Dell XPS Studio 435T needs a UPS

[Anonymous user]

OK, this Dell is a fairly NEW design. It has a 470W PSU that is Energy Star 4.0 compliant and is a PFC PSU.

I now have an APC XS 1000 600W UPS. When I run the TEST it doesn't hold up the Dell, power drops. The XS 1000 Powerchute s/w reports only 218W draw? I've read all the articles and they say for a PFC PSU you need 1.25 rating of the PSU to work for sure due to surge in-flux. That would be about 590W's.

OK, maybe it is MARGINAL... and I called APC and we ran some tests, the switch works fine, a monitor and printer and some other stuff like external drives on it hold power under test?

To be sure I purchased from Best Buy a 1285VA (750W) UPS. It didn't hold either?

Now Dell tells me I need either an ON-LINE or DOUBLE CONVERTER UPS. These, compare to the cost of the computer are rather high. In addition, I've not found anyone near me that carries these, all the stores carry 'home' UPS's. Ordering on-line would be OK IF* I knew it would work. The cost of returning a UPS is not cheap.

How can I be sure even that a specific UPS will work? I'm thinking an APC SUA1000 should do the trick, but I can't tell for sure what type it is. I assume the main problem is the TRANSFER time from power line to battery and back? Most UPS's, that state transfer time, is 2 - 4 milliseconds, which I assume it too slow?

Thanks,
Irv S.

[Anonymous user]

OK, this Dell is a fairly NEW design. It has a 470W PSU that is Energy Star 4.0 compliant and is a PFC PSU.

I now have an APC XS 1000 600W UPS. When I run the TEST it doesn't hold up the Dell, power drops. The XS 1000 Powerchute s/w reports only 218W draw? I've read all the articles and they say for a PFC PSU you need 1.25 rating of the PSU to work for sure due to surge in-flux. That would be about 590W's.

OK, maybe it is MARGINAL... and I called APC and we ran some tests, the switch works fine, a monitor and printer and some other stuff like external drives on it hold power under test?

To be sure I purchased from Best Buy a 1285VA (750W) UPS. It didn't hold either?

Now Dell tells me I need either an ON-LINE or DOUBLE CONVERTER UPS. These, compare to the cost of the computer are rather high. In addition, I've not found anyone near me that carries these, all the stores carry 'home' UPS's. Ordering on-line would be OK IF* I knew it would work. The cost of returning a UPS is not cheap.

How can I be sure even that a specific UPS will work? I'm thinking an APC SUA1000 should do the trick, but I can't tell for sure what type it is. I assume the main problem is the TRANSFER time from power line to battery and back? Most UPS's, that state transfer time, is 2 - 4 milliseconds, which I assume it too slow?

Thanks,
Irv S.

[Erasmus_apc]

You do not need a true online or double conversion online UPS. The issue with the Back-UPS units is that they often have trouble with PFC power supplies, most notably Active PFC power supplies. As you have already stated, load vs. UPS power factor can also come into play when sizing a UPS.

The transfer time for a Back-UPS unit is slower than for a Smart-UPS unit. Typical Back-UPS transfer time can be as "slow" as 4-8ms. Smart-UPS transfer time is typically 2-4ms. Smart-UPS' also provide a pure sine wave output vs. the Back-UPS' stepped approximated sine-wave, which can cause some loads to hiccup or drop. Additionally, the SUA units have a better overall power factor and handle inrush better than Back-UPS units can. All of the above adds up to the SUA units being a better solution for sensitive power supplies or supplies that use PFC.

The SUA Smart-UPS units have been in production for years now and only the most sensitive equipment (nuclear medical cameras, highly sensitive laser equipment, IT equipment that requires 0ms transfer time) have ever had trouble with them going on battery and back.

[Anonymous user]

I have been following the APC / Dell XPS 435T saga. I hve spent some time on the phone with tech support. I purchased a second APC Back-UPS XS 1500. As soon as commercial power drops the PC has a siezure...I measure the on-battery output voltage and it holds steady @ 115v...the output Frequency drops immediately 30hz......load or no load..... Has anyone had any similar measurement results...?
Thanks

[BillP]

Hi Joe,

I hope you can answer the following questions:
# What is the exact model of the Dell XPS PC and the power supply that it comes with?
# Where did you buy the UPS from?
# What is the exact load on the UPS?
# Who recommended the UPS/solution?

[Anonymous user]

I installed the SAU1000 yesterday and it works JUST GREAT..

At this point, I'm assuming the problem was not the transfer time but the shape of the sine wave, simulated vs. true.

Interestingly, the load factor between the Back-UPS 1000 and the Smart-UPS 1000 differ? The Back-UPS reported 216W used most of the time. The Smart-UPS shows about 20 to 30% load, but usually around 28% or so. At 600W capacity of the UPS, this is around 168W's or so, much lower than the Back-UPS. I guess the UPS is more 'efficient'?

Anyway, the PSU works with this UPS.

Irv S.

[Erasmus_apc]

Minor differences in the current sensors on the Back-UPS and Smart-UPS units can cause this, as well as the fact that your load can and will fluctuate in small degrees. Good to know that the Smart-UPS unit is working!

[Anonymous user]

Hi, Thought I would say thanks, I have a XPS435T to and last night my APC1000 would not run it during a thunderstorm outage last night, was going to call dell but read your post here. So I had to get one ASAP but 400 dollars for one is alot to spend, Found a SUA1000 on ebay sold by a non profit group computers for kids, for 158 used and free shipping, they had a few, and come with new batterys, and cables and a 1 yr warrenty, not bad. Can't wait to get it.

Message was edited by: whstark

[Anonymous user]

I bought from the same seller. But I got it for $128. I noticed he listed them at various prices, I just bidded for the lowest priced one! Just received it today. The top is a little scratched up, but overall it's in good condition. I opened it up but can't tell if it's a new battery. I don't think so, as it's the original APC battery, and there was a layer of dust on the battery! You would think if they put in a new battery, it would be a cheaper aftermarket.

I'm having a problem downloading PowerChute Business Edition v8.0.0. Download page wont' load. Don't know if it's some bug with IE8 compatibility issues, or on the APC site. I have registered and did log in.

Anyone knows where or how I can download the software?



Thanks.

[Anonymous user]

I can't get it down now either. Suggest you contact APC directly or ask in maybe the General Discussion forum? Something has to be wrong on APC's end...

Irv S.

[BillP]

yes i think there was a problem with the website.

if you still cant get to it by tomorrow, we can get you a different link.

[Anonymous user]

I was very pleased to find this thread while attempting to diagnose spontaneous reboots with a Dell Studio XPS 435t on an APC UPS related to minor power fluctuations. At first I thought this was because my APC was undersized, but I purchased a new one, an XS 900VA, and had the same issue. Any time I either 1) unplugged the UPS from the wall or 2) switched on my stereo's power amp, the Dell 435t would suddenly reboot. What is interesting, however, is that the 435t does NOT reboot if plugged in to the wall when I switch on the power amp, and it also does not reboot under either of the two circumstances if plugged into a CyberPower 685AVR UPS...a relatively, inexpensive home UPS. This latter finding tends to suggest that the problem is not with home UPS units per se, but is with the nature of the power wave put out by APC's home UPS units, and/or the way APC units respond to brownouts or outages. I have experimented repeatedly, and I have not yet seen the Dell reboot when connected to the 685AVR, whether I unplug the 685AVR entirely, or whether I cycle power on my power amp.

[Anonymous user]

I am confused...but the CP685AVR only has a 390-watt output, not enough to power your 435T system.

[Anonymous user]

The 685AVR works as a temporary measure at least, but most likely because I am not taxing my 435T during testing, and not drawing anywhere near the potential maximum. I also have a 22" LCD monijtor powered through the UPS. What is significant is that the 435T does not reboot no matter how many times I disconnect the 685AVR from wall power or no matter how many times I switch on the stereo power amp. I have not tried to run the system on the 685AVR battery power for more than 10 or 15 seconds, as the only issue I am trying to resolve is the spontaneous reboot issue during transition.

I have decided to gamble that a newer 1000VA/600W CyberPower model will behave similarly, and I have ordered one online. If it does not work, I will have to eat return shipping costs of course.

I would very much like to understand specifically what is at play here. Why do two different APC UPS models trigger Dell 435T reboots during transition to battery, yet the CyberPower 685AVR (although undersized for permanent use with this system) does not?

Message was edited by: retronaut

[Anonymous user]

What confuses me is that the XPS 435T has a 470W PSU...does that mean it uses 470W?? If that is correct, then how does the 685AVR with only 390W support the 435T on battery backup?? Don't forget there are other peripherals on top of the computer, like monitor, sound system, etc.

I don't really understand the terms and physics behind all this, but the APC UPS models uses the term "Stepped" while CyberPower uses "Simulated" for the sine wave output. I don't know if there is a difference.

Now if the CyberPower (higher capacity model that you ordered) works, I'd be Darned! I should have ordered that brand new one for less instead of a used APC SUA1000 for more money! Let us know how it works out.

[Anonymous user]

I looked at the Cyberlink page. You'd probably need the 1200 model to be safe, costs $160.

Cyberlink uses the word 'simulated' for the sine wave. APC uses 'stepped' on the BACK-UPS and 'pure' on the SMART-UPS.

I did try from Best Buy a 'Geek Squad' UPS which is a re-badged Tripplite that was a 1200VA unit and it failed too.

If the Cyberlink works, great.

The power requirements are always worst case don't forget, and PFC supplies can draw 25% more than rated at startup. However, my old APC which reports the actual power said the nominal draw for the 435T and a 21" LCD display was less than 250W's. A 400VA unit might be capable of handling the load, but the run-time would be low on battery.

Irv S.

[Anonymous user]

The article at http://www.mainresource.com/tipups.html MAY offer an explanation as to the difference between the APC and the CyberPower:

"When a power failure occurs and the UPS switches power to the battery, the signal generated by the inverter and supplied to the equipment may not be a true sine wave (it may be called a "modified" or "simulated" sine wave). The modified or simulated sine wave can range from a square wave to a trapezoidal wave. These pseudo-sine waves are the product of less expensive components than those used to make true sine wave generators. Computers using the power supplies found in most desktop PCs and servers will work fine with square waves, better with trapezoidal waves, and best with true sine waves. Generally speaking, only continuous on-line UPSs require the true sine wave to be effective."

Perhaps the simulated wave from APC home UPS models is a square wave, while that from the CyberPower (at least from the 685AVR) is trapezoidal.

??

Message was edited by: retronaut

Message was edited by: retronaut (correct typos)

[Anonymous user]

retronaut wrote:The article at [http://www.mainresource.com/tipups.html] MAY offer an explanation as to the difference between the APC and the CyberPower:

"When a power failure occurs and the UPS switches power to the battery, the signal generated by the inverter and supplied to the equipment may not be a true sine wave (it may be called a "modified" or "simulated" sine wave). The modified or simulated sine wave can range from a square wave to a trapezoidal wave. These pseudo-sine waves are the product of less expensive components than those used to make true sine wave generators. Computers using the power supplies found in most desktop PCs and servers will work fine with square waves, better with trapezoidal waves, and best with true sine waves. Generally speaking, only continuous on-line UPSs require the true sine wave to be effective."

Perhaps the simulated wave from APC home UPS models is a square wave, while that from the CyberPower (at least from the 685AVR) is trapezoidal.

{color:#0000ff}--------------------------------------------------------------------------------------

That makes sense....possibly the case.{color}

[Anonymous user]

Sir,
So, what's the bottom line on buying a UPS for my Dell 435T? The Trip-Lite I bought at Costco didn't work. The APC XS 1300/1500 doesn't work either.

The APC doesn't switch to battery power; it just turns OFF. I need one which can operate on the battery for a while.

Please restrict the sine-wave jargon and just tell me what will work with my computer.

Barry

[voidstar_apc]

Get a a [4G SmartUPS 1500|http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SUA1500] or a [5G SmartUPS 1500|http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SMT1500].

[Anonymous user]

Thank you. I'll try that. Barry

[BillP]

Hi,

found this thread and I have XPS 435T also and have Back-UPS XS-1200. Hopefully this provides some insight. I initially had XPS 435 tower plugged into Battery sockets on UPS and tower would always power-off when UPS switched to battery power. The Back-UPS uses modified sine-wave and power supply in XPS 435T cannot handle that (we know this). However, I moved the XPS 435 to the 'surge only' outlets on UPS and now I withstand many of the little power dips we get here. Of course, anything longer than a second (or two), the XPS 435 will shut off since it is on 'surge only'.

But this has saved me from buying Smart-UPS at about $600. If you want XPS 435T to run on battery, you will need to get Smart-UPS (pure sinewave output). It might be possible to change power-supplies on the 435T to one that tolerates a modified sinewave signal, but probably not recommended unless you really know what you are doing.

[BillP · Accepted solution]

[this may of be some help too|http://nam-en.apc.com/cgi-bin/nam_en.cfg/php/enduser/std_adp.php?p_faqid=1493]