1
\$\begingroup\$

As I'm looking at some resistive-load applications I'm wondering whether there are any rules that put an upper-limit on supply-matching efficiency. Because if a load isn't matched to the power supply in terms of voltage, current, and waveform (typically AC vs. DC) then we have to add all sorts of components, and depending on how mismatched the supply and the load are we may end up cooking off a lot of waste power to bring them into alignment. To what extent is this just for design simplicity, and to what extent is it unavoidable?

  1. Is there some minimum waste power cost involved in stepping voltage up or down? My understanding is that in the AC realm it is negligible even in practice with transformers, as it is in the DC realm with multipliers. But are these limited to cases where the input and output are multiples, and it gets more expensive in terms of waste power to hit a specific voltage that isn't an integer (AC) or binary (DC) multiple of the input?

  2. Current seems to be an efficiency killer. I.e., once current in a circuit becomes significant it seems like there's no way to control it without generating a lot of waste. What theoretical constraints are there are on current throttling efficiency, and on what variables do they depend?

\$\endgroup\$
  • \$\begingroup\$ Have you come across switching regulators? \$\endgroup\$ – Andy aka Feb 28 '16 at 10:09
  • \$\begingroup\$ @Andy - Yes, I was thinking there's a "digital" solution to the current problem, but if the supply current capacity is very high relative to the desired output isn't there a limit to how you can PWM to limit current before the waveform has too many gaps relative to any reasonable specification? \$\endgroup\$ – feetwet Feb 28 '16 at 15:42
4
\$\begingroup\$

There is no theoretical limit other than the laws of thermodynamics. Efficiency cannot be 100%.

Typically switching supplies designed to a reasonable cost have an efficiency of >80% over a range of outputs and inputs. Typically efficiency falls as the output load is reduced because there is some overhead in the supply, and it may fall a bit at the highest rated current as I^2R and other losses mount.

Transformers can be very low loss (especially if they are designed for low losses) but that is not necessarily true - in fact relatively recent State of California legislation basically outlaws the sale of many cheap transformer AC adapters because their losses are higher than switchmode types, especially at low loads. Some run warm to the touch even with no load. As an Engineer we have some control over this- by specifying the better (and more expensive) laminations the transformer can have dramatically lower core losses, but this is unsuitable for a product that must sell for a very low price.

In your earlier question you were asking about linear supplies, which generally (but not always) have a low efficiency. A perfect linear regulator supplying 1A at 1V from a 12V source has an efficiency of 1W/12W = 8.3%, which is much lower than 80%. On the other hand, a perfect LDO (linear regulator) supplying clean 5.0V from a noisy 5.6V source has an 89% efficiency, possibly better than a switching supply.

Current in itself is not a problem- supply currents in modern PC motherboards are probably in the 100A range (power of a CPU alone might be 120W+), and the supplies themselves do not require heat sinks.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ I quite like your answer, the average pc psu usually do not provide 100A though, only the high end units reach that current. I think that a theoretical limit exist though, and it is lower than 100%. I'm not sure if it is possible to build some superconductor based fancy power converter, other than that you can make wires only so big, and pass devices only so wide, some Joules are gonna be turn into heat for sure. \$\endgroup\$ – Vladimir Cravero Feb 28 '16 at 10:14
  • \$\begingroup\$ Vladimir Cravero, average modern PC are actually needs 100A to run its CPU. The main smps may supply 3.3V at 30Amps but the on board regulator switch it down to around one volt at hundreds of amps. \$\endgroup\$ – soosai steven Feb 28 '16 at 14:43
  • \$\begingroup\$ @soosaisteven Look at the size of the PCB and IC conductors on these high-power PCs: I have a hard time seeing how they can carry 100A! Doesn't that require a conductive cross-section on the order of 5AWG? (Not disputing that they do; just don't see how!) \$\endgroup\$ – feetwet Feb 28 '16 at 15:34
  • 1
    \$\begingroup\$ @feetwet They use multiphase supplies from the +12 rail, so there might be 4, 6 or 8 supplies effectively in parallel, each with 1-2A in (and 10-12x that much current out). The supplies surround the CPU chip. \$\endgroup\$ – Spehro Pefhany Feb 28 '16 at 15:58
  • 1
    \$\begingroup\$ Yes, the poly phase switching power supply interleaved in 12 phases is now common in high end computer boards. The connection between the source and load are made through different copper clad layers in the 8 layer circuit board. The efficiency of the poly phase regulator can be as high as 97%. You may check such products listed in linear technology website. \$\endgroup\$ – soosai steven Feb 28 '16 at 17:58
1
\$\begingroup\$

The problem in your linked question is that you specified an low-frequency (60 Hz) 10 to 20 mA constant-current power supply with sinusoidal output implied. (OP's application is an electrolytic chemical reactor.)

A switched mode supply with high efficiency would be possible but a lot of trouble to design. For a one-off application like this an analog PSU is easier to design but you live with the cost and problems of wasted energy.

Is there some minimum waste power cost involved in stepping voltage up or down? ... limited to cases where the input and output are multiples, and it gets more expensive in terms of waste power to hit a specific voltage that isn't an integer (AC) or binary (DC) multiple of the input?

No. In AC the voltage can be adjusted to the nearest turn on either winding of a transformer. In practice this will be very close and much less than the variation from your incoming supply. In DC it can be set to any value and is infinitely variable without penalty.

Current seems to be an efficiency killer. I.e., once current in a circuit becomes significant it seems like there's no way to control it without generating a lot of waste.

Not so. All the modern LED lamps are constant-current switched-mode power-supplies designed for high efficiency. It's just that regulation of the supply is based on current feedback rather than voltage feedback. See my answer to your linked question for an example of current feedback.

Finally, it's possible to use switched-mode power supplies to regulate both voltage and current efficiently.

| improve this answer | |
\$\endgroup\$

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.