Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Cascading amplifier with a line driver for an analog signal transmission

Status
Not open for further replies.

doncarlosalbatros

Full Member level 2
Joined
Nov 26, 2016
Messages
143
Helped
1
Reputation
2
Reaction score
1
Trophy points
18
Activity points
1,758
Hello,

I'm planning to implement the circuit below I have drawn. But because there is no SPICE for such components I couldn't verify or test any overall workings.

difffd.png
(left-click to enlarge)

From left to right I will first summarize the signal chain:

The transducer is a uni-polar and balanced transducer and it has three terminals marked as SG_out, SG_ref and AGND. It has normally swing of +/-200mV. I want to amplify this transducer signal and send to a 20 meters far away data acquisition input in a balanced way. Signal interest range can vary between +/-100mV to +/-10V and freq from DC up to 30kHz.

The power supply is a +/-15V dual supply where I marked its terminals as +15V, -15V and PWR_GND.

The fixed gain differential amplifier INA106 is chosen for a stable gain and CMRR for temperature changes.

Balanced line driver is right after the differential amplifier is chosen to ensure the balanced output for the transmission.

STP cable carries line driver's outputs all the way to a differential ended and balanced data acquisition inputs. The STP cable's shield is grounded to the ground of the data acquisition ground AIGND.

The data acquisition receives the signals Vo+, Vo- from the STP cable.

If the overall wiring is correct, my questions are:

1-) As you see the STP cable carries Vo+, Vo- and shield but not the analog ground of the incoming signal chain. Does that mean I should ground the balanced driver's ground(from its pin3) with a separate wire to the AIGND as drawn in the schematic? And at what point of the signal chain ground(from AGND to AIGND) is better to connect and ground the power ground terminal PWR_GND? An edit(maybe with a image editor) on my diagram would really help better.

2-) Is my wiring correct or is there any fundamental problem you see?

I would be very glad to hear your opinions before implementing this.
 

Hi,

The first that comes into my mind:
* Why do you want to amplify the signal ... before the cable?
I assume you want to improve signal to noise. If so, then I don't think there is much benefit. Even in best case you gain only a couple of dB. And only if there is differential noise along the cable. Mind: You might expect common mode noise along a twisted and shielded cable. --> thus I don't see much benefit

*******
When I see your schematic there is:
* a differential signal source
* a differential to single ended amplifier
* a single ended to differential amplifier
--> Why the "single ended" mode?

Why don' t you just use a fully differential amplifier? Differential_in differential_out?

You even may build such a thing with two standard amplifiers: ...the left side part only -with greenish backcolor.
opamp21.gif

****
Regarding Ground wiring:
In opposite to single ended wiring: here - with differential wiring - the GND wiring is not very important. Ideally there is no GND current and GND does not act as signal reference. The GND just acts to prevent common mode voltage from runaway.
You may connect GND anywhere.

******
My recommendation:
I'd use no amplifier, but send the already differential signal without amplification across the wire. Use a twisted pair wire with good shielding. Thus you also need no power supply (for the amplifiers) at the sensor side.

****

In every case: use good differential mode filters as well as single mode filters to maintain best signal to noise.

Klaus
 

The problem has been extensively discussed in your previous thread https://www.edaboard.com/showthread.php?382511-Need-help-with-fully-differential-amplifier, two solutions have been already suggested:
1. Using pseudo differential connection of the data acquisition system
2. Using an inverting amplifier to generate Vo-

The bandwidth requirement has now increased from 150 Hz to 30 kHz, but the said solutions will still work for the relative low frequency range. Using a true differential driver is however a valid option.

Why don' t you just use a fully differential amplifier? Differential_in differential_out?
We didn't yet get a full specification, but analog precision (offset, offset drift, low frequency CMRR, gain drift) is probably better with low frequency instrumentation amplifiers.

Does that mean I should ground the balanced driver's ground(from its pin3) with a separate wire to the AIGND as drawn in the schematic?
Absolutely.

And at what point of the signal chain ground(from AGND to AIGND) is better to connect and ground the power ground terminal PWR_GND?

Not so important. I would expect a solid analog ground plane through the preamplifier.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top