High end UPS for Expensive audio eequipments


I just had a sudden power outage while my expensive audiophile system was running.

It really gave me the scare of my life thinking of the aftermath.

Would you folks know of a good protector of high powered amps?

Thanks in advance,


Much depends on where you live.
Here on the East Coast, our power is pretty stable ( though I am now typing this on UPS supplied current due to an unexpected power outage).
Some parts of the US have more variable power and a power conditioner is of greater value.
As far as adding something in front of an amp, the guys at McCormack said, Nay-nay.- Just turn it off should the power go out.
I have had the power go off and on a few times while listening, and the amp was fine. (My other equipment was behind a power conditioner).
I have installed a whole house surge protector last year. I believe that it protects surge that comes from the outside power line. But it does not help if the surge is from within your home (start-stop of fridge, aHVAC unit, washing machines, etc). My power amp is direct to the wall, which is a 20A dedicated power line. The rest of the electronics are connected to Brick Wall PW8R15AUD surge protector, which is attached to another dedicated line.
If we’re talking purely on the basis of investment protection, Series Mode Surge suppressors are the only one’s I truly trust.

The technology is licensed by PS Audio, Surge-X and Furman, but Furman tends to have the lower prices.

A UPS often introduces more problems than they solve, and they are no guarantee of protection from surges.


Without prejudice to the performance degradation issues from the insertion of the surge protector unit and its filtering effects, High-end audio gear usually already has high-end quality build power supplies that do not benefit from any intermediary plug-in surge protection or power conditioners.

if surge protection gives you added peace of mind, then only add it as a quality build system-wide insertion by a licensed electrician at the Electrical panel rather than ANY wall plug-in accessory.
the best protector is anticipating a problem; were you aware of questionable weather conditions and ignored the warning signs of a potential problem?  

were you scared enough to learn this lesson: disconnect, and listen safely later, when the atmosphere is stable

Live on the east coast USA. No warning at all. Weather was sunny.

Just a sudden blip followed by outage.

Thanks for all the responses guys. Please keep it coming.


get insurance?
I have the same setup as milpai, so whole house surge protector, dedicated lines, and Brick Wall on sensitive (digital) gear. The power lines here are underground, I have stable power, but I still pull the plugs when I hear that thunder getting close.
akg_ca is unfortunately incorrect when it comes to surge protection. A lightning bolt, or induced voltage from an EMF will happily go through most modern equipment's power supplies.

HOWEVER, some is going to be more sensitive than others. Tube amps and pre-s for instance are more tolerant due to naturally high working voltages to begin with. Tubes are easy to replace too. :)  Devices with CPU's to control displays, remotes and EQ however are most at risk.

It is however up to a great deal of debate as to whether or not improving the AC signal will have any subjective effect, and many power supply designers feel that so long as the voltages are within spec (110V-130VAC) the user will not be able to perceive any difference.

Personally I find the addition of Furman's LiFT to really open up the treble.  Maybe I'm high. :)  I'm no power conditioner cognoscenti though,


APC S-10 or S-15. Proven, dead quiet, reliable, and not a rip-off like so many power conditioners are. Get one and don't look back.
@bojack -
I had terrible experience with an APC UPS. Not for audio, but for my desktop PC.. It worked well for about a year, and slowly I started having issues with my PC crashing or hanging for no apparent reason.
I pulled out my (non existent) hair trying to figure out the problem. Replaced memory cards, reinstalled the OS, replaced the power supply... So much frustration. So many hours wasted. Even took it to the shop which gave it clean bill of health and charged me $60..
It ended up being the UPS polluting the line with so much noise that caused the PC to crash. I hooked it up to an oscilloscope, was shocked at what I saw.
UPS removed, problem solved.
I will NEVER buy another APC product again.
There are many different grades of UPS. PC/electronic are usually the very noisiest, but APC does make A/V specific, sine wave output versions. 

Severely overpriced for my needs.


akg_ca is correct when it comes to surge protection. Power supplies already contain robust protection that makes most surges (including mythical ones from household appliances) irrelevant.  Concern is for something that might occur once every seven years.  A rare anomaly that can blow through electronics must be earthed at the service entrance.

Others have recommended a 'whole house' protector.  But did not recommend THE most critical component in that protection system - earth ground.  An effective protector must connect low impedance (ie less than 10 feet) to single point earth ground.  Essential even if incoming wires are underground.  Absolutely necessary to protect plug-in protectors or a UPS.

No protector does protection.  Effective protectors (as installed for free on phone, satellite dish, and TV cable) always have a low impedance connection to single point earth ground.  Protection is the item that harmlessly absorbs hundreds of thousands of joules.  Plug-in boxes have no earth ground and will not discuss it. A protector is only as effective as its earth ground.


Please describe which power supply design you are talking about? I haven’t done an exhaustive study, but after examining say, two dozen devices, including DAC’s, mostly amplifiers and preamps, headphone amps, etc., I have yet to see a single one that included "robust" surge protection. I will occasionally see UL listed "noise" filters which are not the same thing, so I really would like to see what you are talking about. The one location where I have seen consistently and extensive use of surge protection is in PC power supplies.

Westom, you are also misinformed about the latest surge technology. Series mode does not connect to ground. They use a very large coil (compared to most surge protectors) instead which becomes the high voltage point in the circuit, and dissipate surge energy as heat instead of current. Their UL listing proves the effectiveness, and the normal "joule" rating becomes irrelevant.

A ground circuit is not needed in an SMP, so the quality of the ground for this application is no more needed than for GFCI circuits. As for your parallel devices being "harmless" we had about 2 dozen of these "harmless" protection devices (MOV’s?) in surge strips flame when a glitch in the local switching station occurred. They did protect the systems attached, but the speed of the surge was a lot slower than lightning.

But the cost / value item is another issue. I find the Furman devices inexpensive enough, and my gear precious enough to worry about it.


Power supplies already contain robust protection that makes most surges (including mythical ones from household appliances) irrelevant.
That's absurd. Few devices contain serious protection against power surges. And I'm not sure why you call appliance surges "mythical" when they can be both heard and measured.

An effective protector must connect low impedance (ie less than 10 feet) to single point earth ground ... Effective protectors (as installed for free on phone, satellite dish, and TV cable) always have a low impedance connection to single point earth ground. 

That's not only mistaken, but typically a violation of current code, at least in the US. Multiple ground rods are now required at a service entrance and if that's not part of an installation now, it will almost certainly be required if the service is ever upgraded. Also, the length of the connection to the ground rod is not usually the determining factor of the safety ground's impedance - that's established by soil conditions. Moreover, a household's ground isn't really through those grounding rods - those are just the safety grounds. The real ground is back through the utility connection - as required by code - unless you have some type of electric problem or malfunction.

@cleeds I think @westom is talking about how traditional surge protection works, which is to short over-voltages to ground.

I should note, just for being complete, that the series-mode protection is (by design) a low-pass filter so it works great on power lines but if applied to Cable TV or satellite signals it would block the signal too, so it's application is limited to AC line protection.

Because of this, units for home have multiple filtering "modes" for POTS (copper phone), LAN, Cable, etc. even if they use series-mode for power.



>  Please describe which power supply design you are talking about?
Long before PCs existed, international design standards required 120 volt electronic to withstand up to 600 volts without damage.  Today's PSU are more robust.  I recently saw a Seasonic spec that defined protection up to 1800 volts.  

Appliances already have robust protection.  To know otherwise means a denial comes with numbers.

Series mode filter is not a surge protector.  Lately those manufacturers have hyped a filter as a protector. Hyping it in sales brochures as a surge protector means a massive profit increase.  As naive consumers spend $hundreds only because they are told they need it.  How many consumers now believe a UPS does surge protection?  Most.  Subjective lies (rebranding it as a surge protector) can massively increase profits since so many ignore numbers.  Furman is another example of big hype and near zero protection.  Your reply should include Furman numbers that define protection.  Good luck finding them.

How many joules does that series mode protector absorb?  About 600.  A 600 joule surge is easily converted by electronic power supplies into rock stable, low DC voltages to safety power semiconductors.   That series mode protector is protecting only from something that is not destructive.  Otherwise you would have posted numbers for it rather than a subjective (advertising) expression.

Nobody said an earth ground is needed for a SMSP.   Where did that come from?

MOVs that flame mean a protector was grossly undersized.  Designed in violation of what MOV manufacturers require. That catastrophic failure must never happen.  MOV datasheets make that obvious.  Plug-in protectors typically claim to protect only from near zero surges (ie 600 joules).  Contain circuits to disconnect protectors parts as fast as possible.  And leave that surge still connected to equipment.  Due to superior protection, equipment protects itself from tiny surges that destroy or flame a near zero protector.

Effective protection is for surges that might overwhelm robust protection already inside every appliance.  Facilities that cannot have damage properly earth a 'whole house' solution.  It costs tens or 100 times less money.

A surge too tiny to overwhelm protection inside equipment turned a near zero (ineffective) protector into flames.  Fires are created by near zero joule protectors.  APC recently admitted some 15 million protectors must be removed immediately due to fire.

And finally, UL says nothing about hardware protection.  UL is only and completely about protection of human life.  Who is the UL?  National Fire Protection Association.  An ineffective (near zero) protector can still be UL Listed.  UL says nothing about protecting appliances.  Never cite UL as proof of effective surge protection.

Once learned, those many facts expose myths that promote obscenely profitable and near zero protectors.

>  Also, the length of the connection to the ground rod is not usually the determining factor of the safety ground's impedance - that's established by soil conditions. 

You have confused resistance with impedance.  Wire length, sharp bends, metallic conduit, etc can significantly increase impedance and not change resistance.

From Jensen Transformers' "Understanding, Finding, & Eliminating Ground Loops in Audio & Videa Systems":
> An EARTH ground is one actually connected to the earth and is necessary for LIGHTNING protection.

From Mike Holts  "Grounding vs Bonding":
>An effective, low-impedance ground path is critical for the successful operation of an SPD. ... Therefore, an evaluation of the service entrance grounding system at the time of the SPD installation is very important.

From Dr Standler's book "Protection of Electronic Circuits from Overvoltage":
> It is essential that every arrester be connected to ground, because the charge in a lightning strike flows to ground. However, providing a low-impedance connection to ground is one of the most difficult practical problems in installing a surge arrester.

From George Kauffman in Electrical Engineering Times:
> Another aspect of impedance ... of a wire is predominately related to
> its length and weakly related to its diameter. ... The length of the
> cable increases the impedance dramatically.

From CRITEC's Technical Note TNCR019:
> The rule of thumb is each foot of wiring adds an additional 50-200V of let-through voltage.

Mike Holt even defines voltages for a tiny 500 amp surge on a 14 AWG wire:
> 6 inches  460 V; 20 inches 582 V; 30 inches 782 V. ...

You can do the math.  That is not resistance.  That is impedance.

Denials come only from hearsay; not from basic electrical knowledge and not from informed professionals. Power outage does not damage hardware. Adjacent protectors do not provide effective protection.  Effective protection always means a potentially destructive transient is not inside a building.  Then robust protection inside all equipment is not overwhelmed.

600V is what I would consider an over-voltage, not a surge. A surge can be kilovolts, though in some cases very low current potential. However that is all it takes to fry silicon. These same circuits that you are touting as being good surge protection is absent in most power supplies (with the exception of PCs).  I would also need to research this as I believe this may apply to electric / motor driven appliances as opposed to electronics. In any event, there's usually a time limit for exposure.

SMP devices will clamp well below that, around 200-300V I believe, but that's besides the point.

Um, your comments about series mode protection are in contradiction with the UL surge protection ratings, and your insistence on relying on Joule ratings means you don’t understand the physics involved. No current = no joules. The same for your ridiculous claims of undersized MOV’s. I’m kind of done with discussing this with you, since you seem to be unwilling to do your homework and are using a dual standard. Of course manufacturers will hype their product, whether they make series or traditional surge protectors.

Have a nice day.


westom09-14-2016 1:40pm You have confused resistance with impedance.
No, I'm not confused at all. But like @Eric_Squires, I'm done arguing with you and your notions of a "single ground point" through a grounding rod and other misinformed - and potentially lethally dangerous - "theories."
cleeds has not even asked what 'single point earth ground' means.  And somehow is an expert?  Single point earth ground must both meet and exceed code requirements.  As was clearly obvious.  He could change tact and start learning what these expressions really mean.  But he won't.  Some people get attached to what advertising has told them.  And then are incapable of learning science; learning how easy advertising brainwashes.

Single point earth ground is standard in every facility that does not have damage. Even implemented in the earliest days of telephones so that operators worked without harm through every thunderstorm. 

cleeds is an example of one entrenched by propaganda.  Unable to learn simple science concepts that even Ben Franklin demonstrated in 1752.  A classic naysayer who knows only he is right.  He cannot even post one manufacturer specification number.

Plenty more to learn. Blackouts and brownouts do not damage hardware.  Most anomalies are already made irrelevant by what is inside electronics.  Anomalies that can overwhelm existing protection are rare.  So informed consumers spend $1 per protected appliance on protection that is far superior to any 'magic' plug-in' box.  Since strip protectors can even compromise superior protection inside appliances if not used in conjunction with a proven and properly earthed 'whole house' solution.

A simple rule applies.  Protection is always about where hundreds of thousands of joules harmlessly dissipate.  A protector is only as effective as its earth ground.

>  Um, your comments about series mode protection are in contradiction with the UL surge protection ratings, and your insistence on relying on Joule ratings means you don’t understand the physics involved. No current = no joules. The same for your ridiculous claims of undersized MOV’s. 

Again UL says nothing about hardware protection.  Anyone with basic electrical knowledge knows that.  UL is only concerned with human safety.  A protector can be near zero protection, not protect anything, fail catastrophically, and still be UL listed.  Because it did not spit sparks and flames.  UL says nothing about protection.  If I say it again, will you finally grasp it?  UL says nothing about hardware protection.  If you cannot get past that, then I must conclude you are incapable of learning even simplest stuff. 

UL1449 is about human safety.  Unfortunately some near zero joule protectors still create house fires. To meet the latest UL1449 upgrade, many protector manufacturers had to increase their joules.  A recent example were some 15 million APC protectors. Why ignore examples of near zero protectors that even create house fires.  Do you need another 50 examples to finally admit advertising lies?

Series mode protectors will 'absorb' up to 600 joules.  If not, a denial would quote a spec number.  No numbers quoted because the denial is an emotion; not based in facts or numbers.  Series mode filters protect from surges that typically cause no damage.

600 volts is protection standard in electronics before PCs existed.  Appliances today are even more robust.  Since you deny specifications (because you don't like it), well, this quote is from Dr Standler's 1992 IEEE paper:  :
> This paper is apparently the first publication in the peer-reviewed, archival engineering literature that specifically discusses the ability of unprotected electronic equipment to survive surges in a laboratory. Surprisingly, consumer electronic equipment was able to survive surges with peak voltages of 2000 V,

600 volts.  1800 volts. Those were specification numbers.  Standler found 2000 volts is a more realistic number.  Why remain entrenched on 600 volts - a number from 40 years ago? Apparently only to argue..

Protection is not about a voltage.  High voltage only exists when someone foolishly tried to 'block' or 'absorb' a surge.  Protection of expensive equipment is never provided by urban myths - 'magic boxes'.  Effective 'whole house' protector never foolishly tries to 'block' a surge - therefore create a high voltage.  Instead, near zero voltages exist when effective protection connects a current (not a voltage) low impedance to earth.   At what point should we mention it is science proven over 100 years ago.

Facilities that cannot have damage always use a 'whole house' solution.   That is, beyond doubt, OP's best recommendation.  It even comes with numbers that say so.

Impedance - not resistance.  Naysayers do not even know what impedance is.  Do not even know what single point earth ground is.  And then make bogus accusations because they did not know it;  They still do not know what single point earth ground is.  Do not have a clue what is or is not dangerous.  Apparently near zero protectors creating fires is not dangerous - their reasoning.

Best protection for the OP is a 'whole house' solution from other companies known by any guy for integrity.  Furman and APC are not on that list.  A protector is only as effective as its earth ground - including that low impedance (ie less than 10 foot) connection. Reality has not changed because near zero joules manufacturers said so.  A protector is only as effective as its earth ground - item that does the protection.

Short of getting an electrical engineering degree or repping one of the major power conditioning companies (w/huge discounts on the entire line), this topic boils down to common sense and exhaustive completeness:
1 - Get a whole-house surge supression device at the electrical panel...something very robust that can take a big spike. An electrician can install it. It's almost certainly better than any local outlet devices
2 - But audiophiles being the OCD creatures we are, go ahead & install any pricey surge suppressors or high end power modification devices you want ahead of the big $$$ gear, paying special attention to line level tube equipment, as someone else here noted
3 - I have a UPS in the home office--it's perfect for desktop computers, which can lose data in a millisecond's outage. But the real solution is a whole house system. Very expensive, but worth it. After the 3rd or 4th extended winter power outage (upstate NY), I finally installed a propane-fired Generac generator w/automatic transfer switch. It's big enough to carry the entire house (20 KW). My home office billing justified the cost and I never looked back
4 - But even with all that, the vulnerability that remains for many is the incoming cable TV or satellite TV wire. Lighting can light up anything attached to those wires, and typically there's no surge protection for those TVs, receivers, or other audio gear--or the surge protection only stands between the power outlets and the gear. There are surge suppression systems that allow cable connection, but that gets tricky w/2-way communication protocols; most people simply disconnect their cable wires at a central location whenever the weather becomes threatening...


Like everything else in this discussion, your reading is half-baked. UL 1449 includes testing for effectiveness as a surge protection device in addition to life safety issues like not starting fires after a surge. Do some reading before posting.

Series mode surge protection devices have the best (lowest) UL tested "let through voltage" or VPR for test surges. meaning that during tetsting the equipment being protected would suffer a much lower voltage than with MOV parallel devices.

This is why fairly reliable companies such as Furman, SurgeX, and PS Audio rely on them for their best protection, and have NO MOV’s or joule ratings (for AC protection) and have no sacrificial components (in the AC protection).  Again, if you had bothered going through the design, or reading the spec you wrote so eloquently incorrectly about you'd know this.


Well the absolutely best buy for price, performance, reliability and pure sine wave output is a professional rather than audiophile type UPS with double conversion.

I have one running my entire system for 3 years 24hr/day without interruption.
There is one caveat. The fans run continuously and cannot be placed in the same room as the equipment due to the noise of the fans.

I run a line from the basement closet under the stairs to my basement home theater and also to the main floor 2 channel sound system.
LOAD capacity 1500 VA / 1350 WATTS - higher power available -
Can be purchased at AMAZON for $785
or in Canada for CDN $850


CYBER POWER correction
CDN $996
CYBER POWER cheaper price if you discard remote web control
US$ 627


@ Westom

" And finally, UL says nothing about hardware protection.  UL is only and completely about protection of human life.  Who is the UL?  National Fire Protection Association."

Um, not to confuse what you've written with actual facts, but UL (Underwriter's Laboratories) and the NFPA (National Fire Protection Association) are two separate and distinct organizations. 

Further, while UL is obviously concerned with human life, their standards are written to help protect devices and structures as well.  "Only and completely about protection of human life" is an incorrect statement.  Sorry. 
Correctly noted is my mistake with NFPA and UL. Both are separate and address human safety issues. Both address device protection and structure protection relevant to human safety. Neither says anything about quality of product functions. Otherwise a UL number that says "how" good is provided. No number is provided because none exists. UL only tests for threats to human safety.

UL1449 says nothing about protecting appliances from a destructive surge. It only discusses protection of humans from a protector failure. Protector can be grossly undersized and still be UL listed as long as it fails in a manner that does not threaten human life.

A protector can fail catastrophically during a UL1449 test. Still be UL listed even if it does near zero protection for an appliance. No way around reality.

Effective surge protection means one can say where hundreds of thousands of joules harmlessly dissipate. Where does UL discuss that number? It doesn’t. UL is not about appliance protection. UL is only about human protection - ie it spits no sparks and flames.

A plug-in and UL listed protector (ie type 3) must be distant from the breaker box since being too close to earth ground means that tiniest protector can be a human safety issue: http://www.nemasurge.org/what-is-spd/

Effective protection means a destructive surge (ie 20,000 amps) is connected low impedance (ie less than 10 feet) to single point earth ground. Then hundreds of thousands of joules are not inside blowing through near zero plug-in protectors. Where does a naysayer discuss these or any other relevant number?

No numbers means a denial is bogus. At best, probably wild speculation. Informed consumers learn to ignore recommendations that are subjective. Subjectively recommended were ineffective and grossly overpriced products from companies such as Furman and Monster. Not one manufacturer spec number from Furman or Surgex is posted. Because neither claim to protect from destructive surges. Such products are marketed to consumers who only want to be told what to buy. And get angry when spec numbers (reasons why) are discussed.

Posted were direct quotes from professionals who know and do this stuff. Why does one know this stuff? Because one was doing it even before an IBM PC existed. Why is well understood science denied by another who also ’feels’ treble is opened up? Power to an amplifier’s transistors is unchanged by series mode filters. Amplifier power still comes from filtering and regulating well over 300 volt radio frequency spikes. Those spikes exist with or without a series mode filter. Why would sound from transistors change when nothing into those transistors changes? Somehow ’feelings’ justify science. Bulloney.

Circuits make cleanest AC voltages into well over 300 volt radio frequency spikes. Then clean that ’dirtiest’ power. Best protection is already inside high end amps. Consumers concern is an anomaly that can overwhelm that already existing and robust protection. That means earthing a best solution (that costs about $1 per protected appliance). Then even direct lightning strikes (ie 20,000 amp) are made completely irrelevant by this device rated at least 50,000 amps.

Best protection is also a least expensive solution. With numbers that say why.

Best protection for high end amps and other other appliances means one knows where hundreds of thousand of joules are harmlessly absorbed. Done by UL approved devices that meet or exceed above numbers. And not done by UL approved devices that only claim to absorb near zero joules. Both are UL approved since human life is not threatened. But only a properly earthed solution claim to protect from what might be destructive.

Consumers are warned that ineffective and near zero joules devices are only recommended when no supporting facts and numbers are provided. Just as foolish is to assume UL tests product performance like Consumer Reports does. A near zero (ineffective) protector can still be UL listed as long as it does not threaten human life as it fails prematurely.

Best protection for high end amps always was a ’whole house’ solution, rated at least 50,000 amps, and with a low impedance (ie less than 10 foot) connection to single point earth ground.
Series mode surge protection devices have the best (lowest) UL tested "let through voltage" or VPR for test surges. meaning that during tetsting the equipment being protected would suffer a much lower voltage than with MOV parallel devices.

Difference between longitudinal mode and metallic mode currents was never learned. That is even taught to every first year engineering student. So the engineer will again expose a half truth in a let-through voltage assumption.

Assume a 5000 volt surge is incoming on the hot (black) wire. If approaching a typical protector rated at 330 let-through volts, then 5000 volts is still incoming on a hot wire. And 4670 volts is incoming on a safety ground (green) and neutral (white) wire. Where is protection?

If a surge has a lower let-through voltage, then that 4670 volts increases closer to 5000 volts. Without first year electrical knowledge, well, demonstrated is why a near zero (600 joules) filter was recommended to somehow avert hundreds of thousands of joules. And why a near zero joule solution is recommended subjectively. By discussing this one spec number, demonstrated again is why advertising (not science) was the source of that recommendation.

Please learn what a let-through voltage measures before making recommendations. Please post numbers with your every recommendation to learn about basic electrical concepts you did not know.  Recommendations without citing the relevant specification number is best ignored as if junk science.