Asymmetrical Information

Perhaps you should rephrase?

The frequency of AC they take (60hz) has nothing much to do with the amount of power the converter can supply (50 Watts).

I don't know how well 60hz electronics will work with a 50hz source; they might work just fine (since they all just convert it to DC anyway). I've never had to find out.

What are the input and output specs on the bricks and your converter? We'd be looking for the volatge and current ratings here, as well as whether it's AC or DC. Usually, this information is located on the product sticker. For example, I'm looking at a wall-wart that specifies "INPUT: 120V AC 60Hz 22W OUTPUT: 15V DC 1000mA."

To more directly answer your question, it may be possible to use a 60Hz brick with a 50W converter. To know the answer for your specific case requires the above information.

input 120v 60hz 23w
ouput 12vdc 1000mA

and

input 120v 60hz 72mA
output 15v 500mA

I think you'll be fine. Brick #1 is drawing 23 watts (says so right on the sticker!) and Brick #2 is drawing about 9 (voltage * current). Total power demand is well within spec. Don't know exactly how they'll convert if they're set up to take a 60hz input but get 50hz instead. But I don't think you'll blow anything up.

The manufacturer can answer this question easily, and it's probably on the website. :p

you should post all three stickers (bricks as well as converters...)

it does seem to be fine, but more information would help

It's not a good idea to use a transformer designed for 60 Hz with a 50 Hz source. This has nothing to do with the wattage numbers.

Last time I wrote about this, I threatened to use the word "impedance" when I said that this shouldn't be done, and now it looks like I'll have to.

Impedance is AC resistance. But the reason we don't use the same word is that resistance is a number, but properly impedance is a curve. It means that the resistance of the device is different at different frequencies.

Transformers have to be designed for the frequency they're operated at. All coils are filters, and there's an optimum frequency where the impedance is at its lowest. A good design for a transformer will minimize its impedance at the frequency it's expected to operate at.

At other frequencies, the impedance will be higher, and the transformer will dissipate more heat. If the transformer is designed to handle that without smoking, that's fine too. But if it isn't, then trying to operate the transformer at that frequency is risky.

Some power bricks say that they're rated "50-60 Hz". What that means is that they've tuned the impedance for somewhere between the two, and they've made sure that the transformer doesn't overheat when operating at either frequency.

If a power brick is specified as "60 Hz" exclusively, then it means that no attempt was made to make it work properly at 50 Hz. Probably the impedance is optimized for 60 Hz, and thus when used with 50 Hz it's going to get warmer, and no attempt was made in the design to deal with excess heat.

There's a degree of overdesign, and such a transformer won't go up in a puff of smoke within seconds of being plugged into 50 Hz power, but it's not really safe.

It's not just that the transformer itself could smoke. It could cause a fire.

It's your life, your apartment, your power bricks, and your call. but I woud not plug a brick that says "60 Hz" into the output of a 50 Hz stepdown transformer.

Jane: What does this "50W converter" output? Does it say something like "120v 60Hz 50W" on the output? If so, the earlier poster's calcualations answer the question -- you can plug both bricks into the converter simultaneously.

But if it outputs 50Hz frequency (which is a separate figure from the 50W power), then heed Den Beste's explanation: you could fry one or both of those bricks, possibly quietly, possibly accompanied by a release of magic blue smoke, and rarely but still-possible, fire.

If you want to risk it anyway, connect the bricks and their associated device, then feel each brick every couple minutes over the course of an hour or more. Some older brick models may get uncomfortably hot to the touch when operating normally, and most newer brick models a little less-so. (It will be very helpful if you know, from previous experience, about how hot these things get after being plugged in for a long time.) If you decide that you CAN safely use them this way, unplug them whenever you're not home AND actively using them. Ordinarily, this is not good practice since thermal-cycling is the primary cause of eventual electronics failures, but in this case make an exception.

Alternately, if the brick during operation gets so hot that you cannot comfortably hold onto it without feeling burned (55-60C+), the transformer inside is likely heading north of 70C and may end its life prematurely.

Jane, if you are considering taking these power supplies to the U.K. ----DON'T. Acoording to this site http://kropla.com/electric2.htm they use an input voltage of 230vac @ 50 Hz. Since you said your bricks were rated at 120vac at 60hz, they might handle the 50hz with no problem, but the 230vac input voltage might cause maximum smoke. I do want you to return to the US with no injuries, resign yourself to buying an Euro rated brick.

We have prior information from a previous thread. Jane is using a stepdown transformer which is converting 240V at 50 Hz to 120V at 50 Hz.

Bill, your warning is unneeded. A 240V plug is physically incompatible with a 120V plug. They were designed that way deliberately so as to prevent someone from plugging a 120V device into a 240V socket by mistake.

By the way, with respect to anony-mouse's recommendation to test a brick over a period of an hour or so, you need to know that the brick has to be under normal load for that test to tell you anything.

If you plug the brick into the power strip but don't plug anything into the brick, it won't draw much power and won't heat up as much. The issue with impedance is one of efficiency: for every N milliwatts of power provided to the device by the brick, Q milliwatts of heat will be produced at 60 Hz, and Q+R milliwatts of heat at 50 Hz. The thing you're trying to find out is if R is too high an extra amount of heat to be safe.

I don't really agree with his suggestion to make that test, but at the very least I want to emphatically agree that you should not leave such bricks plugged in when you're not in the same room.

Ah, it's been a long time since I thought about this kind of stuff. I just remembered another failure mode.

One of the ways that an overheated transformer inside a brick can die is by shorting. The heat gets too great and the insulation inside the transformer breaks down, permitting adjacent windings to connect.

Effectively, what this means is that the number of windings gets smaller. If that happens on the primary side (i.e. the side which is connected to 120 V) then it means that the output voltage of the transformer will go up.

So instead of producing 12V output or 15V output, for a while it might rise to 25V or 30V or maybe even higher before the transformer dies completely.

Which means it will also destroy whatever it was supposed to be powering, quite possibly with a further and more spectacular release of anony-mouse's "magic blue smoke".

It really has been a long time. Actually, it's unlikely that the output will change. What would actually happen is that the regulator would pop. Sorry; ignore the previous post. (That's what you get when you ask engineers for a simple explanation...)

Well, geez, if you're worrying about the bricks catching fire, just drop them in a bucket of water before plugging them in.

Why is everybody looking at me like that?

Yes and no. 60HZ transformers can be used at 50 hz if the input voltage is reduced about 17%. Reducing the input voltage is not practical, however, if the transformer voltage is rated about +17 %, - whatever% it should be ok.

If the 50W converter is the one that is sold by Radio Shack, and other travel organizations, for use by travelers, then you should have no problem. I used such a converter for about 4 years to power a laptop that was not designed to run on 240v power, and both the laptop and the converter worked quite well. Mr. Den Beste has made some very good points about impedence and heat dissapation, and I would also comment that you may also hear a 50Hz hum from the unit, but that is no cause for worry.

I have nothing to say on this point, but it's great to hear from Steven! Hope you're feeling better.

A reader who never complained,

Not to argue with you, Steven, but small power transformers don't have enough Q to worry about setting the frequency "somewhere between the two".
The reason there are 60Hz only designs is that it is cheaper; a 50Hz xfmr is about 20% more iron and a bit more copper. I've never seen a 50Hz only design; if it works on 50, it'll work on 60 with no sweat.

Hi Jane. I am not an engineer but I have lived overseas a lot with North American appliances, so I'll just share a few personal experiences.

I have frequently plugged 120V 50hz DC adaptors into step-down transformers with no problems whatsoever. They have never overheated on me or otherwise burned out. Having said that, I would not recommend using them on a long-term basis. A better approach would be to buy a replacement DC adaptor from the manufacturer of your appliance that works with local electricity. Just order it as a spare part. Or you could get a generic-brand switchable one that provides several different levels of output, so one brick could be used with more than one appliance (one at a time, of course). I found one here in Switzerland at the local electronics superstore and it works flawlessly. Note: If you try to get a repacement DC adaptor from the manufacturer, be aware that appliances sometimes have different model names/numbers in different markets, so you may have to do a little bit of reserach.

Step-down transformers tend to work pretty well with electronic appliances, but any kind of internal clock or timing device may be thrown off (e.g. clock radio, etc.) I'm not sure why this happens, but it does. However, I would never use a transformer with heating appliances (hair dryers, toasters, coffee makers, etc) or motorized appliances (blenders, food processors, etc). It's probably better to just buy new kitchen appliances and store the old ones until you go back to the States.

A qualification: there are transformers that are specifically designed for heating appliances like hairdryers, but they should NEVER be used with anything electronic or motorized.

Finally, just because a transformer is rated at 1000 Watts doesn't mean you can plug in appliances whose ratings add up to 995 Watts and expect everthing to be fine. Leave plenty or headroom or the transformer may burn out. A good rule of thumb is that the transformer should be rated for twice the power requirements of your appliance, e.g. if your stereo draws 50 Watts you should have at least a 100 Watt transformer.

Good luck...

One more thing: I think the timing problem is caused by the 50/60Hz mismatch, and is probably the biggest thing to look out for. My old computer worked fine with a transformer but the dates stored with the files were always wrong.

Perhaps Steven or someone else can explain why this happens.

Duh! Because a second is defined as 60 cycles when designing for equipment that operates on 60 Hz power, and plugging that equipment into 50 Hz power means that the equipment only receives 50 cycles in a second instead of 60. So it needs 10 more cycles to think that a second has gone by, but then, 1.2 seconds have gone by (1 second plus 10/50 of a second).

A 240V plug is physically incompatible with a 120V plug. They were designed that way deliberately so as to prevent someone from plugging a 120V device into a 240V socket by mistake.

Tell that to my boyfriend, who managed to do that very thing. It was on Christmas Eve. I had bought a new power cord for my computer, and he plugged it in without switching the power supply from 120 to 240V. Not all power supplies can do that, but this one can. His excuse was that the monitor had needed no such switching, which is true.

Yes, I admit this was a special circumstance.

Fortunately it only ruined the power supply, and not the computer. Unfortunately a) Australia usually shuts down between Christmas and New Years; we were lucky to find a computer company that was open for a few days in the middle of that week, and b) it took a couple of tries to find a new power supply that would physically fit into the computer (despite measuring the old one beforehand).

I bought a new (compact) stereo in Oz, one that would work on a variety of power systems (took a while to find, and cost far more than it would have in the US -- it's a Sharp, if you decide to go that route; they had several models like that). It's still churning (more or less) merrily along beside me now that I'm back in the US. I bought that because I was worried that my CD/tape player wouldn't work on 50Hz, although several people told me they'd bought equipment in the US which worked fine on a transformer. Maybe they just got lucky.

Well Jane, you've demonstrated an unpublished "rule"
(rules as in "The Rules" by Ellen Fein and Sherrie Schneider).

One way for a pretty girl to get a guy to "make a move" is for her to ask for help (and to be obviously in need of same).

You go, girl!

Christ, you even got the Bob Dylan of blogging, the enigmatic Stephen den Beste to proffer advice.

JKohn, Rex: Sorry, but no.

Computers don't use AC to set their clocks; the AC frequency is nowhere near accurate enough to do so anyway (nominal 60hz varies a bit; I don't recall how much, but enough that you'd be stark raving mad to use it for a clock!).

Computer clocks, like battery powered clocks, and really all non-radio-controlled or atomic clocks (to my knowledge, unless there's some technology out there that is neither quartz nor radio-signal, not atomic decay, for electronic clocks), use a quartz crystal as their basic timing device (there may, as I understand it, also be frequency divider or multiplier hardware, but the base clock signal they derive from is a quartz crystal dingus).

Just as importantly, the AC is converted to DC before it gets anywhere near your computer's motherboard and circuits.

(If your computer's clock keeps getting off, it probably just has a bad clock. PC clocks are often just not very accurate.

You can tell most OSes to simply set their clocks via the internet directly, and on any non-stone-age OS you can get external software to do so.)

I am a EE, but I'll try to say this in English. Electronic power supplies now come in two main types, linear (or unregulated) with a transformer at the input, and switching. If the power brick for an appliance has an AC output, it's definitely the transformer-first type. If it's DC out, it's probably switching, but could be the transformer-first type.

With a transformer, the main issue is that the minimum amount of iron and copper required is inversely proportional to the frequency. Energy is transferred across the transformer in the magnetic field. The field grows, collapses, and reverses with the input current, and the changing field induces current and voltage in the output coil. With 60Hz, there are 120 magnetic pulses a second, with 50 Hz just 100. That means that the peak magnetic field has to be higher by 20% (a ratio of 120/100) in the 50Hz transformer.

If the transformer was overbuilt by 20% or more, the way most things in electronics are, it wouldn't be a problem. However, it seems like the transformer requirements can be calculated quite precisely, and most consumer products have custom-built transformers that might be just barely big enough. That is, IF your device pulls the full design power, it's likely that there isn't enough iron in the transformer to support the required magnetic field. (Iron only holds so much magnetic field; when you reach or exceed the maximum density, it's called "saturation.")

This magnetic field also regulates the input current on the primary (120V) side. When the iron saturates, the input current can increase by a lot. It's not just 20% over, it can be several times the design current. This generates heat, maybe a lot of heat that can destroy the transformer in a few minutes.
OTOH, the power drawn by an electronic device will vary depending on many things (how you are using it, temperature, variations between transistors and IC's as they are built at the factory.) The transformer has to be sized to supply the possible maximum, but it's quite likely that you'll never require any more than 80%. So John D'Oh has been able to run 60 Hz devices off 50 Hz 120VAC. But you might be unlucky, so do this only if really necessary, and keep checking the transformer temperature.

There are lots of 50/60 Hz rated power supplies out there. They just add 20% to the iron and copper. Of course, that also adds 20% to the weight and cost of the transformer. 50/60 Hz "international" transformers have two primary coils that you wire in parallel for 120V and in series for 240V.

Incidentally, there are some 50Hz transformers that get into trouble at 60Hz. This is a different issue, related to hysteresis and eddy currrents. Hysteresis is the tendency of iron to resist changing magnetic fields. Eddy currents are currents that are induced in the iron, instead of the output coils. Both of these steal power from the input and heat up the iron with it, and higher frequencies obviously increase these losses. Transformer manufacturers try to reduce these effects by using a special iron formulation and by dividing the iron into a stack of thin plates. Sometimes in products designed for 50Hz, they'll do the absolute minimum to reduce eddy and it will overheat at 60Hz.

With a Switching power supply, the input stage rectifies the AC directly to create high voltage unregulated DC. Then the DC current is switched with transistors to create something like high-frequncy AC (with frequencies in the KHz). A much smaller transformer then works to step down the voltage, and the output from the secondary is rectified and filtered to create a DC output. Switchers CAN be much more forgiving. There are "international" switchers where the only thing you have to change is the power cord - the switcher will handle 100 - 250 VAC at 50-60Hz. But these aren't consumer products where every cent of cost has been shaved.

I'd expect that most of the time an American consumer-quality switcher would work on 50Hz, as long as the voltage has been stepped down to about 120. One limitation is the filter caps on the input side. These store charge and energy between the AC half-cycles, and might run short at a lower frequency = longer cycle time. But it's not so usual for these parts to be sized to the absolute minimun required. For one thing, it's hard to control the capacitance better than 20%, so the cheapest way to get a capacitance of at least 80 (say) is to oversize it: 100 +/-20% (80-140). Or even 100 +50%/-20% (80-150). Second, the aluminum electrolytics often used for big capacitors dry out and lose capacitance as they age, so they have to oversize it even more to compensate. So most of the time you've got plenty of extra capacitance. And if you don't, instead of a burned-out supply you should just get low voltage at the output.

However, don't count on it. There might be other failure modes I haven't thought of. And it's possible that a low voltage will cause something to self destruct inside the expensive electronic gismo rather than the cheap brick.

Ralph: Most laptops use a switching power supply; see markm's explanation.

Sigivald: Good explanation on the computer side, but some analog clocks can and do run off the AC frequency. It may not be perfect, but it is close enough for general timekeeping, and what's more, it is self-enforcing because the power supply is a grid. (Any generator or motor drifting slightly out of line will be very quickly brought in-line; and if you try to bring a generator online without being synched to the frequency and phase, you will mechanically destroy it.)

You don't often see consumer-level synchronous AC analog clocks anymore, but they're still common in commercial installations. They use a synchronous AC motor which rotates at a multiple of the AC power supply's input frequency. For obvious reasons, if you change that input frequency, the clock runs at the wrong speed.

As Sigivald said, computer clocks don't depend on the AC frequency. There's a little crystal oscillating at MHz (or possibly GHz nowadays?).

Just plain clocks might work by counting the AC cycles. The cheapest electrical ones with moving parts often had just a synchronous AC motor. That's a permanent magnet on a spinning shaft and a coil of wire next to it - when the current reverses, the magnetic field of the coil reverses, and the motor must turn 180 degrees to follow it. (Getting it started spinning is a lot more complicated - actually, I don't understand that part - but once it's spinning at close to the right rate, it's forced to lock onto the AC frequency.) So for an American clock, the second hand is a 60-1 gear down from the motor, the minute hand is another 60-1 gear down, and the hour hand is 12-1. The power companies keep pretty good time (so they don't get into trouble trying to sync dozens of generators together), so if your clock is on the right power frequency and loses enough time to notice, you probably had a power failure. A European clock would change the first set of gears to 50-1. And an American clock in England would lose 10 seconds a minute, which makes it pretty useless.

Once, lots of other equipment would have similar issues. Tape players or record players (if anyone remembers the vinyl albums) might also have motors that follow the AC frequency, which would give you a definitely out-of-tune and draggy playback. Except, who has tape players that work only off line power anymore? If it's dual powered (AC or batteries), it's got to have it's own clock.

edit: I should have said "any synchronous generator or motor", i.e., the only kind whose application directly interacts with the grid frequency.

markm: According to Wikipedia, a single-phase synch motor may use a small integrated induction motor for startup, or a lightweight rotor with a unidirectional lock to prevent the magnets from being drawn in the "wrong" direction at startup.

http://en.wikipedia.org/wiki/Electric_motor

Computers don't use AC to set their clocks; the AC frequency is nowhere near accurate enough to do so anyway (nominal 60hz varies a bit; I don't recall how much, but enough that you'd be stark raving mad to use it for a clock!).

Others have commented, but I'll chime in. You are correct to say that the frequency varies a bit, but the power company keeps track of the variance, and makes sure that it averages out to 60 hz over time. That's why analog clocks don't ever need resetting (except for power failures.) . At any point in time, they might be off a bit, so they aren't accurate to fractions of a second, but those errors don't add. Ten years later, the clock will be right, plus or minus a fraction of a second.
My father (who happened to design the watthour standard) explained this to me when I was young, and I was fascinated by it.

(As an aside - I cannot say say how happy I am to see den Beste posting)

Computers typically use a crystal oscillator, which run on direct current. Computer clocks don't use a count of AC cycles, which wouldn't make sense in any event because they're on the motherboard (typically) which only sees DC. And typically the oscillator frequency precision is dictated by the size and construction (and temperature; let's not forget that) of the components in the oscillator circuit.

I'm willing to defer to MarkM's comments, since I'm not actually an EE.

There was one thing he said that I did want to comment on: 50/60 Hz "international" transformers have two primary coils that you wire in parallel for 120V and in series for 240V.

Doing that requires a physical switch, and I've seen them. But that also requires a customer who can read, and if Mark is an EE then he knows as well as I do how stupid some customers can be.

Which is why the brick for my laptop works at both 120V and 240V without any switch. The big clue is that it draws the same current at both voltages, which thus means it's using twice as much power at 240V. Where does the excess go?

I think that they're using a fixed step-down ratio (probably 5:1, because the output is 19V) and they've got a big honking regulator with a monster heatsink in there. At 240V the regulator chews up about half the power being drawn from the wall.

Guess that's what I get for reading the thread backward: already been amply covered. In case anyone's interested, quartz-oscillator clocks can be made to be more accurate through calibration and temperature control. But that's way off-topic.

I'm not sure what is meant by '"international" switchers where the only thing you have to change is the power cord' in markm's comment, but I just wanted to add that just about any PC power supply with Active PFC seems to be able to handle switching from one AC standard to another without any help. I'm of the opinion that you'd describe these things as "consumer" devices too. :)

Anonymous: When I lived in Paris (87-90), laptops did not use switching powersupplies - they were large, heavy transformer bricks.

Also, the power companies _count_ the number of cycles each day, and each year, to make sure that synchronous clocks stay in snych. The generators that make the power are LOCKED into synchronization - generators that lose synchronization with the grid can suffer severe mechanical damage. Interestingly enough, if you have a clock that uses a battery as a backup powersupply, then it does NOT count the cycles, and is subject to drift.

Finally, some of us here are probably old enough to remember that mainframe computers in the 1960s did use line frequency as a timing source. You had to pay more money to IBM for a high resolution timer, if you wanted to, say, operate a time-sharing system, and charge clients for each small amount of computer time used.

I am not an electrical engineer but I do play one on television.

I can't help you, Ms. Galt, but it is a grand thing to read comments by Steven Den Beste again. You are missed, sir.

Which is why the brick for my laptop works at both 120V and 240V without any switch. The big clue is that it draws the same current at both voltages, which thus means it's using twice as much power at 240V. Where does the excess go?

Uh...are you completely certain about that? Is this an old unit or some unusual configuration, or is it just badly labeled? (e.g. "120-240VAC 1.2A max")

Just wondering for curiosity's sake, because even a 19V, 3A linear-regulated supply could easily dissipate ~40W of heat between the regulator IC and transformer losses, and of course the outside of the brick could not be permitted to exceed ~45C without becoming a consumer safety hazard.

Not an easy combination for a laptop brick, short of using active cooling or making the brick unreasonably large. AFAIK most autodetecing power supplies are switch-mode and use a detection circuit to set the switcher's operation according to that input voltage.

Comments are Closed.