1
\$\begingroup\$

This question led to this other question, which in turn is now spawning the one you are going to read.

As much as I understand, AC is more efficient for long distance transmission and much more versatile because it allows to be transformed from one voltage to another. To have AC you obviously need it to periodically reverses direction (otherwise is just DC), and for everyday appliance the frequency at which this happens is 50 or 60 Hz mostly for historical reasons.

So once you plug a device this contains a simple rectifier that change the direction of one of the ripple so that the signal is more akin to the DC (usually) needed for the device to work.

Also, I've read that in the old times the supply signal was used as the frequency reference for devices that needed it.

So, all of this raises a question: what happens if you feed a (for example) 50Hz device with a 400Hz supply? Is there something implicit in the rectifier so that it can work only at specific frequencies? Or will it still work and just happen to produce a "more DC" supply (that is, with less pronounced ripples)?

\$\endgroup\$
  • \$\begingroup\$ Not all devices have DC rectifiers, lot of them use AC. \$\endgroup\$ – Marko Buršič Jun 16 '17 at 7:53
  • \$\begingroup\$ @MarkoBuršič yeah, I know, that's the reason for the "(usually)" in the question. But I was just trying to keep things simple, and as far as I know passengers mostly do not board a plane to plug a vacuum cleaner and start cleaning. \$\endgroup\$ – motoDrizzt Jun 16 '17 at 7:55
  • \$\begingroup\$ And you usually don't connect a laptop in the main circuit of the airplane. The seats have standard output voltage/frequency not the 400Hz. \$\endgroup\$ – Marko Buršič Jun 16 '17 at 8:01
2
\$\begingroup\$

Just to correct an assumption: -

As much as I understand, AC is more efficient for long distance transmission

No, it isn't as good as DC for long distance travel (skin and proximity effects) but, because you can use a transformer it becomes the only feasible practical option.

what happens if you feed a (for example) 50Hz device with a 400Hz supply? Is there something implicit in the rectifier so that it can work only at specific frequencies?

To a certain extent there is something about common garden diodes that tend to make them less suitable as the frequency rises and that is called the reverse recovery time: -

enter image description here

So, as the voltage waveform goes from forward conduction to reverse conduction, a diode won't immediately turn into a blocking component and a reverse current will flow.

For a 1N400x rectifier this is about 30 us (some manufacturers): -

enter image description here

See also my answer to this related question.

At 400 Hz, the period is 2.5 ms and half a period is obviously 1.25 ms. 30 us represents 2.4% of the half cycle so it's starting to become a problem. Answer: use better and faster diodes.

For devices that use a mains transformer, operating at higher frequencies means more eddy current induction in the laminates and potential overheating. Answer: use thinner laminates.

\$\endgroup\$
  • \$\begingroup\$ Is skin effect really significant at 50Hz? \$\endgroup\$ – Sclrx Jun 16 '17 at 9:05
  • 2
    \$\begingroup\$ @Sclrx yes, of course: en.wikipedia.org/wiki/Skin_effect and this: electrical4u.com/skin-effect-in-transmission-lines \$\endgroup\$ – Andy aka Jun 16 '17 at 9:09
  • 1
    \$\begingroup\$ @Sclrx not at wire sizes typically used inside portable/domestic appliances. \$\endgroup\$ – Peter Green Jun 16 '17 at 11:20
  • \$\begingroup\$ The reason I pojnted this out is because of the term "long distance transmission" and, skin effects are significant on long distance transmissions. \$\endgroup\$ – Andy aka Jun 16 '17 at 11:23
  • \$\begingroup\$ @PeterGreen true, but we are talking about power lines here, so it does matter \$\endgroup\$ – Sclrx Jun 16 '17 at 11:23
0
\$\begingroup\$

As with all things,

It depends

Old and simple AC-DC power supplies work with a bridge rectifierbridge rectifier, which hyperphysics (linked) explains much better than i could. (and if you did not know about it, this site is a wonderful resource for everything physics-related, including electronics)

The difference in frequencies (not even a full decade) is small enough to allow us to estimate that it would not change the behavior of any of the components use in the bridge.

So yes, a bridge rectifier would still be able to turn AC to DC if the AC frequency was 400Hz, but its efficiency would change.

For other power supplies, i can not answer (there are many different types, each would behave differently)

You mentioned using the 50/60Hz as a reference. a simple clock, for example, (assuming the logic circuit are able to handle the 400Hz input) , would count 8 times faster if it was designed to operate on 50Hz.

\$\endgroup\$
0
\$\begingroup\$

There are some bad things that could happen when giving a bridge rectifier 400Hz mains.Andy aka has covered slow diodes and extra heating on 400Hz .If your rectifier has a X cap rated at 250VAC across the AC terminals it might blow up because the expected ripple current will be 8 times the 50Hz rated value .This means that ESR based heating will go up by 64 .If this is not enough the dielectric insulation losses will probably go up by 8 times .X caps are common these days for EMC reasons .The Y caps which are also common for EMC reasons will also experience an 8 fold current increase .The Y caps may not be rated for this and things are not safe due to the earth leakage current going up 8 fold .Now any circuit inductance has 8 times the reactance which may result in less output or extra heating in line reactors .

\$\endgroup\$
  • \$\begingroup\$ Would the ripple current not reduce by a factor of eight because it is being topped up more often? I agree with the heating problems. It's a problem if there are high harmonics in power-factor correction equipment. \$\endgroup\$ – Transistor Jun 16 '17 at 11:11
  • \$\begingroup\$ Ripple current in the primary capacitors will remain similar in magnitude but at higher frequency. Ripple current in the input filter capacitors will increase linearly with frequency. \$\endgroup\$ – Peter Green Jun 16 '17 at 11:15
  • \$\begingroup\$ @Peter Green and Transistor .The X cap is connected across AC and will see the 400Hz and get more stress .The BULK DC bus cap gets charged more often and discharges more often .The bulk DC cap has a lower depth of discharge giving lower ripple Voltage but similiar ripple current .I am not too worried about the DC cap blowing up but are concerned about AC capacitance . \$\endgroup\$ – Autistic Jun 16 '17 at 11:24
0
\$\begingroup\$

So, all of this raises a question: what happens if you feed a (for example) 50Hz device with a 400Hz supply?

There are several issues.

  1. Diodes will have more losses to reverse recovery.
  2. Input filter capacitors will carry substantially more current (7 times more than at 60Hz).
  3. Wires will have a higher effective impedance due to inductance and skin effect.
  4. If there is power factor correction it may be unable to keep up with the input changes resulting in worse power factor and/or lower efficiency.

Many devices will have sufficient safety margins that they will work anyway but some devices could potentially go up in smoke.

\$\endgroup\$
  • \$\begingroup\$ With 8X faster ripple, the sag will be 88% smaller. Thus the high forward currents will only last 12% as long as at 60Hz. \$\endgroup\$ – analogsystemsrf Jun 16 '17 at 17:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.