Voltage doublers, with rigour

[leadbelly]

I am confused by the design details of voltage doublers, as I think terminology is often an issue.

Based upon https://en.wikipedia.org/wiki/Voltage_doubler

The terminology of half-wave = Greinacher = Cockroft-Walton

Full-wave = Delon = bridge = floating

So for the half-wave, diodes must be rated sqrt(2) * Vac, and caps rated 2 * sqrt(2) * Vac?

And for full-wave, diodes rated 2 * sqrt(2) * Vac, and caps rated sqrt(2) * Vac?

What is the ripple of each type of voltage doubler?

 

[Elvee]

So for the half-wave, diodes must be rated sqrt(2) * Vac, and caps rated 2 * sqrt(2) * Vac?

The diodes need a 2*Vpeak rating.
Only the output cap needs 2*Vpeak, for the input one Vpeak is sufficient

And for full-wave, diodes rated 2 * sqrt(2) * Vac, and caps rated sqrt(2) * Vac?

Yes

What is the ripple of each type of voltage doubler?

The approximate formula remains applicable if you enter the right values: Vripple= Iout/(Fripple*C).
For the first variant Fripple=Fmains and C= Cout and for the second Fripple=2*Fmains and C=C/2
In practice, due to second order effects, the ripple of the second variant will be somewhat smaller.

Note that for me, the distinction full/half wave is not very relevant: both use both polarities in some way, and a true full wave doubler requires twice as many capacitors and diodes

 

[basreflex]

both schematic are identical, being both half wave type doubler.

only the capacitor placement differs, still in series with the coil. the cap ratings are indeed different, as the series cap mode has half the final voltage.

 

[AndrewT]

both schematic are identical,............... the cap ratings are indeed different,...............

???
How can the schematics be identical and yet impose different voltages on the capacitors?

 

[Elvee]

In summary the circuits are identical, except the arrangement of diodes and capacitors is different, as is the voltage seen by one capacitor, the ripple frequency and the internal resistance....

 

[TonyTecson]

for the half wave case ripple frequency is 50/60hz
for the second case, the full wave, ripple frequency is doubled, 100/120hz
i use the second one all the time...

 

[basreflex]

on second inspection of the circuit the full wave is indeed full wave, one half wave is delivered by the diodes, the other half is delivered by the capacitor. so indeed the full wave has the double ripple frequency, just by clever placement of the capacitors. I have used the FW version up to 7Mhz

 

[Ketje]

There can be a little problem with the Greinacher circuit.
If the capacitor on the transfo side is smaller than the other one it can get charged in two directions.If you use a smaller polarised capacitor put a diode across to prevent charging in the wrong direction.It makes also the rise time after turn-on shorter.
Mona