UPS recommendation for late 2013 Mac Pro

[vett93]

I am looking for a new UPS for my Mac Pro, a late 2013 model. I would like to have the following 2 features:

1. Controlled battery powered outlets: I want to shut off the power to external HDs when the Mac Pro is in sleep mode. My current APC BackUPS XS1000 has controlled outlets. But these outlets are only surge protected, but not battery powered.

2. Sinewave output:

I can't seem to find one that meets the above two requirements. Any pointer from others?

 

[MarkC426]

Cyberpower ups’s are ‘pure sinewave’
If your hard drives are connected to the mac do they not go off when your mac sleeps?
I do not know of a feature to isolate certain sockets.....

 

[vett93]

These HDs have their own power adapters. These HDs are powered on 24X7. I have seen UPS that has a master-slave setup in that when the computer is in sleep mode, it will cut off power to these slave outlets. But these slave outlets are only surge protected, not battery powered.

 

[bsbeamer]

You may need to use a power strip and have this work off your monitor's power plug for the control:
https://www.amazon.com/dp/B01DP9M3DO
https://www.amazon.com/dp/B0006PUDQK

Basically whenever your monitor goes to sleep, it SHOULD trigger the power strip to switch the controlled outlets to OFF.
When it wakes up it SHOULD trigger power back ON to the controlled outlets.
Not sure I'd run this off your Mac Pro since power at sleep likely will not trigger the outlets/control since it still is consuming power.

As posted above, many CyberPower are pure sine wave models. Using 1500PFCLCD and 1000PFCLCD for different purposes. I highly recommend CyberPower. Have had fewer long-term issues vs. some other big names.

 

[bjar]

These HDs have their own power adapters. These HDs are powered on 24X7. I have seen UPS that has a master-slave setup in that when the computer is in sleep mode, it will cut off power to these slave outlets. But these slave outlets are only surge protected, not battery powered.

If you did this wouldn’t the Mac say that the drives were not ejected properly when you wake it up?

 

[vett93]

With the power strip solution, the HDs will not have UPS. My concern is that HDs may run into problems during power outage.
[doublepost=1537827035][/doublepost]

If you did this wouldn’t the Mac say that the drives were not ejected properly when you wake it up?

If the Mac goes into sleep mode first and then power gets cut off the HDs, shouldn't it be okay?

 

[bsbeamer]

Power strip gets plugged into battery backup outlet on UPS.
The HDDs and monitor will have UPS protection.

During power outage, nothing SHOULD be triggered since monitor is on battery backup and not triggering (controlling) the additional outlets. The only thing changing is source of power from outlet to battery.

 

[vett93]

Power strip gets plugged into battery backup outlet on UPS.
The HDDs and monitor will have UPS protection.

During power outage, nothing SHOULD be triggered since monitor is on battery backup and not triggering (controlling) the additional outlets. The only thing changing is source of power from outlet to battery.

Thanks for the explanation. This makes more sense. Thanks. However, there is one thing that I need to point out. Most UPS manufacturers are against putting a surge protector on UPS outlet. They claim there would be some interference between the UPS and the surge protector.

 

[AidenShaw]

Thanks for the explanation. This makes more sense. Thanks. However, there is one thing that I need to point out. Most UPS manufacturers are against putting a surge protector on UPS outlet. They claim there would be some interference between the UPS and the surge protector.

The "interference" would certainly occur with non-sine-wave UPS systems running on battery, since the surge protector would look at the square wave current as "noise" and try to smooth it. And "smoothing" it would mean discarding some of the battery-generated electricity - which would reduce run-time on battery. Never seen any claims about damage, though, just warnings about reduced run-times.

With a true sine-wave UPS, the surge protector wouldn't see significant noise.

 

[bjar]

With the power strip solution, the HDs will not have UPS. My concern is that HDs may run into problems during power outage.
[doublepost=1537827035][/doublepost]

If the Mac goes into sleep mode first and then power gets cut off the HDs, shouldn't it be okay?

The Mac Pro May unmount them. I have not tried it. I know on my Mac book if I have an sd card in the slot and remove it while the laptop is asleep, it says the card was not ejected properly when I wake it back up. Just something to think about.

 

[ActionableMango]

These HDs have their own power adapters.

Don't they spin down on their own when not in use? My hard drives are external with their own power adapters and they spin down when not in use for some period of time. This happens even if the computer is on, and definitely when the computer is off.

You might check this hard drive setting:

If the Mac goes into sleep mode first and then power gets cut off the HDs, shouldn't it be okay?

I would think so. When the computer goes to sleep it should flush all of the write caches so there is no pending disk activity.

I'm speaking of normal file system use of course. I wouldn't know what would happen in some unusual scenario, for example the computer being told to sleep right in middle cloning one drive to another, or encrypting a drive, etc.

 

[vett93]

Don't they spin down on their own when not in use? My hard drives are external with their own power adapters and they spin down when not in use for some period of time. This happens even if the computer is on, and definitely when the computer is off.

You might check this hard drive setting:

View attachment 788298​

I would think so. When the computer goes to sleep it should flush all of the write caches so there is no pending disk activity.

I'm speaking of normal file system use of course. I wouldn't know what would happen in some unusual scenario, for example the computer being told to sleep right in middle cloning one drive to another, or encrypting a drive, etc.

Yes, my external HDs would spin down when they are not used. Does it have benefits to shut off power to them when the computer is in sleep?

 

[westom]

Most UPS manufacturers are against putting a surge protector on UPS outlet. They claim there would be some interference between the UPS and the surge protector.

"Interference" is not described with perspective (numbers). They do not want to discuss numbers for obvious reasons. No such thing as a pure sine wave UPS. Anything that creates power from a battery has an 'impure' sine wave. How much? That means also posting a number such as %THD. Notice, nobody posts that or any other relevant number. In part because UPS manufacturers don't want to discuss their 'impure' pure sine wave.

This UPS was also marketed as a pure sine wave UPS. It outputs 200 volt square waves with a spike of up to 270 volts. They did not lie. We all learned this in high school math. Square waves and spikes are nothing more than a sum of pure sine waves. So even that is a sine wave output.

That interference is damage created by spikes, square waves, stepped waves, etc output by a UPS in battery backup mode. It can damage a protector or (less often) damage a UPS.

Why do you need a pure sine wave? Hype. Fear. Urban myths. That 200 volt square wave with spikes of up to 270 volts is problematic to less robust motorized appliances. And just fine for electronics.

More numbers they do not discuss. Your computer filters incoming power. Then converts it to DC. And filters it again. Then converts that to well over 300 volt radio frequency spikes. Why is 'dirty' power from a UPS problematic when your Mac powers semiconductors from well over 300 volt spikes? Best hardware protection is galvanic isolation, filters, and regulators that convert that 300 plus spikes into rock stable, low DC voltages to safely power semiconductors. Semiconductor protection routinely exists inside all electronics.

Best: do not power a power strip protector from a UPS. A power strip without protector parts is safe. Ignore urban myths that promote a need for 'pure sine waves'. That fear comes from hooey; not from engineering knowledge and well proven reality. Always demand specification numbers with each recommendation. So many subjective recommendations are based only in hearsay and wild speculation; not from engineering knowledge that was so well understood long before PCs existed.
[doublepost=1537881766][/doublepost]

Yes, my external HDs would spin down when they are not used. Does it have benefits to shut off power to them when the computer is in sleep?

More fears that need not exist. No HD is told power will be turning off. A HD only learns about power off when DC voltages start to drop. That is how disk drives worked even long ago when disk heads were moved by motor oil.

All disk drives see DC power disappear. Finish what it is doing. And safely shutdown. Nothing warns in advance of a power cutoff. Even those external HD are not 'protected' by a UPS.

Sleep mode means a computer still needs AC power. Only power off modes that do not need some AC power are shutdown or hibernate.

 

[AidenShaw]

How much? That means also posting a number such as %THD. Notice, nobody posts that or any other relevant number. In part because UPS manufacturers don't want to discuss their 'impure' pure sine wave.

http://www.apc.com/shop/us/en/produ...-Backup-Surge-Protection-SMC1500C-/P-SMC1500C

 

[Ludacrisvp]

Anytime I remove a USB drive from a sleeping Mac it will instantly wake the Mac which IMHO would defeat the purpose you are attempting to achieve.

 

[AidenShaw]

Ignore urban myths that promote a need for 'pure sine waves'.

Modern high efficiency PSUs like sine waves...

https://dl4jz3rbrsfum.cloudfront.net/documents/CyberPower_Active_PFC_Solutions.pdf

"Shutdown can occur because simulated sine wave output has a power gap in each cycle. When the UPS system switches to battery current, a power supply with an Active PFC circuit may detect that power gap and shut itself down."​

https://www.dougv.com/2010/03/active-pfc-enabled-psus-are-not-compatable-with-most-low-end-ups/

 

[MarkC426]

Still don’t understand why you want them on a seperate power though?
I have my external drive power connected to ups backup socket.
When mac sleeps, drive also sleeps.

I know when I upgraded a version of OSX (can’t remember version), I originally had drive connected via fw, and the drive would spin even when mac asleep, changed to usb and fixed it.

 

[vett93]

Modern high efficiency PSUs like sine waves...

https://dl4jz3rbrsfum.cloudfront.net/documents/CyberPower_Active_PFC_Solutions.pdf

"Shutdown can occur because simulated sine wave output has a power gap in each cycle. When the UPS system switches to battery current, a power supply with an Active PFC circuit may detect that power gap and shut itself down."​

https://www.dougv.com/2010/03/active-pfc-enabled-psus-are-not-compatable-with-most-low-end-ups/

View attachment 788658​

For the record, my late 2013 model Mac Pro has been running on an APC UPS that generates modified sinewaves for 4+ years. It does not crash when the power goes out and UPS switches to battery power.

My concern is more about the noises from the square waves on the components. I recently had to reformat the SSD. I have no proof that it is related to the UPS. But if I am going to get a new UPS, it might as well be sinewave.

 

[westom]

If AidenShaw knew what number was relevant, then he cited that number. Anyone can post a massive color glossy sales brochure .... that does not even define a pure sine wave. He posted no numbers because recommendations only come from what advertising, hearsay, speculation, and subjective reasoning (also called junk science reasoning) has told him to believe.

Well, if one reads all that stuff, then surge protection provided by a UPS is about 230 joules and never more than 460. A surge that miniscule is routinely converted by electronics into rock stable, low DC voltages to safely power semiconductors. UPS does not even do what is already done routinely and better inside electronics.

He only recommends a UPS for 'cleaner' power because propaganda (without any spec numbers) orders us to believe what we are told.

Just as ridiculous is a myth that power is cleaner because is has no gaps. Again, anyone who first learns technology before making recommendations knows that AC power is always electricity with gaps. In fact, computer standards even define how long power can be removed - and internal DC voltages much not vary even by 0.2 volts.

That modified sine wave is not only perfectly good for all electronics. It gets converted into 300 plus volt radio frequency spikes with many gaps. Even that is perfectly ideal power that results in rock stable, low DC voltages. Since it is all done better inside a PSU.

All this stuff, claimed as necessary for a Mac, was already done better inside computers even over 40 years ago. Claims are only subjective because advertising propaganda has created mythical fears.

Even 'interference" was some subjective myth created when basic electrical concepts were never learned. Myths exist when one does not always demand reasons why with numbers. That 'interference' is really 'dirty' electricity output by a UPS when powered from batteries. 'Interference' that quickly degrades near zero joules inside a power strip protector. And in some cases, might cause UPS damage.

So back to what the OP wants. That (dirty) sine wave from most every UPS is perfectly good for all electronics. Second, some disk drives (or its external drive hardware) can be set to sleep when not accessed after a time period. Third, external disk drives, powered by a USB bus, get powered off when that computer powers off. Four, externally powered HDs can be powered from a 'smart' power strip. That cuts off power to other receptacles when the first item (ie computer) stops demanding power.

Unfortunately, a UPS powered from that power strip would not cooperate. But then a UPS is only implemented to provide temporary and 'dirty' power so that unsaved data can be saved. Is it necessary?

 

[AidenShaw]

For the record, my late 2013 model Mac Pro has been running on an APC UPS that generates modified sinewaves for 4+ years. It does not crash when the power goes out and UPS switches to battery power.

Which model APC?

The cheaper ones us a "Stepped approximation to a sinewave" - which is smoother than the square wave. The more steps, the closer the approximation. An infinite number of steps would give you a sine wave.

 

[vett93]

Which model APC?

The cheaper ones us a "Stepped approximation to a sinewave" - which is smoother than the square wave. The more steps, the closer the approximation. An infinite number of steps would give you a sine wave.

I have the APC Back UPS XS1000.

 

[AidenShaw]

I have the APC Back UPS XS1000.

Do you hearing any buzzing or other noise when it's on battery?
[doublepost=1537904688][/doublepost]

That modified sine wave is not only perfectly good for all electronics.

Proof that that statement is wrong:

https://forums.anandtech.com/threads/can-apcs-stepped-sine-wave-damage-my-power-supply.1661111/

"I have 3 UPSes. One is a cheap "modified sine wave" one that killed the electronic ballasts in 2 compact fluorescent lights within minutes of plugging them in. In both cases, the CFL bulb buzzed for a couple seconds then went out. I've used this UPS with a variety of power supplies, linear and switching, over the years and never had any buzzing nor PSU failures."​

 

[westom]

My concern is more about the noises from the square waves on the components.

Does not matter what the incoming wave is. First, it is filtered. Then it is converted to DC. Then filtered again. Then converted to well over 300 volt radio frequency spikes. Those spike are powering your internal component - regardless what UPS waveform exists.

Then superior filters, galvanic isolation, and regulators convert that 'dirtiest' spikes into rock stable, low DC voltages to safely power semiconductors.

This need for pure sine waves is hooey promoted to naive consumers to increase profits. That 'pure sine wave' is necessary for less robust motorized appliances.

If worried about 'noise', then worry about what "well over 300 volt radio frequency spikes" are doing to your components. Since only galvanic isolation, filters, and regulators separate that from your components.

Notice that engineers, who really know this stuff, are not concerned.

 

[ActionableMango]

All this stuff, claimed as necessary for a Mac, was already done better inside computers even over 40 years ago. Claims are only subjective because advertising propaganda has created mythical fears.

My personal experience says otherwise.

For mission critical communication applications we sell Dell computers and the cheaper sinewave approximated UPS units. This was fine for a long time until about 8-9 years ago when Dell switched to new high efficiency power supplies. When we tested a power outage, everything connected to the UPS stayed on except the computer, which immediately turned off. When we contacted Dell about this, they admitted that the new high efficiency power supplies were too sensitive and that the supply treated the stepped approximation as a power loss. Seeing these computers turn off wasn't a myth caused by advertising propaganda--it was a real problem that we had to solve because obviously this is unacceptable in a mission critical application.

So we tested many computers and many UPS units and discovered mostly that there isn't really a consistent rule as to whether computers would work with simulated sinewave UPS or not. Many computers would work, but some like those Dells wouldn't. In other cases a computer would stay on, but its power supply would loudly "squeal" when using simulated sinewave UPS--yet these same computer were perfectly quiet when connected to a sinewave UPS. Also, from a previous thread, at least one other user here has said his cheesegrater Mac Pro power supply squeals on his simulated sinewave when it is using battery. I have no idea what the squeal is, or whether or not it's bad for the electronics, but it doesn't matter because loud squeals are not acceptable in a communication environment where dispatchers need to talk to radio and telephone users. The only consistent rule in our testing was that sinewave UPS worked with all computers all the time and with no squeal.

So you can say whatever you want about it all being BS. But in the real world we experienced real problems with PCs on stepped sinewave UPS, Dell confirmed to us that this issue was a known problem with a reason that matches what @AidenShaw has stated, and switching to sinewave UPS fixed the problem. Note this was a loss for Dell, as we returned all of the UPS units and switched to a different UPS vendor with a wider selection.

Now it wouldn't surprise me if newer high efficiency power supplies have improved to the point where this isn't a problem any more. But (A) that wouldn't make the problem a myth. (B) This is the Mac Pro forum where many many people are using old computers. The most common model talked about here is the 4,1 and 5,1, which are 8-9 years old, right in that period where we first started noticing the problem with high efficiency power supplies. And (C) for mission critical applications we aren't going to take the chance in case the customer switches to a PC with incompatible high efficiency power supplies in the future.

 

[AidenShaw]

Does not matter what the incoming wave is. First, it is filtered. Then it is converted to DC. Then filtered again. Then converted to well over 300 volt radio frequency spikes.

Whether this is universally true is not the point.

The point is that some active PFC power supplies monitor the input power and decide that the long zero voltage gap between the square waves is a power outage. They're happy with noisy sine wave inputs, but not 4.16 msec of zero voltage.