Can anyone tell me what is meant by a reference signal and where it comes from?

[leeb_965]

Im trying to get my head around lock-in amplifiers and it says that a reference signal is multiplied by a modulated signal but it doesnt actually say what a reference signal is.

 

[Harald Kapp]

The reference signal is a signal of the same frequency and phase as the signal you want to detect with the lock-in amplifier.
This signal could come from the original source where the modulated signal is generated. Or it could be regenerated from the received signal e.g. by a PLL.

Some theory is here: http://en.wikipedia.org/wiki/Lock-in_amplifier

Harald

 

[leeb_965]

Hi Harold, Do you know much about lock-in amplifiers.

Im tearing my hair out trying to find what the ref signal is. Every book Ive looked at says multiply the reference signal and the modulated signal but doesnt actually say what it is or where it comes from.

I have a full-bridge strain gauge set-up that is followed by a 3 op-amp instrumentation amplifier. I have the modulated signal amplified at the output but I dont know where this ref signal will come from.

 

[Harald Kapp]

I'm not particularly versed in lock-in amplifiers. I once learned about it during my university time.

From your description a lock-in amplifier is probably not the right tool. I assume your strain-gauge bridge operates from a DC supply voltage. If so, there is no signal to modulate and thus no signal to demodulate. A lock-in amplifier is useless here.

You could use a lock-in amplifier if you operate the strain-gauge bridge from an AC supply voltage. You can then use the same AC source as the reference signal. The signal from the strain-gauge bridge is then an AC signal modulated by the bridge unbalance. By multiplying this signal with the original AC frequency, the lock-in amplifier cancels all error components in the signal that do not have the same frequency as the AC source.
While this is technically feasible, it will not make sense in terms of effort/use ratio.

In your application it is probably better to stick with an instrumentation amplifier (http://en.wikipedia.org/wiki/Instrumentation_amplifier) which will cancel any offset voltages (and common mode disturbances) but will amplify the differential signal froim the strain-gauge bridge.
You don't have to vuild one yourself, you can buy them as integrated circuits, too.

Harald

 

[leeb_965]

I read that an AC source is better, so Im going to try this. So will I take the reference signal straight from the AC source? The AC source is 10V and 225HZ and the amplified signal at the output of the inst amplifier will be alot smaller so will it not matter that the amplitude of the source is much higher than the amplified signal as long as they are the same frequency?

 

[davenn]

Im trying to get my head around lock-in amplifiers and it says that a reference signal is multiplied by a modulated signal but it doesnt actually say what a reference signal is.

The reference signal is usually a VERY stable oscillator. eg a TCXO (temp controlled crystal oscillator) a rubidium freq standard, even a GPS derived signal. These oscillators can either be an intergral part of the equip, or they may be an external source that may feed several pieces of equip.

Dave

 

[duke37]

In this case, a highly accurate frequency standard is not necessary, just use the energising signal as the reference as you say.

You will need to get the phasing right.