5

The textbook explanation is not doing it for me.

The half-wave voltage doubler shown in Fig. 4-27(a) offers some improvement in safety over the full-wave voltage doubler. Compare Figs. 4-24(a) and 4-27(a). The chassis is always hot in the full-wave doubler. In the half-wave doubler, the chassis is hot only if the connection to the AC outlet is wrong.

Textbook is Electronics Principles and Application, Ninth Edition, by Charles A. Schuler.

Could someone elaborate on how 4-24a is chassis hot but 4-27a is not? Google has not been helpful.

4-24a

4-24a

4-27a

4-27a

MicroservicesOnDDD
  • 2,557
  • 1
  • 13
  • 48
Al T
  • 61
  • 3

2 Answers2

8

A better way to say it would be "the output is common-ground with the source".

Hot Chassis

This term was relevant back in the Bad Old Days, when electronic equipment was allowed to be mains powered, without isolation or grounding, and may have had lax insulation. (I forget if actual metal chassis were allowed; wooden and Bakelite enclosures were common at least, offering some insulation in that case.) The advantage was, since the radio set (for example, the AA5 ("All-American Five") used five vacuum tubes, with their heaters wired in series, and supply voltage (B+) derived from a half-wave rectifier) could be directly mains powered, no transformer was required, saving on cost.

The difficulty is, circuits weren't wired well back in the day, and even if you plugged in the radio [what should've been] the right way around, maybe you got the "hot" line on the chassis anyway. And then 120VAC is touchable, and unhappiness ensues...

Mind, this remains true to this day, long after the adoption of polarized two- and three-prong receptacles. Polarized connectors make this confusion less likely, but the circuit may always be wired incorrectly (amateur, or incompetent professional, wiring), or indeed not be possible to wire "correctly" (some commercial or industrial installations may wire 208 or 240V between phases of a 3-phase circuit, in which case both leads will be hot).

Further reading on US receptacles: https://en.wikipedia.org/wiki/NEMA_connector

For these reasons, present standards are to treat mains as hazardous, regardless of how it is wired at the panel. This also addresses the condition that transients (surges) may raise the line-to-line or line-to-earth voltage to hazardous levels (several kV). Compliant equipment is insulated and isolated from mains, so that the user can never come in contact with it, even if a single point of failure occurs (say, one layer of insulation becomes compromised).

Common Ground

Since we will not use such a circuit for saving cost of equipment (i.e., the insulation is required regardless), we can still consider the value of this circuit by other means.

For example, in a switching power supply (SMPS), a high-frequency transformer is used to provide isolation, and we may have that one side of the secondary winding shall preferentially be grounded. This can happen due to the voltage on the winding, relative to others near it inside the transformer, and the capacitance between windings. (This is high-level stuff, so don't worry if it's going over your head; suffice it to say, high frequencies can have different ideas about what "grounding" means.) Voltage doubler circuits are not as often used in SMPS, but similar reasoning applies to the rectifiers that are used, as far as where the transformer winding's ground return should connect.

Author's Intent

While I can't speak for Mr. Schuler, it almost seems like one of those things that academics might latch onto -- something that is easily identified, and is (or at least used to be) useful. By this reading, it suggests an outdated sense of importance, or a lack of practical awareness; mains should most definitely not be tied to chassis these days. It does seem like it would be more reasonable to call it simply what it is: common ground. But I'm far from academics myself, and maybe these terms are still in common use in those circles. A definition or glossary would seem prudent, to address exactly such confusion as yours.

Tim Williams
  • 22,874
  • 1
  • 20
  • 71
  • 2
    Good observation, I was going to say something about how even the supposedly "safe" version looked like a deathtrap to me, but a lot of these textbook examples have a tenuous connection with reality. – vir Mar 22 '23 at 19:57
4

In 4-27(a), you can see that the ground symbol, by which I am assuming they mean chassis ground, is connected directly to one of the AC inputs, which I am assuming is the neutral one. Neutral is emphatically not ground, though they are bonded together at your service panel so there should only be a volt or two difference between them.

In 4-24(a), the neutral input is connected to the node between C1 and C2. Ground with respect to the load is also assumed to be connected to chassis ground and if C2 is charged then it is at line potential relative to neutral.

vir
  • 14,718
  • 13
  • 28
  • 1
    Awesome, I understand it now! Thank you. – Al T Mar 22 '23 at 19:30
  • 1
    Neutral should never be treated as ground. Both of these should be treated as hot chassis. One really good reason is: many plugs are reversible, so half the time live is neutral and neutral is live. – user253751 Mar 22 '23 at 19:44