Isolation between two devices

[FreshmanNewbie]

I want to get an input from a transformer and feed it into an op-amp which then gives it to an ADC.

This is a simple draing . The input on the secondary of the transformer is 10mV pk-pk/










This is the Op-amp used.

Can the transformer be connected directly to the inputs of the op-amp?

I have seen that some resistor has to be connected in series with the output of the transofmer?

Can someone tell me which is the correct method and why?

 

[stenzer]

Hi,

• I assume you want to amplify your signal, so you will need at least some input resistors to set your gain and of course a feedback network (external or the internal one).
• What's about opamp supply? Split-rail or single supply?
• Which transformer are you using? Does it provide a center tap to provide an input bias voltage?
• What is your frequency range of interest?
• How does your input signal looks like?
• Output resistance/resistor depends on the following/stage load e.g. a BALUN, 50Ohm matching, ADC input .....

BR

 

[KlausST]

Hi,

If you do a quick research on OPAMP amplifier circuits ... you will immediately see that all circuits use resistors.

What you need to know:
* amplifier power supply
* amplifier input voltage peak-peak (already given)
* amplifier output voltage, min and max
--> output_voltage_range / input_voltage_range = gain
* output offset voltage
* frequency range of interest
* maybe other requirements, like distortion, (erroneous) overvoltage, precision

****
For beginners I recommend to use a standard circuit "OPAMP differential amplifier circuit" using 2x resistor pairs (4 resistors in total).

****

When I do an eletronics design, I first decide the application´s requirements. Without looking for a solution.
Btw: by writing this, I don´t meant that you need to do it the same way as I. You are free to go your own way.

So I´m just guessing what your "basic requirements are". Correct me if I´m wrong.

AC input --> unknown analog processing --> ADC --> unknown digital processing --> result.

The most important:
* define the AC input
* define the expected result

You already come with a transformer. I guess the transformer is not a requirement, it is a solution. It could be a good solution or not.
I personally did some measurements involving transformers (many years ago), but since then I avoided them and used a resistive measurement solution..
Reasons for avoiding: used high_impedance measurement instead of true isolaton (no safety risk). I did not like the phase shift and the distortion caused by transformers.
And as a benefit the resistive solution was cheaper and is able to process DC.

Klaus

 

[D.A.(Tony)Stewart]

Define all the following before you choose a design;

1. Source impedance, signal range , amplitude, frequency range
2. Interferences ( Line + RF ) Rejection required ?
3. Output range, impedance