Wide input range Amp

crutschow said:

RMS is the DC power equivalent voltage of a sine wave. It's approximately 0.707 of the peak voltage (not peak-to-peak).

So is your value RMS, peak, or peak-to-peak?

Click to expand...

Hi Crutschow,

I was reading an Oscilloscope 10V Peak to Peak.

Cheers.

I was reading an Oscilloscope 10V Peak to Peak.

Click to expand...

Than its not 10 volts RMS.

Here are some buffer circuits to use with the AD8310 log circuit.

The input op amp attenuates the 10Vpp to the 4Vpp maximum input of the AD8310 and drives the relatively low input impedance of the AD8310.

The output op amp provides gain to increase the 2.7V max AD8310 output to the 5V of the PIC A/D converter input.

The op amp's power is not shown but it must be the appropriate voltages for the op amp chosen.

THAT2252

Input has about 100db of dynamic range. VERY LARGE

Output is =-.3 to +0.3 volts. A op-amp and resistors can turn that into 0 to 5 volts.

Only works with AC input it will not work with DC. Change the input resistor so that 10V AC input = +0.3 volts.
This is like the other log-amp but built for audio.

hi camerart.
There is a low cost way which should give results close to your requirement.

Using a TS924 quad rail2rail OPA, configure one stage with a gain of 5, the output of this drives a peak detector/hold.
The output of peak hold circuit is connected to one adc input of the PIC, via a 4k7 current limiting resistor.
This means the 1mV input signal is amplified and will give a peak of 5mVdc at the adc1 input, as the adc is set for 1023, this will give a resolution of 1bit LSB.

When the AC input signal is 1Vpk the adc1 will be 5V, as the the input signal exceeds 1Vpk, the 4k7 will limit the input to adc1 [ you could use a 5.1V zener as a clamp is required]

With a second TS924 configure one stage with a gain of 1, drive a second peak/hold detector.
The output of this second peak detector is connected to a second adc2 pin input via a 2 to 1 resistive divider, so the adc2 input range is 0V thru 5V.
Also set for 10bit resolution.
Both these circuits are driven by the same 1mVpk thru 10Vpk signal.

In the PIC program the program tests the voltage at adc1, if its less than 5Vdc it will use that value, if its 5V [ or over] then it will use the adc2 voltage input.

It sounds more complicated than it really is to build, just 2, quad OPA TS940 and a few components.

The PIC's adc is set for a 10bit conversion

Hi Crutschow,

OK, Thanks. I'll see if there are any cheaper options first, and if not I'll try this.

Cheers.

Hi Eric,

Your idea sounds interesting. Since embarking on this project quite a while ago, I have bought loads of different Opamps for it, I am assuming that the TS924 is different in some way.

It does sound complicated to me but, If I attempt to draw a circuit and post it, will you correct errors for me, as I am certain it won't be right immediatelly.

It shows rail to rail on the data sheet, will it work from a singal supply (Battery)?

It seems that this circuit uses a PIC chip? So will this chip aoutput 0V to 5V dc for the A/D input on my main board.

Cheers.

camerart said:

Hi Eric,

Your idea sounds interesting. Since embarking on this project quite a while ago, I have bought loads of different Opamps for it, I am assuming that the TS924 is different in some way.

It does sound complicated to me but, If I attempt to draw a circuit and post it, will you correct errors for me, as I am certain it won't be right immediatelly.

It shows rail to rail on the data sheet, will it work from a singal supply (Battery)?

It seems that this circuit uses a PIC chip? So will this chip aoutput 0V to 5V dc for the A/D input on my main board.

Cheers.

Click to expand...

hi,
Beware the 'rail2rail' claims of the datasheets, they really mean 'nearly rail2rail', within a few tens of mV, which for most applications is acceptable.

Get the d/s for TS924 and see if the 'near to 0V output' is going to be OK for your project.
They are designed to work on a single power rail.

As I explained for the adc1 input, the OPA would swing close to +10V when the Vac input is 2Vpk, all one has do is connect a 4k7 resistor in series and use a 5.1V zener to hold the adc1 input to 5.1V

The second circuit output OPA would swing to +10V, but this would be connected to 4k7/4k7 resistive divider , so +5V at the adc2 input.

If excess over volts was a problem, say +12V max , again a 5.1V zener clamp could be used.

IF you really must have 1mV [ 5mV] close to 0V output you will have to use a dual supply rail and a different type of OPA.

With say two 9V batteries for +/-9V you could get down to 0V.
If you decided on that, consider the CA3240 dual low power OPA's

OK

ericgibbs said:

hi,
Beware the 'rail2rail' claims of the datasheets, they really mean 'nearly rail2rail', within a few tens of mV, which for most applications is acceptable.

Get the d/s for TS924 and see if the 'near to 0V output' is going to be OK for your project.
They are designed to work on a single power rail.

As I explained for the adc1 input, the OPA would swing close to +10V when the Vac input is 2Vpk, all one has do is connect a 4k7 resistor in series and use a 5.1V zener to hold the adc1 input to 5.1V

The second circuit output OPA would swing to +10V, but this would be connected to 4k7/4k7 resistive divider , so +5V at the adc2 input.

If excess over volts was a problem, say +12V max , again a 5.1V zener clamp could be used.

IF you really must have 1mV [ 5mV] close to 0V output you will have to use a dual supply rail and a different type of OPA.

With say two 9V batteries for +/-9V you could get down to 0V.
If you decided on that, consider the CA3240 dual low power OPA's

OK

Click to expand...

Hi Eric,

This seems a good idea, is it because of the wide range of input voltages make things difficult.

Now I have to figure out Peak detector/Hold.

Have a look at the attachment. Is this something like what you suggest?

Will LT1013 Op amps work for testing, as I have some?

Cheers.

hi,
The LT1013 will swing close to 0V, but the upper swing is ~ +Vs - 2V.
From the datasheet the power output swing appears limited, especially at 50KHz.

I have moved the Zener diodes.

You do realise that you will have to resistively attenuation the input signal.?
Its not permitted to drive the inputs with a signal that swings close to the power supply voltages.

Lets see your peak detect and hold circuit.

ericgibbs said:

hi,
The LT1013 will swing close to 0V, but the upper swing is ~ +Vs - 2V.
From the datasheet the power output swing appears limited, especially at 50KHz.

I have moved the Zener diodes.

You do realise that you will have to resistively attenuation the input signal.?
Its not permitted to drive the inputs with a signal that swings close to the power supply voltages.

Lets see your peak detect and hold circuit.

Click to expand...

1/ So no to LT1013

2/Zeners, OK.

2/ No, but I'll put a resistor in.

3/I had to Google Peak and hold, have a look. I presume the Peak and hold part takes up 2X Opamps form the Quad.

Hi Ron,

Looks interesting, I'll concentrate on EricGibbs circuit for now, thanks.

Cheers.

camerart said:

Hi Ron,

Looks interesting, I'll concentrate on EricGibbs circuit for now, thanks.

Cheers.

Click to expand...

hi,
I would be interesting to help with the design but I would need to know a lot more about the application.

Top of list would be,, when you have the project running and you are able to read the adc values into the program, what output function will the program control.?

Basically whats the purpose of detecting the ~55KHz sine wave 1mVpk thru 10Vpk.? and what are you going to do with it.??

You have mentioned in your posts that the input signal is 1mV,, 1mVpk etc and it dosn't swing thru zero volts.
Can you post a simple diagram showing the waveform and the voltage points.?

Also how is the 1mV to 10V coupled to the input of the OPA's, is it a pick up coil, capacitor or some other connection.?

Give us more details of the application, we may be able to suggest simpler alternative solutions.

hi,
This should help get you started.
In the zip is the circuit diagram and 1mV thru 10V plots.

The images are 1mV,10mV and 1V ......... [ *5]

EDIT:
Added the JFET CC.

ericgibbs said:

hi,
I would be interesting to help with the design but I would need to know a lot more about the application.

Top of list would be,, when you have the project running and you are able to read the adc values into the program, what output

function will the program control.?

Basically whats the purpose of detecting the ~55KHz sine wave 1mVpk thru 10Vpk.? and what are you going to do with it.??

You have mentioned in your posts that the input signal is 1mV,, 1mVpk etc and it dosn't swing thru zero volts.
Can you post a simple diagram showing the waveform and the voltage points.?

Also how is the 1mV to 10V coupled to the input of the OPA's, is it a pick up coil, capacitor or some other connection.?

Give us more details of the application, we may be able to suggest simpler alternative solutions.

Click to expand...

Hi Eric,

The project, hopefully will be versatile and be used for a few other applications.

One use it to aim and focus Cameras, where other methods have failed. I use wires with the signal passing through, and coils to

pick up the signal. The feedback from the coil goes through the A/D input to a program that combined with another program moves

the Camera at a certain time by switching motors,and is controlled by the distance read from the coils.

This is all working apart from when using long wires, where the signal is too small (1mV ish)

The existing set up uses 1X 12V Battery.

I was wrong when I said the signal doesn't go through zero volts (Sorry). One end of the coil is GND and the other end swings

about Zero Volts.

So with the long wire the signal can be as low as 1mV and with other setups and Camera movement, the signal could go up to 10V.

I have sent people round in circles before with my explanations, so probably best to stick to the circuit.

Hope this helps.

Cheers.

ericgibbs said:

hi,
This should help get you started.
In the zip is the circuit diagram and 1mV thru 10V plots.

The images are 1mV,10mV and 1V ......... [ *5]

EDIT:
Added the JFET CC.

Click to expand...

Hi Eric,

I've been swatting up on Opamps, but I would have never ended up with this circuit, it and the readings looks great.

One problem is that my system uses 1X 12V Battery. Can your circuit be altered to suit?

If this is possible, I'm sure it would fit the bill. Thanks.

Cheers.

camerart said:

Hi Eric,

I've been swatting up on Opamps, but I would have never ended up with this circuit, it and the readings looks great.

One problem is that my system uses 1X 12V Battery. Can your circuit be altered to suit?

If this is possible, I'm sure it would fit the bill. Thanks.

Cheers.

Click to expand...

hi,
Its really got to be a dual supply.


This sim shows the 1V to 10V part

ericgibbs said:

hi,
Its really got to be a dual supply.


This sim shows the 1V to 10V part

Click to expand...

Hi Eric,

It certainly looks good. I'll have to have a good think.

I quick thought!!! The existing 12V Battery is quite heavy, but would a small second one paralleled to the first, just for this circuit be ok? If so, as it needs to last some time, how much would this whole circuit consume?

Cheers.

camerart said:

Hi Eric,

It certainly looks good. I'll have to have a good think.

I quick thought!!! The existing 12V Battery is quite heavy, but would a small second one paralleled to the first, just for this circuit be ok? If so, as it needs to last some time, how much would this whole circuit consume?

Cheers.

Click to expand...

hi,
Look at this image , it shows the +V and -V supply currents.

Approx 11mA for +12V and 7mA for the -12V.

The two 12V batteries are in Series not Parallel,

I have added the two adc inputs to show how they are related, note I have a used a 1mV to 10V DC sweep to show the plots.

ericgibbs said:

hi,
Look at this image , it shows the +V and -V supply currents.

Approx 11mA for +12V and 7mA for the -12V.

The two 12V batteries are in Series not Parallel,

I have added the two adc inputs to show how they are related, note I have a used a 1mV to 10V DC sweep to show the plots.

Click to expand...

I nearly understand it, an it looks good to me.

Did I say batteries in parallel! Of course it is series. So the second battery @7mA won't need to be too big either. A couple of A/hrs = A week or so. That's fine.

I will need to Divide ADC /2.

It will be w hile before it is set-up, with circuits and programs etc, but I'll let you know the outcome.

Thanks for you efforts.

Cheers.