No load output voltage of a bridge rectifier

[sridhar10chitta]

Will a bridge rectifier circuit without a filter or load resistor connected across its output deliver a pulsating unidirectional voltage ?
An oscilloscope trace (dc input enabled) did show a trace of a full wave output though the half sine did not swing down to zero voltage completely and showed a noisy trace before following the next half sine. This is repeated in all the half sines.

 

[Phrak]

Between AC peaks the terminals are loaded by the scope probes and the diode capacitance--and leakage.

 

[vk6kro]

The bridge rectifier does have a load. It is the input resistance of the oscilloscope.
Because this is about 1 megohm, the reverse capacitance of the rectifier becomes important so you get apparent reverse voltages as part of the output.
Also at this load, stray capacitances, even from the oscilloscope's probe and lead provide some filtering.

Try it with 1000 ohms as a load and it should look more like the classical full wave rectified output

 

[sridhar10chitta]

Thanks. I can appreciate that there is a loading effect due to the oscilloscope probe.
My question is whether a pulsating voltage exists that cannot be seen using the oscilloscope ?
Or is it that the voltage makes an appearance only if a resistive load is assumed however light it is ?
Sridhar

 

[vk6kro]

The diodes in the bridge rectifier have reverse capacitance and, when the load is very high, even a small amount of capacitance will allow a reverse voltage to get to the output at the same time as the forward voltage from the other side of the transformer winding. In the extreme case, these might cancel each other out.

If that happened you would get no output at all.

Putting the oscilloscope or a large resistor in starts to give some recognisable waveform out.

As you make the load smaller,the diodes start to behave normally and the forward resistance of the diode is more able to conduct the larger current than the reverse capacitance is. So you get the traditional rectified sinewave pattern out.

My question is whether a pulsating voltage exists that cannot be seen using the oscilloscope ?

NO.

Thanks. I can appreciate that there is a loading effect due to the oscilloscope probe

This effect is negligible compared to the feedthrough of the reverse capacitance.

 

[sridhar10chitta]

Thanks.

 

[sridhar10chitta]

I am sorry for adding this.
Is my assumption that the reverse capacitance effect is felt at 50 to 60 Hz and will be more at higher frequencies right ?

 

[vk6kro]

Yes.

At higher frequencies, the reverse capacitance of normal silicon rectifier diodes makes them unsuitable. It makes them appear to conduct in both directions so the DC output is lower than it should be.

Switch mode power supplies use Schottky diodes because they have less reverse capacitance.
These supplies operate at much higher frequencies than 50 or 60 Hz.

 

[Phrak]

Switch mode power supplies use Schottky diodes because they have less reverse capacitance.
These supplies operate at much higher frequencies than 50 or 60 Hz.

I haven't heard that. Can you support it? You will need to compare similar reverse voltage rating and similar forward current rating. Schottky diodes have less forward drop. This is significant if you are rectifying a DC output of 5 or 3.2V or even 1.8V. The current trend is toward replacing the freewheeling diodes with switched FETS. Affordable celled FET series resistance has droped very nicely over the course of a couple decades and Schottkys are trending out for their comparatively high forward drop.

 

[Phrak]

My question is whether a pulsating voltage exists that cannot be seen using the oscilloscope ?

NO.

This cannot be right. With zero load, the applied AC voltage will follow through the diode capacitance only limited by leakage current.

 

[vk6kro]

Switch mode power supplies use Schottky diodes because they have less reverse capacitance.
These supplies operate at much higher frequencies than 50 or 60 Hz.

I haven't heard that. Can you support it? You will need to compare similar reverse voltage rating and similar forward current rating

They have several advantages like low capacitance, low forward voltage and fast recovery time:
http://www.nxp.com/#/pip/pip=[pfp=41751]|pp=[t=pfp,i=41751]

My question is whether a pulsating voltage exists that cannot be seen using the oscilloscope ?

NO.

This cannot be right. With zero load, the applied AC voltage will follow through the diode capacitance only limited by leakage current.

If you only had one diode that would be right.
However he was talking about a bridge rectifier where some diodes would be getting forward voltages and some would be getting reverse voltages. The capacitance of the reverse biased diodes would have as much effect as the conductivity of the forward biased diodes if the load was infinite.
Without trying it, or wanting to, my instinct would be that the two would tend to cancel out giving a greatly reduced or zero output.

 

[Phrak]

I have to admit, vk, that I forgot about the other diode. Good catch! As it turns out, the diode capacitance varies inversely with the reverse voltage. When the diode has zero volts across it, it's capacitance is greatest. Both diodes are reversed after the peak voltage on the primary, or transformer secondary--whichever it is. So the diode that was immediately conducting and has reversed, drives the open node the greatest.

This is mildly intersting. 90 degrees after the peak, the other diode capacitance is matched and the other diode starts driving the node higher. So ignoring leakage, the ripple should should be driven down no more than 100-70.7 percent from the peak.

 

[vk6kro]

Yes, it became pointless after a while.

I suppose you could model it with enough data, but life is too short to worry about waveforms out of open circuited bridge rectifiers!

 

[Phrak]

Yes, it became pointless after a while.

I suppose you could model it with enough data, but life is too short to worry about waveforms out of open circuited bridge rectifiers!

cheers, vk.