How to determine wattage of Amiga PSU?

[Amiga1992]

I am very confused as to how to determine how many watts my Amiga power supplies consume.

I Am trying to get voltage converters to get these things running in the US and I am utterly confused.

Somewhere online I read you have to multiply the voltage (220V) by the amperage of the unit. This leads to some ridiculous values. And how to deal when Amiga supplies have two different amperages for each voltage line?

Is this really the method to calculate how many watts the units will consume?

I also have a C64 power supply rated at 13A, is it REALLY 220V*13A??? That's ridiculous.

[ajk]

What exactly does it say on the label..? Wattage is indeed volts times amps, but you have to look closely what values you use. My A600 power supply for example says:

INPUT: 230V 300mA

So that's 230V x 0.3A = 69W or so

It also says:

OUTPUT: +5V, 3A +12V, 500mA -12V, 100mA

So those amount to 15W, 6W and 1.2W respectively, or about 22W total. Not great efficiency if it really draws all of the 69 watts from the grid.

The 13A value could be for a fuse, meaning that if the PSU is shorted or something it will try to draw a huge current and the fuse will blow.

[FOL]

Yep, 13Amp is the UK standard for rating fuses.
Normally, 3Amp, 5Amp and 13Amp.

[Amiga1992]

Originally Posted by ajk

So those amount to 15W, 6W and 1.2W respectively, or about 22W total. Not great efficiency if it really draws all of the 69 watts from the grid.

But which one really matters for getting a voltage transformer?

[prowler]

Originally Posted by Akira

But which one really matters for getting a voltage transformer?

In the example given by ajk, it is the mains input power consumption (230V x 300mA) = 69W at full rated ouput load, i.e. ((5V x 3A) + (12V x 500mA) + (12V x 100mA)) = 22.2W, which provides the answer you require.

As ajk mentioned, these figures imply a PSU efficiency of less than 33%.

[Amiga1992]

Thanks for that guys.
Now every website recommends getting DOUBLE of what you use, some times TRIPLE. It seems a bit overkill to me, what's the fact about that?

[Hewitson]

I personally think you're going about this the wrong way. You'd be much better off buying a US power supply.

Originally Posted by prowler

As ajk mentioned, these figures imply a PSU efficiency of less than 33%.

Wow... How could anyone have thought that was acceptable?

[prowler]

Originally Posted by Akira

Thanks for that guys.
Now every website recommends getting DOUBLE of what you use, some times TRIPLE. It seems a bit overkill to me, what's the fact about that?

I can only imagine that recommendation is to cover the initial surge when you first power on your PSU working at very near its full rated load, caused by all the smoothing capacitors charging up.

This is a very brief surge only, of course, so there is really no need to specify anything much greater than the full rated load condition, unless your voltage transformer is of cheap, East European origin.

Originally Posted by Hewitson

Wow... How could anyone have thought that was acceptable?

Only because transformers are notoriously inefficient.

[Amiga1992]

Originally Posted by Hewitson

I personally think you're going about this the wrong way. You'd be much better off buying a US power supply.

You can't do that to a 1084 monitor.

Anyway, what about the CD32 power supply? The load it says on input (350mA) is smaller than what it says on each line (5v=2.2A, 12V=500mA)

[ajk]

@Akira

That sounds like it's in the same range as that A600 one (which draws a maximum current of 300mA at 230V).

Transformers of any kind trade voltage for current, or the other way around. Since the Amiga only really needs a lot of current at 5V (a low voltage), it means that you don't need much current at 230V (a much higher voltage).

For the purposes of power, 1A at 100V is equivalent to 100A at 1V, both being 100W. Electronics tend to need a low voltage, which means the current will be relatively high.

In practice you can't make that transformation happen without losses, which is why the Amiga PSU draws about 70W even though the Amiga can only use a maximum of 22W or so. Modern switch mode power supplies should run at 80% or more efficiency under the right circumstances.

Also, if the power rating of a transformer is 70W, it doesn't mean that it's going to take 70W all the time. That's just the maximum that will be drawn if the load (the Amiga or whatever) needs it all. Still, you should size your converter so that it can cover even the full power use.

As prowler says, they probably recommend twice or even three times the rated power because many devices can draw higher power for short periods of time, even if the average power use is much lower. Also the converter probably runs cooler and lasts longer if it doesn't have to work at or near its maximum capacity all the time. You probably don't need to triple the intended wattage, but it really depends on the quality of the converter.

[Amiga1992]

Thank you very much guys, you have explained it all very perfectly.

I currently have an LCD TV rated at 179W connected to a 300W transformer, do you think it'd be safe to add the CD32 power supply too? I'm not using this setup all the time, usually it's just the TV on the 300W transformer (and i turn everything off with each use).

This is all a temporary solution while I manage to find time and money to make a picoPSU power supply for that one and the other to Amigas that still have a 220V old brick!

[demolition]

For reference, my A600 uses about 25W on the 230V side of the PSU, and this is with one of the heavy transformer PSUs, so that rating of ~70W is only to provide proper startup. I wouldn't worry about hooking your CD32 up to the same transformer as the TV, since the current consumption is tiny in relation.

Regarding a 1084S, you should probably verify that it will work with 60Hz mains. A transformer cannot do anything about that.

[Amiga1992]

I'm not sure. Even the Amiga power supplies say "50Hz" only on the brick but they work.

I have had experience before using a 110V 60 Hz 1084 with a transformer stepping it down from 220V, and it worked. I am hoping the opposite will prove true. I can't do anything right now but to wait and see what happens.

Thanks everyone for being so helpful.

[ajk]

The transformer in the PSU will at first change the AC (sine wave shaped) mains voltage to a lower voltage, but it still remains AC. After that there will be a diode bridge which changes the voltage into a pulsating waveform (all the negative pulses are inverted). That is then filtered through a capacitor (plus maybe further components like regulators) to get a smooth DC voltage.



(image not mine)

Transformers will be designed with a target frequency in mind, so running the PSU at a different frequency will have some impact on the efficiency and heat generation. You can probably get away with it but it's hard to say how each particular device copes.

The Amiga will only care about having the correct DC voltages. In my understanding the 1084 internally works also with DC (it's just got the PSU built in) but I'm not 100% sure on that.

[Toni Wilen]

1084 (and probably all 1980s+ solid state TVs) have internal switch mode power supply, SMPSs normally don't care about mains frequency, first stage of switch mode power supply is DC conversion (diode bridge).

[Amiga1992]

Well the 1084 seems to be working fine!



I had it on for at least an hour or more and the transformer is not ridiculously hot or something like that, it's barely lukewarm. It all seems to operate fine!

The C64 will get problems with the clock signal I guess (because I am feeding it 60Hz instead of 50Hz) but as someone told me in IRC, "who cares":P

Amigas should work just fine!

[pandy71]

Originally Posted by prowler

Only because transformers are notoriously inefficient.

In fact transformers are very efficient (for high power transformer efficiency can be above 95% - http://en.wikipedia.org/wiki/Transformer#Energy_losses )

Mentioned 230 0.3A can be rating for internal fuse - frequently there is fuse integrated with transformer (primary winding) that prevent fire in worst case scenario - primary wounding have fuse which can prevent large damage by simply interrupt current flow, sometimes this fuse is thermal type (http://en.wikipedia.org/wiki/Thermal_cutoff) sometimes ordinary resistor like (single use - transformer need to be replaced or repaired).

[prowler]

Originally Posted by pandy71

In fact transformers are very efficient (for high power transformer efficiency can be above 95% - http://en.wikipedia.org/wiki/Transformer#Energy_losses )

Power distribution transformers handle massive throughput, so it is worth spending a significant amount of money to minimize losses. The design of transformers used in consumer electrical goods is a much less exact science, but things are improving since the introduction of energy ratings...

[Hewitson]

I believe it's the regulators that are inefficient rather than the transformers.

[Toni Wilen]

Yeah, linear regulators are very inefficient. Transformers unfortunately need to be very big and very heavy if frequency is low and power capacity needs to be high.

High-end PC power supplies also have >90% efficiency.