Problem with a UPS unit and an Antec Power Supply

[Charlie Hoffpauir]

First, this is a home-built computer, with an Antec case and PS. The
3rd Antec case I've used for a homebuilt. The PS is a 500w Antec unit
that came with the case. (The PS is labeled as Antec Earthwatts 500w
max, model EA 500D.)


Problem description...

My UPS unit (APC 1250) failed to keep the computer running when we had
a severe power dip, so I concluded that the UPS was bad... (I had used
this same UPS with my previous computer, but since it seemed to be
working fine, I kept it when I built the new computer).

So I got a new APC 1500 (figuring bigger is better). Then another
power blip, and again the UPS failed to keep the computer on-line. So
I disconnected everything except the computer itself from the UPS unit
and ran a test by unplugging the UPS.... The UPS fired up but the
computer shut down.... but not immediately.

So I phoned APC support and they had me do a few more tests, including
unplugging the UPS with a normal resistive load connected (a lamp),
and the UPS worked fine, not a blink of the lamp. Another test with
the computer connected, and this time me watching the display on the
UPS, and the output voltage actually went crazy, varying quickly from
120 to less than 60 then immediately up to as high as 250 and back
down again, over and over until we unplugged the computer.... the
voltage then stabilized at 120.

The APC tech said that some power supplies needed a special UPS that
produced pure sine waves, rather than square waves.... and that they
were available from APC but were rather expensive. An alternative
would be to change the PS in the computer.

So, I exchanged the PS with an old one from another Antec case (Antec
TruePower 2, 480 w) and the computer seems to be once more in sync
with the UPS, that is I tested the setup by unplugging the UPS and it
powered up the computer just fine. Is there something fundamentally
different between the Earthwatt and the TruePower series of Antec PS?

Has anyone else heard of this, or was the tech just blowing me off?

If this is plausable, what would be a good replacement PS? I can
continue to use the 480w TP2 I suppose, but then what to use for that
computer?

Since I now seem to have 2 workable UPS units, would it "help" to have
both units connected, in series, so that power failure the unit that
the computer was connected to would get the square wave from the first
UPS, and perhaps filter it enough to keep the computer on-line?



I'll appreciate any insight.

 

[Bryce]

First, this is a home-built computer, with an Antec case
and PS. The 3rd Antec case I've used for a homebuilt. The
PS is a 500w Antec unit that came with the case. (The PS
is labeled as Antec Earthwatts 500w max, model EA 500D.)


Problem description...

My UPS unit (APC 1250) failed to keep the computer running
when we had a severe power dip, so I concluded that the
UPS was bad... (I had used this same UPS with my previous
computer, but since it seemed to be working fine, I kept
it when I built the new computer).

So I got a new APC 1500 (figuring bigger is better). Then
another power blip, and again the UPS failed to keep the
computer on-line. So I disconnected everything except the
computer itself from the UPS unit and ran a test by
unplugging the UPS.... The UPS fired up but the computer
shut down.... but not immediately.

So I phoned APC support and they had me do a few more
tests, including unplugging the UPS with a normal
resistive load connected (a lamp), and the UPS worked
fine, not a blink of the lamp. Another test with the
computer connected, and this time me watching the display
on the UPS, and the output voltage actually went crazy,
varying quickly from 120 to less than 60 then immediately
up to as high as 250 and back down again, over and over
until we unplugged the computer.... the voltage then
stabilized at 120.

The APC tech said that some power supplies needed a
special UPS that produced pure sine waves, rather than
square waves.... and that they were available from APC but
were rather expensive. An alternative would be to change
the PS in the computer.

So, I exchanged the PS with an old one from another Antec
case (Antec TruePower 2, 480 w) and the computer seems to
be once more in sync with the UPS, that is I tested the
setup by unplugging the UPS and it powered up the computer
just fine. Is there something fundamentally different
between the Earthwatt and the TruePower series of Antec
PS?

Has anyone else heard of this, or was the tech just
blowing me off?

If this is plausable, what would be a good replacement PS?
I can continue to use the 480w TP2 I suppose, but then
what to use for that computer?

Since I now seem to have 2 workable UPS units, would it
"help" to have both units connected, in series, so that
power failure the unit that the computer was connected to
would get the square wave from the first UPS, and perhaps
filter it enough to keep the computer on-line?

I'll appreciate any insight.

Cascading ("in series") the two UPS units will not help
and is a bad idea in general.

Antec's Earthwatt uses active power factor correction to
improve electrical efficiency. Electronics added to the
supply switches back and forth between charging a
capacitor from the rectified AC line and then discharging
into the DC storage capacitor in the UPS. The switch
points must match certain thresholds in the (normally)
sinusoidal AC voltage input waveform.

I suspect the active power factor correction used in the
Earthwatt is behaving very badly when it sees the square-
wave produced by the UPS operating on battery. As APC
indicated, a sine-wave UP$ $hould $olve the problem.
Hopefully, Antec will build power supplies that won't
misbehave this way.

Until then, avoid the Earthwatt series.

Bryce

 

[Charlie Hoffpauir]

Cascading ("in series") the two UPS units will not help
and is a bad idea in general.

Antec's Earthwatt uses active power factor correction to
improve electrical efficiency. Electronics added to the
supply switches back and forth between charging a
capacitor from the rectified AC line and then discharging
into the DC storage capacitor in the UPS. The switch
points must match certain thresholds in the (normally)
sinusoidal AC voltage input waveform.

I suspect the active power factor correction used in the
Earthwatt is behaving very badly when it sees the square-
wave produced by the UPS operating on battery. As APC
indicated, a sine-wave UP$ $hould $olve the problem.
Hopefully, Antec will build power supplies that won't
misbehave this way.

Until then, avoid the Earthwatt series.

Bryce

Bryce,

Thanks for that information. I'll look for another PS, and hopefully
it will specify whether it uses active power factor correction. Until
then the old Antec 480w unit seems do be doing fine... we had another
power blip this afternoon, and the computer didn't falter.

 

[Bryce]

Bryce,

Thanks for that information. I'll look for another PS, and
hopefully it will specify whether it uses active power
factor correction. Until then the old Antec 480w unit
seems do be doing fine... we had another power blip this
afternoon, and the computer didn't falter.

Antec ought to be interested in your experience. Lots of
desktop computers are powered from square-wave UPS units.
If their APFC circuit is incompatible, they have a problem
that needs addressing. It's also possible that your unit
might be defective. Perhaps they would be willing to swap
it for another: you get a (hopefully) useful unit and
they get one for quality control to look at.

Give them a call.

Bryce

 

[GlowingBlueMist]

Antec ought to be interested in your experience. Lots of
desktop computers are powered from square-wave UPS units.
If their APFC circuit is incompatible, they have a problem
that needs addressing. It's also possible that your unit
might be defective. Perhaps they would be willing to swap
it for another: you get a (hopefully) useful unit and
they get one for quality control to look at.

Give them a call.

Bryce

Good suggestion Bryce, any company that wants to pursue business computers
will need to be compatible with the existing base of UPS machines.
Expecting a company to upgrade all UPS's to "pure sine wave" technology
would be cost prohibitive, making the power supply company the cheapest to
replace...

With luck it is just a defective power supply and not an entire line of them
being sold by a company that does not have a clue...

 

[Charlie Hoffpauir]

Good suggestion Bryce, any company that wants to pursue business computers
will need to be compatible with the existing base of UPS machines.
Expecting a company to upgrade all UPS's to "pure sine wave" technology
would be cost prohibitive, making the power supply company the cheapest to
replace...

With luck it is just a defective power supply and not an entire line of them
being sold by a company that does not have a clue...

Thanks for the suggestions.

I submitted a problem report to customer support, describing what has
happened, and asked specifically if the Earthwatts series were
incompatible with UPS units. Since the PS is only 10 months old, it
should be covered under warranty (if indeed it is defective). However,
I guess it depends on how they define defective.... since it is
effectively powering my old computer, and it powered my new computer
since Feburary (except for when I had a power failure).

 

[Paul]

Thanks for the suggestions.

I submitted a problem report to customer support, describing what has
happened, and asked specifically if the Earthwatts series were
incompatible with UPS units. Since the PS is only 10 months old, it
should be covered under warranty (if indeed it is defective). However,
I guess it depends on how they define defective.... since it is
effectively powering my old computer, and it powered my new computer
since Feburary (except for when I had a power failure).

The reviews are always a good source of info. There are two models of
the EA500, a Seasonic unit and the latest is from Delta. Antec contracts
their manufacture to others.

http://www.newegg.com/Product/Produ...=False&VendorMark=&Keywords=(keywords)&Page=1

"nickname Worst PSU Ever 8/8/2009 1:04:21 PM

As many have reflected, this product is not compatible with the
majority of UPSs

Including: APC BR1500LCD, an (older) 900VA Cyberpower - Even when used
with (or w/out) an APC LE1200 Voltage Regulator.

Pull the plug & the machine re-boots. (again)
Repeatedly caused physical damage to devices within.

I called APC. IMO APC knew the PSU was the issue but instead of saying
such they just try and upsell you to their smart* (or better) $400+ UPS
line claiming the improved 'transfer time' will solve the issue.

I didn't have the guts to break the warranty/seal on the server until
recently & that's when i (looked up &) found other people reflecting
the (exact) same issues with this PSU."

The word gets around. And this is not the only supply design with this
issue. There are others. What doesn't make sense to me, is Europe has
a much higher percentage of units with power factor correction, and the
people there must own UPSes as well. So what has changed ? Do the
Europeans not get square wave UPSes ?

Picture here of a "modified sine wave", to show how crude the
approximation is. It is a square wave, with a bit of pulse width
(PWM) modification to roughly approximate a sine wave plus
harmonics.

http://img338.imageshack.us/img338/3759/waveformln6.gif

Paul

 

[Charlie Hoffpauir]

Ach. Forgot. In case you're interested.

http://en.wikipedia.org/wiki/Power_factor

Steve,

Thanks for your comments and the links. My intention now is to find
another PS that does not use active power factor correction rather
than spend a lot more for a UPS that produces pure sine waves.
However, I appreciate the link to the CyberPower unit. Form the other
link (about power factor) it seems that PS that use passive power
factor correction would be a better choice.... however.... I've looked
at several different PS of different makes on the newebb site and
nowhere does any of them state whether active power factor correction
is used. If this is such a problem, then it seems that the specs
should indicate the "feature". I guess I'll have to try the
manufacturers sites on any PS I'm considering.

 

[Charlie Hoffpauir]

It was the least expensive one I could find.


Granted that a new non active pfc would be cheaper, I got the UPS
because APFC is supposed to lower your power drain from the mains.
It looks like Europe has alrady gone to APFC for new computers and
I expect the USA and others will follow. It's greener, you see.



Most do if you dig down far enough.


Go to newegg:

http://www.newegg.com/Product/PowerSearch.aspx?N=2010090058
&SubCategory=58&GASearch=3


http://preview.tinyurl.com/y8t9892

and down the list is a place to check for no PFC.

That's a great search aide....

Before I got your message on that , I went to the Enermax web site and
entered a question about their Power supplies with Active Power Factor
Correction; told them I was interested in a 500 w model. Here's their
response:

If you use our ECO80+ 500W EES500AWT you will not have an issue with
square sine wave UPS.

So apparently it is not a problem with "all" Active Power Factor
Correction models or brands. Now I'm waiting to see what Antec says
before I buy another PS.