents have a bigger lag.

Quote:
If you're
driving a motor,


Yes. In a motor it can give torque pulsations and additional heating
(not proportional with the voltage harmonic distortion), but enough to
overheat the motor if it's operated at its rated power.
It has been a problem with PWM-driven motors in the past (apart from
the high dV/dt-s that are associated with switching converters, leading
to partial discharge in the insulation and eventually to failure).

> transformer

In a transformer it will potentially generate all kinds of sounds and
can lead to the same overheating problem as in the motor.

Quote:
, or have a switcher with a low pass input
filter,


In that case the switcher has solved the problem for you: the higher
harmonic voltages will not lead to (significant) higher harmonic
currents.

Quote:
or running a device that really hates harmonics, you're going
to have problem.


Yes, negative emotions like hate mostly lead to problems. :)

> Resistive loads are ok.

Yet those are the one to which your statement applies most.

Quote:
Switching power supplies
are usually ok.


Although their input capacitor can get severely stressed by the higher
harmonic voltages, leading to higher currents than the design may have
anticipated.

> Choose your loads wisely.[1]

That's funny. :)


joe

}snipped probably the most informative part{

Tom Del Rosso
Guest

Tue Jan 19, 2016 9:06 pm   



AncientOfyears wrote:
Quote:
Then I plan to hook up two 12V marine deep discharge batteries in
series to provide 24VDC using appropriate wire gauge.


Series? Did you mean parallel?

No way would it work with a higher voltage. Even in parallel the
charger might not like putting out more current.

--

Joe Hey
Guest

Tue Jan 19, 2016 9:38 pm   



On 19 Jan 2016 10:10:00 GMT
Rob <nomail_at_example.com> wrote:

Quote:
Steve <nospam_at_nowhere.org> wrote:
On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears
AOE_at_AOESPAM.com> put together some random words that came up with:

I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in
series to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.

Did the same thing using a APC SmartUPS. Used the biggest marine
batteries I could find. Used it to power a phone system and alarm
system. The thing supplied power for around 4 days before the
batteries went dead.

I did this years ago using an APC SmartUPS 1000 and two used sealed
lead batteries the size of a car battery (don't remember the
capacity, was like 60Ah or so).

It did not work well. The charge current output by the APC (designed
for the standard 17Ah batteries) was not correct for those big
batteries, and they quite quickly became very unbalanced (one fully
charged the other one still mostly empty) and the UPS became confused.

I don't know if it was caused by the batteries already going dead
(they had been in a large PBX for some years) or by the mismatch
between the UPS designed capacity and the actual capacity. Maybe
both were a factor.

Anyway, whatever you do make sure the connections are safe. The
voltage of the batteries is safe but the short-circuit current isn't!


And if you get a loose contact it might arc and not stop, unlike AC
where an arc extinguishes itself during a zero-crossing of the current,
leading to fire and no fuse blown.

joe

Jeff Liebermann
Guest

Tue Jan 19, 2016 11:10 pm   



On Tue, 19 Jan 2016 13:20:20 +0000, Joe Hey <joehey_at_mailinator.com>
wrote:

Quote:
On Mon, 18 Jan 2016 18:36:39 -0800
Jeff Liebermann <jeffl_at_cruzio.com> wrote:

On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com
wrote:

I need to fixup a 1500VA USP to run longer.

How long? If more than a few hours, I suggest a Honda inverter
generator.

This is a pseudo sine wave unit.

On the bottom of my APC BE550G UPS is the following inscription:
The output of this device is not sinusoidal. It has a
total harmonic distortion of 45% and a maximum single
harmonic of 35%.
Assuming APC is talking about voltage distortion, a THD of 45% means
that
0.45^2 = 0.20 = 20%
of the total output power is wasted in harmonics of 60 Hz.

Sorry, but I disagree with that statement.
The voltage distortion is just that, and how much of those higher
harmonics are converted into heat is totally up to the load's
characteristics.


The harmonics can also be reflected back to the UPS. In extreme
cases, the harmonics can also be radiated. As I previously mumbled,
the ability of the load to use the harmonic power is dependent on what
manner of load it presents to the UPS. A purely resistive load can
use it all. Something with a 60 Hz low pass filter on the input will
have problems.

Quote:
Loads with an inductive component (i.e.: R + jX) tend to have a much
higher impedance for the higher harmonics than for the base harmonic
(60 Hz), hence the power lost in the higher harmonics is lower than the
percentage of voltage distortion for that harmonic.


Those same inductive loads often have filter and bypass capacitors
across the line, which lowers the input impedance at higher
frequencies and are useful for passing FCC Part 15 conducted emissions
tests.

Quote:
An additional factor that limits the Watt-losses (real power, or active
power) is the fact that higher harmonic currents have a bigger lag.

If you're
driving a motor,

Yes. In a motor it can give torque pulsations and additional heating
(not proportional with the voltage harmonic distortion), but enough to
overheat the motor if it's operated at its rated power.
It has been a problem with PWM-driven motors in the past (apart from
the high dV/dt-s that are associated with switching converters, leading
to partial discharge in the insulation and eventually to failure).

transformer

In a transformer it will potentially generate all kinds of sounds and
can lead to the same overheating problem as in the motor.

, or have a switcher with a low pass input
filter,

In that case the switcher has solved the problem for you: the higher
harmonic voltages will not lead to (significant) higher harmonic
currents.

or running a device that really hates harmonics, you're going
to have problem.

Yes, negative emotions like hate mostly lead to problems. :)

Resistive loads are ok.

Yet those are the one to which your statement applies most.

Switching power supplies
are usually ok.

Although their input capacitor can get severely stressed by the higher
harmonic voltages, leading to higher currents than the design may have
anticipated.

Choose your loads wisely.[1]

That's funny. :)


joe

}snipped probably the most informative part{

--
Jeff Liebermann jeffl_at_cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Joe Hey
Guest

Tue Jan 19, 2016 11:39 pm   



On Tue, 19 Jan 2016 08:10:23 -0800
Jeff Liebermann <jeffl_at_cruzio.com> wrote:

Quote:
On Tue, 19 Jan 2016 13:20:20 +0000, Joe Hey <joehey_at_mailinator.com
wrote:

On Mon, 18 Jan 2016 18:36:39 -0800
Jeff Liebermann <jeffl_at_cruzio.com> wrote:

On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears
AOE_at_AOESPAM.com> wrote:

I need to fixup a 1500VA USP to run longer.

How long? If more than a few hours, I suggest a Honda inverter
generator.

This is a pseudo sine wave unit.

On the bottom of my APC BE550G UPS is the following inscription:
The output of this device is not sinusoidal. It has a
total harmonic distortion of 45% and a maximum single
harmonic of 35%.
Assuming APC is talking about voltage distortion, a THD of 45%
means that
0.45^2 = 0.20 = 20%
of the total output power is wasted in harmonics of 60 Hz.

Sorry, but I disagree with that statement.
The voltage distortion is just that, and how much of those higher
harmonics are converted into heat is totally up to the load's
characteristics.

The harmonics can also be reflected back to the UPS.


Yes, harmonic currents can do some bad things when encountered by a
power line filter.

Quote:
In extreme
cases, the harmonics can also be radiated.


Yes, that's a useful addition.

Quote:
As I previously mumbled,
the ability of the load to use the harmonic power is dependent on what
manner of load it presents to the UPS.


Maybe you should consider writing your mumblings down, because from
here I can only read what you write, not hear what you mumble. Smile
But IF I read you mumbling, then it was that 20% would be wasted in
harmonics (of 60 Hz--in Europe that would be 50 Hz, I guess). You
didn't write it as depending on the type of load. In what follows you
only wrote about the possibility of problems, depending on the load.

joe


Quote:
A purely resistive load can
use it all. Something with a 60 Hz low pass filter on the input will
have problems.

Loads with an inductive component (i.e.: R + jX) tend to have a much
higher impedance for the higher harmonics than for the base harmonic
(60 Hz), hence the power lost in the higher harmonics is lower than
the percentage of voltage distortion for that harmonic.

Those same inductive loads often have filter and bypass capacitors
across the line, which lowers the input impedance at higher
frequencies and are useful for passing FCC Part 15 conducted emissions
tests.

An additional factor that limits the Watt-losses (real power, or
active power) is the fact that higher harmonic currents have a
bigger lag.

If you're
driving a motor,

Yes. In a motor it can give torque pulsations and additional heating
(not proportional with the voltage harmonic distortion), but enough
to overheat the motor if it's operated at its rated power.
It has been a problem with PWM-driven motors in the past (apart from
the high dV/dt-s that are associated with switching converters,
leading to partial discharge in the insulation and eventually to
failure).

transformer

In a transformer it will potentially generate all kinds of sounds and
can lead to the same overheating problem as in the motor.

, or have a switcher with a low pass input
filter,

In that case the switcher has solved the problem for you: the higher
harmonic voltages will not lead to (significant) higher harmonic
currents.

or running a device that really hates harmonics, you're going
to have problem.

Yes, negative emotions like hate mostly lead to problems. :)

Resistive loads are ok.

Yet those are the one to which your statement applies most.

Switching power supplies
are usually ok.

Although their input capacitor can get severely stressed by the
higher harmonic voltages, leading to higher currents than the design
may have anticipated.

Choose your loads wisely.[1]

That's funny. :)


joe

}snipped probably the most informative part{


Rob
Guest

Wed Jan 20, 2016 3:54 am   



Tom Del Rosso <fizzbintuesday_at_that-google-mail-domain.com> wrote:
Quote:
AncientOfyears wrote:
Then I plan to hook up two 12V marine deep discharge batteries in
series to provide 24VDC using appropriate wire gauge.

Series? Did you mean parallel?

No way would it work with a higher voltage. Even in parallel the
charger might not like putting out more current.


No, those popular "shoebox size" APC and other makes UPS do have two
12v batteries in series to get 24v to operate the UPS.

Larger ones have even more batteries in series. Running at a higher
DC voltage makes it easier to get an efficient inverter to AC.

AncientOfYears
Guest

Wed Jan 20, 2016 8:07 am   



So what I am getting out of this is

Mod a UPS that uses only one battery i.e. 12V
Get a higher capacity SLA.

These UPSs I already have so I would not need to buy a larger USP.

May need several UPS to get the VA I need.

This is for an unattended security system.

I really need to do a cost benefit analysis on this.

Neon John
Guest

Wed Jan 20, 2016 8:22 am   



On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com>
wrote:

Quote:
I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.


Probably no reason to do that. Does it have a cooling fan? Put a
thermocouple on the main heat sink and see where the temperature
trajectory levels off.


Quote:
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.


Don't waste your money on consumer batteries. If you plan on using
wet cells, buy T-124 240 amp-hour 6 volt batteries. They'll easily
last a decade if properly cared for.

I strongly recommend AGM-type batteries. I have a set on my
whole-house UPS that are over 10 years old and still test out to near
full capacity.

Quote:
Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.


I can't speak to the new ones but I have an old APC 1500 VA unit that
treats the batteries with kid gloves. The charging voltages are
adjustable and I've set them to optimal for my particular battery's
chemistry.

John

John DeArmond
http://www.neon-john.com
http://www.tnduction.com
Tellico Plains, Occupied TN
See website for email address

pedro
Guest

Wed Jan 20, 2016 8:30 am   



On Mon, 18 Jan 2016 16:48:39 -0800, AncientOfyears <AOE_at_AOESPAM.com>
wrote:

Quote:
I need to fixup a 1500VA USP to run longer.
(snip)
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?


From the personal safety viewpoint, DO check that the battery system
is at or close to earth - ground to you furriners - and/or UPS output
neutral potential. What you *may* discover is that a lot of UPS
inverters have the batteries at half-AC-output voltage. Whether you
are at ease with that is something you should consider.

Cydrome Leader
Guest

Wed Feb 17, 2016 10:46 pm   



In sci.electronics.equipment AncientOfyears <AOE_at_aoespam.com> wrote:
Quote:
I need to fixup a 1500VA USP to run longer.
This is a pseudo sine wave unit.
First I plan to derate it by 50% so I will draw no more that ~700VA.
Remove the existing internal batteries and set aside.
Then I plan to hook up two 12V marine deep discharge batteries in series
to provide 24VDC using appropriate wire gauge.

So far maybe so good?
Suggestions please.

Using CyberPower rather than APC since I am hearing that APC eats
batteries before their time with too much testing.


Make sure the UPS has proper fan cooling for extended run.

I tried just this years ago and the UPS overheated and then exploded (all
the power transistors blew up). It was never designed for extended use off
larger external batteries. Whoops.

Goto page 1, 2  Next

elektroda.net NewsGroups Forum Index - Electronics Equipment - UPS Upgrade

Ask a question - edaboard.com

Arabic versionBulgarian versionCatalan versionCzech versionDanish versionGerman versionGreek versionEnglish versionSpanish versionFinnish versionFrench versionHindi versionCroatian versionIndonesian versionItalian versionHebrew versionJapanese versionKorean versionLithuanian versionLatvian versionDutch versionNorwegian versionPolish versionPortuguese versionRomanian versionRussian versionSlovak versionSlovenian versionSerbian versionSwedish versionTagalog versionUkrainian versionVietnamese versionChinese version
RTV map EDAboard.com map News map EDAboard.eu map EDAboard.de map EDAboard.co.uk map