Built in AVR on UPSes - is it really necessary?

[Anonymous user]

I've read that all switch-mode power supplies, those used on PCs and most eletronics for a long time now, are switch mode power supplies and they're are known to work just fine with voltage fluctuations of about 30% above or below it's nominal rating, 115V for instance. AVRs usually corrects the voltages fluctuations to 6-12% up or down depending on the input voltage. So if 130 volts is coming in, the AVR reduces it for about 115V at the best cases. 130 volts is well withinn the tolerance of a swtich mode power supply and it could withstand a 150volts without being damaged, as I've read. I've also read that AVRs that make use of electromagnetic relays (I don't know if I'm calling it the right way) to switch between one voltage input or another, introduces wanted interference on the power being fed to the power supply and can be harmful at some cases.

With the advent of power supplies with active PFC, the need for voltage regulation is even less necessary I mean, not necessary AT ALL! Monitors and many printers and scanners power supplies also work at a range of 100-240volts. They adjust themselves on the fly to work with any voltage fluctuation, so no need for voltage regulation as well.

I've been seeking for an UPS that is 220V in/out without AVR that also outputs a true sine wave, but I can't seam to find one in my country and even overseas. I even tried other international APC websites, but I can't find a cheap unit without voltage regulation and a true sinewave. Most of the power supplies on my set up are 100-240V capable, the power supply of my PC is one with Active PFC, so that's why I'm trying to find one good and cheap UPS without voltage regulation and also with a true sinewave because my Zalman 460watts is not happy with the stepped sinewave. It makes a buzzzing noise when on battery and I fear this will shorten its lifetime.

Could anyone indicate me one UPS that fits my needs?

Thanks for any reply!

Message was edited by: rau

Message was edited by: rau

[Anonymous user]

rau wrote:
I've read that all switch-mode power supplies, those used on PCs and most eletronics for a long time now, are switch mode power supplies and they're are known to work just fine with voltage fluctuations of about 30% above or below it's nominal rating, 115V for instance. AVRs usually corrects the voltages fluctuations to 6-12% up or down depending on the input voltage. So if 130 volts is coming in, the AVR reduces it for about 115V at the best cases. 130 volts is well withinn the tolerance of a swtich mode power supply and it could withstand a 150volts without being damaged, as I've read. I've also read that AVRs that make use of electromagnetic relays (I don't know if I'm calling it the right way) to switch between one voltage input or another, introduces wanted interference on the power being fed to the power supply and can be harmful at some cases.

With the advent of power supplies with active PFC, the need for voltage regulation is even less necessary I mean, not necessary AT ALL! Monitors and many printers and scanners power supplies also work at a range of 100-240volts. They adjust themselves on the fly to work with any voltage fluctuation, so no need for voltage regulation as well.

I've been seeking for an UPS that is 220V in/out without AVR that also outputs a true sine wave, but I can't seam to find one in my country and even overseas. I even tried other international APC websites, but I can't find a cheap unit without voltage regulation and a true sinewave. Most of the power supplies on my set up are 100-240V capable, the power supply of my PC is one with Active PFC, so that's why I'm trying to find one good and cheap UPS without voltage regulation and also with a true sinewave because my Zalman 460watts is not happy with the stepped sinewave. It makes a buzzzing noise when on battery and I fear this will shorten its lifetime.

Could anyone indicate me one UPS that fits my needs?

Thanks for any reply!

A lot of PSUs are indeed rated for 86 to 264v. Those with a selector are often 90 ~ 132V on low, and 180 ~ 264V on high.

I'm not aware of sinewave UPS without a AVR, but the AVR circuit serves two purpose, so it's not entirely a waste.

There is a large transformer inside the UPS that powers the charger as well as serve as the main transformer for on-battery operation and the primary side is multi-tapped C, -12%, 120v and + 12% (usually). The transformer is just idling under normal conditions, but when the input voltage falls out of range, relays re-routes the output to -12 or +12 tap as the controller sees fit. This is the AVR portion.

When it's on battery, the transformer is used backward. The secondary becomes primary. With freshly charged battery, the normal tap can provide enough voltage at output and PWM will accommodate declining battery voltage to maintain output voltage. As battery voltage keeps declining as the capacity is depleted or the load increases, it will reach a point where PWM compensation no longer maintains the output within limits. Instead of shutting down, relays mechanically switch taps to increase the transformer boost ratio bringing output voltage back within range again. It allows the UPS to squeeze a little more runtime on battery.

This is going beyond what you asked, but older APC BackUPSs actually had two transformers. One large one dedicated for on-battery up-converting and a small one for overnight charging. This topology offers no AVR, but since the large transformer is not regularly energized it uses substantially less phantom power just to sit around. A 450VA BackUPS uses only a 2-3W fully charged and idling, while a 1400VA SmartUPS 20 watts under same conditions using about $20 in energy per UPS/year just to sit idle.

[Anonymous user]

Relays.

[Anonymous user]

I understand the transformer part you explained.

In my country the use of AVRs is almost compulsory. It is inconceivable for most of the computer users and even technicians to use a computer without an AVR. They don't even cogitate the possibility of not using one. There is this big belief here that AVRs protect the PC from any kind of power problem except for power outages, of course. Most of the AVRs sold here doesn't even come with surge suppression! Some people are beginning to change their minds, I myself was one of these "believers" a year ago that are were "converted".

So the AVR market in my country is huge. We have many brands making them, including APC. On computer forums some people are educating others that AVRs are in most cases a completely waste for modern PCs. The side affects they might bring and the things o use to protect your PC are also being discussed. Some may say the utility power is unstable in my country, but that is not totally true. What is true is that many homes have their wiring installation done by people who does not understand what they're doing and are done completely out of the standards.

Please, read these statements below and tell me what your opinion is:


+"Myth #4. VOLTAGE-REGULATING TRANSFORMERS ARE HELPFUL WITH COMPUTERS.+

+Most modern desktop computers use switch-mode power supplies rather than older style, linear designs. A switch-mode power supply draws from the AC power line only as much energy as it requires to maintain its output power. In this sense it responds spontaneously to voltage fluctuations. If the line voltage drops, the power supply draws current for a longer period, until it replenishes the energy it put out since the previous cycle of the power wave. Because of this natural ability to accommodate varying source voltages, a switch-mode power supply gain no benefit from a voltage regulating transformer. However, switch-mode power supplies are more sensitive to source impedance than source voltage, and the increased impedance inserted into the line by the transformer may actually hinder the power supply by restricting the current available. A tap-switching, voltage-regulating transformer may also introduce noise if the tap switch hunts back and forth between adjacent output taps. Computer switch-mode power supplies often have a wider tolerance for input voltage than do regulating transformers themselves. Thus the primary benefit of a voltage regulating transformer is its leakage inductance, which is much less than that of an isolation transformer, but the regulator introduces offsetting disadvantages. Moreover, the transformer's promary function, voltage regulation, offers no material benefits."+

And please read what this Voltage Regulator manufaturer says about the implications of using one:

Voltage Regulator Problems - Full Text at: http://www.teal.com/products/App%20note%20AN-2.htm

+Voltage regulators are relatively simple devices, but they may reduce the reliability of an electrical system. In addition, the cost vs. benefit ratio of a voltage regulator is sometimes much higher than other power conditioning technologies. As a result, voltage regulators should be prescribed with care.+

Efficiency: Some voltage regulators expend a lot of energy in order to operate. This affects the electrical cost-to-operate, and also increases the facility air conditioning load. At low power levels (such as a workstation) this may not be a problem, but at power levels above 1 kVA the efficiency becomes significant. Ferro-resonant and magnetic synthesizers in particular are very inefficient; while tap-switchers and electro-mechanical units may be quite efficient.

Reliability: Voltage regulators with active components are susceptible to failure rates much higher than transformers and other passive electronic components. Electro-mechanical regulators require regular maintenance for the motors, belts, and brushes that provide regulation. Tap-switching regulators contain large switching SCR's or Triacs that can fail due to line or load problems. Designers and end-users must factor in the cost to maintain and service a voltage regulator into the "cost of ownership" of such a device.

Load Interaction: Any voltage regulator may cause a load interaction, especially with sensitive equipment that contains pulsing loads. Examples of this type of equipment include industrial process machines, medical imaging equipment, and printing presses. When load interaction occurs, load changes or internal voltage regulation may cause the external voltage regulator to resonate or mis-operate. This actually causes voltage problems (sags and surges) on the output of the regulator that are not seen on the input.

Load interaction caused this voltage "sag" - actually a voltage regulator misbehaving.

Load interaction is most common with tap-switching, magnetic synthesizer, or Ferro-resonant regulators. Unless the application has been approved by the manufacturer of the sensitive equipment that contains pulsing loads, such regulators should not be used.

Response to Severe Sag or Voltage Outage: Some regulators that contain electronic switching components will shut down or restart when they experience a severe voltage sag or outage. As a result, these devices could convert a short (1 cycle) disturbance into a several cycle outage. The sensitive load, which might ride-through a 1 cycle outage, will see the longer outage and shut down.

The Need for Voltage Regulation

The use of Voltage Regulators was widespread in the 1970's. At that time, most sensitive equipment was powered by Linear Power Supplies, that required a tightly regulated input voltage. In addition, electronic process equipment either had no internal voltage regulation, or very simple regulation that was affected by utility voltage changes. During this time, external voltage regulation often made a large improvement in system reliability and uptime.

However, in today's electronic world, the requirement for voltage regulation has dropped greatly, due to several factors. Linear Power Supplies have been largely replaced with Switched Mode Power Supplies (SMPS). These SMPS have a much higher dynamic range than the Linear Supplies, and can provide a regulated output voltage over a much greater input voltage range. Modern industrial process equipment also contains substantial voltage regulation through the use of phase-controlled SCR's, switching power supplies, etc. As a result, the need for external voltage regulation has been eliminated in most areas. Only the most unstable electrical systems (such as found in developing countries), or older equipment designs have a need for voltage regulation.

[Anonymous user]

KVAr, let me ask you a question. I've read on your profile that you used to fix UPSes, so do you know if the APC SmartUPS series use thyristors or relays to switch between one transformer output tap to another?

thanks!

[Anonymous user]

Thanks KVAr and Dirtfoot for the answers.

I still think voltage regulation is a waste for modern power supplies, at least for most part. And if we are talking about those with active PFC it's clear voltage regulation is something of the past. Power supplies with active PFC are becoming more and more common these days.

I think voltage regulation relays make the power supply work twice as much. The response time of a relay is about 5 to 8ms. When for exemple a fast voltage fluctuation occurs, let's say there is a 10 volt fluctuation, so until the relay "thinks" it has to change its state, this 10 volts fluctuation has gone through the power supply and it corrected it, since the fluctuation is much faster then 5ms, so then the relay reacts and again, the power supply has to adapt itself to the "new" voltage fluctuation caused by the AVR. Voltage fluctuation are much, much faster then 5ms on correcting these anomalies. Okay, you can say the fluctuation lasts longer then 5ms, sometimes for hours, but I've never seen fluctuation higher or lower then 30% off the nominal 120V or 220V as it is in my country. That does not justifies the use of an AVR in my opinion. I think most americans do not have any kind of voltage regulations in their homes for their PCs and yet they don't face any problem and most of them use regular power strips, sometimes with no kind of protection inside.

You also said something about changing the sensitivity of the UPS. As far as I know, you can only change the sensitivity of the battery transfer point, the lowest and the highest point and it has no effect on the internal logics of the AVR itself.

What I really think is, with Active PFC power supplies and many others being able to work from 100-240V, and there is still a percentage of tolerance in this number, manufactures like APC should make available different choices for the consumers. They should release in the market models without AVR, letting the customer choose if they want it or not or make available an option on the Powerchute software that would allows the user to turn off the voltage regulation and choose a broader lower and higher transfer point to the battery, so when there is big fluctuation in the power line the battery is not used without a real necessity. Let's say, an option to set the UPS to not go on battery until a 30% threshold is reached, would be nice! I'd buy an UPS like this as long as the price is not too high. 🙂

So once again, thank you all for the answers!

[Anonymous user · Accepted solution]

rau wrote:
KVAr, let me ask you a question. I've read on your profile that you used to fix UPSes, so do you know if the APC SmartUPS series use thyristors or relays to switch between one transformer output tap to another?
thanks!

Please, read these statements below and tell me what your opinion is:

+"Myth #4. VOLTAGE-REGULATING TRANSFORMERS ARE HELPFUL WITH COMPUTERS.+

+Most modern desktop computers use switch-mode power supplies rather than older style, linear designs. A switch-mode power supply draws from the AC power line only as much energy as it requires to maintain its output power. In this sense it responds spontaneously to voltage fluctuations. If the line voltage drops, the power supply draws current for a longer period, until it replenishes the energy it put out since the previous cycle of the power wave. Because of this natural ability to accommodate varying source voltages, a switch-mode power supply gain no benefit from a voltage regulating transformer. However, switch-mode power supplies are more sensitive to source impedance than source voltage, and the *increased impedance inserted into the line by the transformer may actually hinder the power supply by restricting the current available*.+

Yep. Switching power supply is constantly monitoring the secondary voltage and feeds back to primary side PWM control. Although, I believe operating on low line voltage for a long time is harmful if the power supply is heavily loaded as the I^2R will increase internally. This is particularly so for consumer grade PSUs as they're often built to less than specs. There are many consumer grade PSUs rated at say 500W, but will fry in less than 24 hours if used at spec maximum output. Slight over voltage, to the extent that doesn't exceed the specs of primary bulk storage capacitor is relatively harmless in my opinion.

+A tap-switching, voltage-regulating transformer may also introduce noise if the tap switch hunts back and forth between adjacent output taps. Computer switch-mode power supplies often have a wider tolerance for input voltage than do regulating transformers themselves. Thus the primary benefit of a voltage regulating transformer is its leakage inductance, which is much less than that of an isolation transformer, but the regulator introduces offsetting disadvantages. Moreover, the transformer's promary function, voltage regulation, offers no material benefits."+

The Smart-UPS is something that was available in the 90s and it hasn't changed much today with an addition of few computer interface related functions. Power supplies from 90s were almost always dual range (that must be selected with a switch) for 90-130v and 180-260v and exceeding 130v can push the primary bulk storage capacitors close to the limit and below 90 can cause it to stall (not able to maintain spec output) under full load.

I think newer UPSs are designed with this in mind as there are many newer(consumer units) with boost function only and no drop function. However, keep in mind that server/professional grade UPSs are often used to power pre-existing infrastructure with less tolerance and must be accommodating of them as well. While switching regulators do well with gradual voltage swing, notches and spikes can still cause crashes and errors.

Modern electronic fluorescent ballasts are just like computer UPS and can handle 120v to 277v without any switch and will maintain the same output regardless of input voltage, however they will still flicker when a big machine is started on the same circuit and cause a notch.

If the voltage regulator is hunting back and forth on a Smart-UPS you can always lower the sensitivity (I believe it alters the dV/dT sensitivity profile) and change the upper and lower bounds through serial commands with PowerChute or directly.

And please read what this Voltage Regulator manufaturer says about the implications of using one:

+Voltage Regulator Problems - Full Text at: [http://www.teal.com/products/App%20note%20AN-2.htm+

It's still necessary to keep switching power supply from receiving large spikes and notches and voltage regulator design to alleviate these problems can work,but better solution is to isolate the cause, for example, don't put laser printers and computers on same circuit as the pulsating load used by laser printer fuser controller can cause errors.

rau wrote:
KVAr, let me ask you a question. I've read on your profile that you used to fix UPSes, so do you know if the APC SmartUPS series use thyristors or relays to switch between one transformer output tap to another?
thanks!Tap switching and transfer is all done through relays. There are quite a few relays inside each Smart-UPS unit.