I have a "panel" connected to my main circuit board via conducting lines. I intend to transmit small voltages (of the order of say 0.125 volts) from my main board to the panel.

For my model this is no problem, since i would not be transmitting over long distances, the distance of separation will be << 1m. If the circuit will be expanded for real-world implementation the conducting line will be much longer, something of the range of > 3m (perhaps), or even more.

So I thought a form of line driver might be necessary if these small transmitted voltages won't get significantly attenuated. I thought of making them ride upon a higher voltage, 5v, where I'll kind of "demodulate" them at the receiving panel.

I thought of using an opamp adder (summing the small voltage with 5V) at the transmitter (cct board) and using a subtractor (subtracting 5v form 5v+Vsmall) at the receiving panel to extract the original small voltage.

Is this the method usually used? How brilliant is this approach?

cannot comment with the above details . do u need a binary waveform.?, and why a low voltage transmitted to ~3mtrs , and how about the noise interference.?

The 0.125v is dc. I'll be driving an LM 3914 (which in turn drives 10 LED's) with it. I'll actually be sending ten discrete voltage levels of 0.125V, 0.25V, up to 1.25V.
The panel is going to be remote from the main circuit board.
Yes, of course the lines are wires!

Emm, yes, noise - well that's the main purpose of the post anyway. I intend to amplify the voltages to make them more perceptible so as not to be mistaken for noise.

I hope this has clarified the ambiguous.

It's still pretty vague?.

But as you're amplifying the voltages anyway?, do it at the beginning, rendering this question pointless!. This will also give a low impedance output, which will help prevent noise or stray pickup.

as nigel adviced , it would be better to amplify the i/p (say x10 ) and use a voltage divider in the 3914 ( /10) to reduce the noise problems

Originally Posted by akg

as nigel adviced , it would be better to amplify the i/p (say x10 ) and use a voltage divider in the 3914 ( /10) to reduce the noise problems

yes this is brilliant and a cheaper choice than using adder/subtractor combination which uses 2 opamps as against 1 for this one. Yes Nigel I'm amplifying the voltage, definitely, what I was asking was not about amplifying or not but on the method I intended to achieve the amplification.

another question, to what cable length should dc voltage amplification be considered necessary? is there any standard? I know about 500m for Coax but this is due primarily to frequency/bandwidth.

Originally Posted by aibelectronics

yes this is brilliant and a cheaper choice than using adder/subtractor combination which uses 2 opamps as against 1 for this one. Yes Nigel I'm amplifying the voltage, definitely, what I was asking was not about amplifying or not but on the method I intended to achieve the amplification.

I think you're making this hard work? - amplify it before you send it down the cable, a simple opamp is all you need.

another question, to what cable length should dc voltage amplification be considered necessary? is there any standard? I know about 500m for Coax but this is due primarily to frequency/bandwidth.

DC will lose voltage due to the resistance of the cable as it gets longer - easiest thing to do is use CURRENT rather than voltage, this overcomes the problem (until it gets too high a resistance).

Exactly, use a transconductance amplifier at the source to convert the voltage to a current and a resistor at the destination to convert this current to a voltage.