Compatibility between PSU and UPS...

[jones_r]

I want to protect the investement I did in my computer with buying a UPS (Uninterruptibe power supply).

The problem is, I read somewhere that UPS that has a "simulated sine wave", is not compatible with PSU's which have PFC (and the PSU I'm using - SS 550HT - has active PFC).

Now, there are also UPS devices that have real sine waves (not simulated), but those have two problems: 1. cost more, 2. have a fan built in, that works all the time, which creates a lot of noise.

So, what is the solution ?.

[burebista]

I have an APC Smart-UPS 420VA 230V and it work flawless with S-12 430 and M-12 500.

[ayemooth]

my Belkin works fine with my S12-430, and the 400W Super Tornado I used previously. The UPS' specs sheet describes its output as "Simulated Sine Wave".

[jones_r]

This is the definition of "PFC"

[jones_r]

burebista,

I checked your UPS, and it is providing *real* sine waves, not simulated. For this reason it is also more expensive, and have a fan which makes it noisy all the time.


ayemooth,

It could be that you PSU runs hotter than it would have run, if you connected it straight to the wall without the UPS.

[burebista]

PFC has nothing to do with efficiency of a PSU, it doesn't make it nor hotter nor colder.
In fact for a home user a PSU with PFC is useful only to squeeze all output Watts from an UPS. For example my UPS is 420 VAR and rated for 260W output. With an APFC PSU I'll have ~400W output from my UPS.
That's all.

[jones_r]

burebista,
But a manufacturer who specifically makes UPS devices, wrote that PSUs with PFC should be used with real sine-wave UPS devices (like the one you have). Why did he say that ?.

[Oleg Artamonov]

_Some_ PSU's with active PFC (old FSP such as PLN and PFN series, some Delta PSU's, some Hipers and Thermaltakes) may be incompatible with mid-range UPSes. Most modern PSUs works well with any UPS.

Seasonic PSUs are ok, there should be no problems with any UPS, no matter sine wave or approximation.

P.S. "When the PSU was working with an APC SmartUPS SC 620, a load of 360W and 345W for the mains and batteries, respectively, was the highest allowable for the PSU. The switching to the batteries was performed without problems"

[jones_r]

Thanks Oleg.

[jones_r]

Another question: Does anyone know if a typical line-interactive UPS can provide backup to a computer in "Stand by - S3" mode ?.

The voltage requirement in S3 mode is extremely low, which leads me to believe the UPS might think the computer is close and hence not provide current from the batteries in order to save the information in the computer's RAM, in case the electricity goes out when the computer is in S3 mode.

[Oleg Artamonov]

As far as I know, most UPSes will provide output voltage even without any load at all.

[EsaT]

I don't see why any UPS wouldn't work without load and why any maker would even want to do that.

[burebista]

OK, maybe I' wrong, but from here we have:
True Power=Power Factor x Apparent power

On my UPS: 260W= Power factor x 420 VA, hence we have a PF= 0.62
With my APFC M-12 I'll have True power = 0.98 x 420VA, hence total output = 411W.

I miss something?

In fact here on SPCR we have a conclusion:

[Oleg Artamonov]

Things not so simple.
Active PFC is a step-up converter designed in a way that current is proportional to AC voltage, so it works like an almost linear load with PF ~ 1.

But A-PFC designed to work with sine wave input, while UPS output voltage (except models with true sine wave) is far from sine, it has steep slopes and I can't say how efficiently A-PFC will work with such voltage.

I'm going to begin UPS tests next month, may be there will be more experimental data including power factor measurements with "simulated sine wave" input. Up to now all I test is PSU and UPS compatibility ("work with load up to ...W" or "don't work"), nothing more.

[EsaT]

[burebista]

OK. If I understand correct PSU's APFC has nothing to do with UPS's output, right?
So, me as home user which is billed in kWh when I buy an APFC PSU I pay for a "feature" which is useless for me, even it reduce PSU's efficiency by 2-3% and it's useful only for my local electricity company to reduce their reactive load.
Now that's nice.

[jones_r]

First of all it seems to me that if more electricity is circulating, more heat will be generated due to losess. So to me it seems higher PFC = less heat = less noise.

Second, regarding this:

[EsaT]