I am dealing with a signal that produces an AC voltage ranging from 0-5 volts. The analog device that I am using requires a DC analog input of zero to 5 volts DC. I have produced "semi" successful results using a bridge rectifier design with one problem. When the AC signal voltage is below 1 volt or so, the DC voltage drops to zero due to the voltage drop across the diodes. I need a simple way to get the full range of the DC voltage to match the AC signal. FYI; The AC signal is being generated from a current transformer which produces the 0 - 5 Volt AC voltage. I appreciate any help.

 

You need an “ideal diode” or precision rectifier. There’s an op-amp circuit for that. Somebody else will likely find it for you faster than I do.

 

There are several op-amp circuits called "precision rectifiers" that will do exactly what you are in need of. Mostly they use diodes in the opamp feedback to cancel the effect of the diode drop. Mostly they work quite adequately, but you may need to match diodes to have similar forward voltage drops.You can find at least one version at the "schematics for free" site. And probably another at the "free information society" site.

 

There are several op-amp circuits called "precision rectifiers" that will do exactly what you are in need of. Mostly they use diodes in the opamp feedback to cancel the effect of the diode drop. Mostly they work quite adequately, but you may need to match diodes to have similar forward voltage drops.You can find at least one version at the "schematics for free" site. And probably another at the "free information society" site.

Thank you very much! Opamps are foreign to me but I somewhat understand their principle. Thank you for the link. I will build and test the circuit.

 

Thank you very much! Opamps are foreign to me but I somewhat understand their principle. Thank you for the link. I will build and test the circuit.

Note that we’re assuming from what you said that you want to process an “information” level, low current and low power signal. This won’t work if you expect significant power at the output.

 

Note that we’re assuming from what you said that you want to process an “information” level, low current and low power signal. This won’t work if you expect significant power at the output.

Yes, it is likely less than 20mA.

 

Is that 5Vpk AC or 5Vrms AC?

 

Mostly they work quite adequately, but you may need to match diodes to have similar forward voltage drops

A precision rectifier puts the diode(s) in the feedback loop, so the effective forward voltage is generally reduced by the open loop gain of the opamp.
This makes the forward drop voltage usually negligible (typically <10μV), thus matched diodes would offer no significant accuracy improvement to the rectified voltage.

 

Note that we’re assuming from what you said that you want to process an “information” level, low current and low power signal. This won’t work if you expect significant power at the output.

Given that the AC derives from a current transformer that is the assumption that I made, that it is for instrumentation purposes.

 

Given that the AC derives from a current transformer that is the assumption that I made, that it is for instrumentation purposes.

Any chance you could help me specify an OPAMP for this application? I forgot to mention that I only have a single voltage supply of +24VDC on my PCB board. No sense in making this easy!!

 

Below is an LTspice simulation of a full-wave precision rectifier with a 2-pole active filter to give a smooth DC output.
The gain of U1 is adjustable by U3 from 1 to 3.
The simulation is shown with a gain of 1, which gives a DC output that is the rectified average of the input (0.636 of the peak sinewave value). This output value is valid even if the waveform is distorted, since the filter inherently gives the average of the rectified input, regardless of the waveform shape.
The gain adjustment can be used to give a DC output equal to the RMS input (0.707 of the peak sinewave), the peak sinewave input, or whatever is needed, however those values will be inaccurate if there is significant distortion in the input sinewave (or it is other than a sinewave).

The cheap and common LM324 quad op amp can be also used, with some sacrifice in accuracy.

 

Before we go any further maybe one should ask, "What are you going to do with the 0-5VDC signal?"
If you are attempting to measure the amplitude of the AC signal then you are taking the wrong approach.

 

If you are attempting to measure the amplitude of the AC signal then you are taking the wrong approach.

And why is that?

 

And why is that?

Sorry, I was referring to the TS. I know you are working on a precision rectifier which is the proper approach.
But I am thinking that there might be a simpler solution if we knew what the TS intends to do with the 0-5VDC signal.

 

The need for converting the AC 0-5volt signal to DC as the input of my analog device (PLC ANALOG MODULE) requires DC to operate.

The device that I am measuring is a Current Transformer which puts out an AC 0-5volts. These are a the "donut" type units where the MAIN power lead passes through the transformer. 0 amps = 0 volts to 100amps = 5 volts (AC).

As mentioned, I only have a 24VDC+ supply power to work with. The good news is that once I get the DC signal into the analog device, I have the ability to scale and filter the signal. I just need to get the signal converted to DC voltage in some repeatable fashion. I can do the scaling (and some filtering) internally with programming.

Lastly: I mentioned that I need to keep it simple (less complex) as I have to convert 12 channels of this on some limited PCB space. If anyone knows of an IC that will help in the reduction of space or complexity, I am all in.

I really appreciate everyone's input.

 

Analog Devices company probably has an IC that will do the conversion, although the circuit shown in post#11 can do it using one quad op-amp IC per channel.And I am not convinced that the intermediate stage is really needed, since most precision rectifier circuits only use 2 op-amps. It should be possible to scale the voltage in the filter stage. I recommend dropping the 24 volt supply to 15 volts with a regulator to both reduce the op-amp power and to gain some headroom versus the maximum allowable IC supply voltage. And if you build the package with surface mount ICs and components it will be quite small.

 

Question: Why not just add a DC offset to the AC signal so that it’s lowest peak is roughly 0V relative to DC ground?

 

Question: Why not just add a DC offset to the AC signal so that it’s lowest peak is roughly 0V relative to DC ground?

Finding the peak value would be a real challenge, and also it would need to be scaled. In addition to those problems, that offset voltage would need to be perfectly stable. Besides that, adding the offset with an op-amp would take up almost as much area as the precision rectifier circuit. Adding 5 volts would assure that the entire sine wave would be over range, at least before scaling. Even if the input range covered zero to ten volts, offsetting would immediately cut the resolution in half.

 

Take a look at precision peak detector circuits:

http://www.electronics-tutorial.net/analog-integrated-circuits/peak-detector/

 

Even if the input range covered zero to ten volts, offsetting would immediately cut the resolution in half.

I see your point but depending on the application, that might not be a problem. Other error sources might far exceed a small resolution loss.