Transformerless Power Suppply

[T.Jackson]

Short answer is because it's much safer. But there are also other advantages
as I will spell out here...

1. Cleaner power, typically around 5mVPP of ripple - (AC component) - with the usage of sufficiently sized reservoir caps. General rule of thumb: 10uF for every 5mA of current. So, at 1 Amp you want at least 2200uF.

2. Well designed transformers are extremely efficient. Typically 95% or better.

3. Power transformers are reasonably immune to interference.

4. Isolation transformers are critical for the novice for when it comes to taking
high voltage measurements on an oscilloscope. In fact, no one should really take measurements without using it.

Trent Jackson

[Melanie]

I agree that it's better for hobbyists to include a transformer, but I do disagree with some things you write...

A Transformer will RARELY hit 95% efficiency. The reality is a figure somewhere between ZERO (yes ZERO percent efficient!!!) and somewhere approaching 90%. Let me explain...

The higher the frequency, the greater the transfer ratio between the Primary and Secondary. Hence, the most efficient are Switching Power Supplies which usually run at a few tens of kHz. However at the frequencies of the incomming AC Mains Supply (ie 45-65Hz depending where on the planet you happen to be), the efficiency is governed by the amount of iron (and quality of iron) you can shove into your transformer. Generally, the more iron, the greater the transfer ratio which conversely applies the law that the smaller the transformer the more inefficient it will be.

Now for this ZERO percent efficiency (100% waste of energy) bit...

The efficiency of a Transformer is simply calculated as Power Out divided by Power In. Actually we can reduce that to just Current In viz Current Out with regard to the turns ratio. So in a 100% efficient Transformer, 1A at 12V OUTPUT should equal 50mA at 240V INPUT as there's a 20:1 reduction ratio. Sadly, take any Transformer out of your parts box, put a series Ammeter with the Primary and plug it in completely OFF-LOAD. You will have a standing current... ie you are sucking energy in, but not taking any out. All this is doing is heating up the core, and it's worse the smaller the Transformer becasue of the smaller amount of iron. So from this you can gather, that the less current we draw from a Transformer from it's optimum design point, the more inefficient it will be. If a Transformer is designed to be most efficient at say 100mA, then any current draw less than that is heading towards the ZERO efficiency point when it is completely off-load. Likewise if you exceed the optimum design point (which is usually just below the Magnetic Saturation Point of the core), you then risk driving the core into saturation and again your efficiency plumets as all you end up doing is pumping more current in which just heats up the core.

From the above, you can gather that a Transformer is really the root cause of global warming as before it was invented we didn't have the problem!

Notwithstanding that, it does give you galvanic isolation and as proof that countless hobbyists are alive today posting silly questions on internet forums that otherwise wouldn't be, is it's most commendable feature.

Finally, there is a place for everything... even the humble Transformerless design. For small current supplies, the size and cost beats the transformer every time. For a typical Washing Machine or Freezer or Cooker or Microwave or Boiler or Central Heating or Air Conditioning or ..., or... or (add your own domestic appliance to the list) controller, losing the transformer would save about 50 cents (and save some weight too), multiplied by a typical 300,000 production run for the design and you can do the math.

As for hobbyists that play with mains... well we all gotta die sometime (it's a hereditary trait), just some sooner than later...

[T.Jackson]

You will have a standing current... ie you are sucking energy in, but not taking any out. All this is doing is heating up the core, and it's worse the smaller the Transformer becasue of the smaller amount of iron. So from this you can gather, that the less current we draw from a Transformer from it's optimum design point, the more inefficient it will be. If a Transformer is designed to be most efficient at say 100mA, then any current draw less than that is heading towards the ZERO efficiency point when it is completely off-load. Likewise if you exceed the optimum design point (which is usually just below the Magnetic Saturation Point of the core), you then risk driving the core into saturation and again your efficiency plumets as all you end up doing is pumping more current in which just heats up the core.

I think you've have described the term "Eddy loss current" ?

Just having a quick flick through some of my old college notes, I have %95 written down as a typical rate of efficiency...

Trent Jackson

[T.Jackson]

Extract of taken from college notes 16/06/98

Transformers are very efficient components. Typically they boast efficiency levels of better than 95%. This means that only a mere 5% of the total power is lost. Consumed / wasted in the form of heat. Transformers work on the principle of self-induction, and because of this, there is almost perfect isolation between the primary & secondary windings. With this infinite impedance between the two windings, power transformers are deemed very safe.

By Faraday's law of magnetic induction, when an alternating current is applied to the primary windings, a voltage will be self-induced in the secondary. These windings must be close wound. Secondary power is introduced from the primary because of the changing magnetic field between the two.

When a voltage is applied to the primary a magnetic field is set up, and if the frequency is high enough the primary windings will begin to radiate its power to the secondary. The voltage that is produced on the secondary is directly proportional to the turns ratio.

In short, minus all the math, magnetism theory, that's how a transformer works.

Trent Jackson

[Melanie]

Time college lecturers got out a bit more and got dose of the real world.

In theory everything is lovely.

In practice I challenge you to take ANY transformer you've got, and plot Input Current & Output Current from off-load, thru to full rated load in say 5% steps. From that you can calculate and plot a graph of efficiency to prove my point. Enjoy the experience. Then take the efficiency graph and tell your college that life in the real world is far from theory - so far that it's not even close!

For example... let's try to prove the theory of the International Ampere (you know Amps... it's that unit of measurement all of us use every day of our lives... so really we should try to prove the definition before we use it)...

Quote from my University Notes "That constant current which when flowing down two straight parallel conductors of infinite length and negligible cross section, separated by a distance of one meter in a vacuum, produces a force between them of 2 × 10-7 newtons per meter of length."

You get full marks for reciting this like a Parrot in your exam... but let's try to prove it...

1. You go down to your local store and ask for two drums of cable of infinite length and negligible cross-section... you lug them home...

2. You borrow the space shuttle and unravel them in the (supposed) vaccum of space from planet Earth out towards infinity.

3. When you've reached infinity, you join the far ends together and come home.

4. Then you hook up the wire to your PSU with a DVM in series switched to the famous "Newtons per Metre Length" setting on the dial.

5. Take a second DVM in series and switch it to Amps.

6. When DVM No.1 shows 2x10-7 Newtons the other should be showing 1.0000 Amps (assuming it's in calibration).

Worked when I tried it...

[T.Jackson]

Unfair game here. Someone has a UNI degree...

Where's the referee?

I'm in the process of doing a degree in Applied Science In Information Technology. Only done two years of electronics in college. I'm only at trade level
with most of the theory.

Cheers

Trent Jackson