voltage divider and rectifier circuit

[spectrum]

hello all,

hope i'm not annoying you with these stupid circuit tests, i studied electronic at school, but i have never developed a circuit with op amps and impedance problems, so i am looking for some help and clarifications.

My problem is: i have a sinusoidal wave 200Vpp signal that is applied as a carrier to an antenna.
In particular circumstances this carrier signal sometime can have 210Vpp and some others 200Vpp. I need to reduce this voltage and have as output in some way a dc level <5V, or a 1/0 logic state to give it as input to a microcontroller.

But i'm not looking for the best ready solution, i want try to do a circuit by myself, understanding things step by step.

I was trying dividing and rectifiying directly with a schottky, without success, so attached there is a diagram with my new approach.




So i've tought to reduce the voltage with some MOhms resistors, to avoid to create problem to antenna transmission. So, antenna still works fine. Then, i've tought to give the signal to an op.amp, to have low impedance and more current available as output. Then to rectify and have a DC level between 0 and 5V.

So i have some questions:

1) in the point "A" i was expecting a 20Vpp sinusoidal wave, but with x10 oscilloscope probe i see a 2Vpp signal.

2) in the point "B" i was expecting a non-inverted signal with same as the signal in "A". I see a 5 Vpp only positive signal, with negative part cutted and a little cutted at 5V. I think that is becouse i used an only positive power for the op-amp, but i don't understand the amplification effect, idoes it should have gain 1 in this config ?

3) in point C is finally see a wave like in B, with a more clean cut near 0,2V.

In particular, i can't understand why i have so different voltage in "A" from what i was expecting.


Thanks, angelo

 

[Nigel Goodwin]

You don't give anywhere near enough information, such as exactly what the source is, what frequency it is, and what it's impedance is.

 

[ericgibbs]

hello all,

hope i'm not annoying you with these stupid circuit tests, i studied electronic at school, but i have never developed a circuit with op amps and impedance problems, so i am looking for some help and clarifications.

My problem is: i have a sinusoidal wave 200Vpp signal that is applied as a carrier to an antenna.
In particular circumstances this carrier signal sometime can have 210Vpp and some others 200Vpp. I need to reduce this voltage and have as output in some way a dc level <5V, or a 1/0 logic state to give it as input to a microcontroller.

But i'm not looking for the best ready solution, i want try to do a circuit by myself, understanding things step by step.

I was trying dividing and rectifiying directly with a schottky, without success, so attached there is a diagram with my new approach.

View attachment 16469


So i've tought to reduce the voltage with some MOhms resistors, to avoid to create problem to antenna transmission. So, antenna still works fine. Then, i've tought to give the signal to an op.amp, to have low impedance and more current available as output. Then to rectify and have a DC level between 0 and 5V.

So i have some questions:

1) in the point "A" i was expecting a 20Vpp sinusoidal wave, but with x10 oscilloscope probe i see a 2Vpp signal.
If the Zsrc is very high, you must add the Zs to the 10M0 value, then calculate the Vdrop across the 1M0[R2]
2) in the point "B" i was expecting a non-inverted signal with same as the signal in "A". I see a 5 Vpp only positive signal, with negative part cutted and a little cutted at 5V. I think that is becouse i used an only positive power for the op-amp, but i don't understand the amplification effect, idoes it should have gain 1 in this config ?
You have to bias the input, to say half the Vsupply, if you want a high/low swing on the ouput of the opa, when using a single supply
3) in point C is finally see a wave like in B, with a more clean cut near 0,2V.
The opa output is always positive and its reduced by the forward volatge drop of the diode.
In particular, i can't understand why i have so different voltage in "A" from what i was expecting.


Thanks, angelo

Do you follow this?

 

[ericgibbs]

hi,
This is an option, you may need to add a small value cap across the zener,,, really need more information,

 

[spectrum]

Nigel Goodwin,

sorry, src frequency is 125Khz, is a pwm generated from a PIC, and amplified to 200Vpp.

ericgibbs,

many thanks, now is more clear.

About the impedance of the source, how can i measure the value of it ?
Since i have to rectify then, could i use the fact that the opa caut the negative part, avoiding the schottky diode after ?

thanks again,
Angelo

 

[ericgibbs]

Nigel Goodwin,

sorry, src frequency is 125Khz, is a pwm generated from a PIC, and amplified to 200Vpp.

ericgibbs,

many thanks, now is more clear.

About the impedance of the source, how can i measure the value of it ?
You say that the Vpk of the sinsoid is 200V and you measure 2Vpk across a 1M0 resistor. If the input resistor is 10M0 this would indicate that the Zsrc must be about about 100M0, as an estimate!.

Since i have to rectify then, could i use the fact that the opa caut the negative part, avoiding the schottky diode after ?
You could do it that way, BUT the OPA output could/would switch at the 125KHz rate, is that acceptable??

thanks again,
Angelo

hi,
Is it a continous 125KHz input? and are you requiring to detect every half cycle??

 

[spectrum]

yes, is a continuos pwm 125Khz, 50% duty cycle, generated every second, for about 100msec., i need to detect the level after a fixed amount of time like 20msecs.

thanks angelo

 

[ericgibbs]

yes, is a continuos pwm 125Khz, 50% duty cycle, generated every second, for about 100msec., i need to detect the level after a fixed amount of time like 20msecs.

thanks angelo

hi,
If you are getting 2V across the 1M0, I would suggest a 2.2M0 to give about +/-4Vin. The opa is connected as non inverting amp with a gain of 1.

I would recommend that you dont allow the input swing to exceed +5V as you are working with a +5Vsupply... if you are not sure of that the input voltage is constant at 200V, it would be wise to use a 4.7V zener to give a constant +4.7Vout.

When you say the level, do you mean the amplitude of the output or the presence of the output, that is '1' or '0'..??

 

[Nigel Goodwin]

Nigel Goodwin,

sorry, src frequency is 125Khz, is a pwm generated from a PIC, and amplified to 200Vpp.

ericgibbs,

many thanks, now is more clear.

About the impedance of the source, how can i measure the value of it ?
Since i have to rectify then, could i use the fact that the opa caut the negative part, avoiding the schottky diode after ?

Post the circuit of the unit, so we can see what's happening.

An obvious solution, without knowing what's happening, would be to use a suitable tuned transformer to reduce the voltage (like a crystal set - no opamp or power required).