Lock-in Amplifier

[Krzysztow]

I couldn't find anything concerning my case, so I write it here...

I'm working with photon multiplier (Hamamatsu R928) - detecting a light signal over the scanning sample. Because the signal is really low I want to add lock-in amplifier (Standford Research Systems SR830). Unfortunately the signal from PMT is about 3V (constant value) modulated with changes of light intensity due to changes in scanning sample's surface. The problem is that lock-in amplifier's input is only +-1V. I tried to get rid of that "background" voltage by using scheme below, but it didn't work (and also don't know why).

Moreover I came accross the next problem, while I was trying on simpler light and PMT output was within a range. When I connected PMT to lock-in amplifier (input parameters 10MOhm/25pF) it had problems with detecting signal and phase lock it. But if I connected PMT through simple amplifier (Philips 1MOhm/20pF; output: 50 Ohm) with amplification 0dB the signal was OK. So, in principle, it works but I have to use amplifier, which does nothing. It's stupid. Can someone explain me the reason for that?

Thank You in advance,
Chris.

 

[crutschow]

I'm working with photon multiplier (Hamamatsu R928) - detecting a light signal over the scanning sample. Because the signal is really low I want to add lock-in amplifier (Standford Research Systems SR830). Unfortunately the signal from PMT is about 3V (constant value) modulated with changes of light intensity due to changes in scanning sample's surface. The problem is that lock-in amplifier's input is only +-1V. I tried to get rid of that "background" voltage by using scheme below, but it didn't work (and also don't know why).

Have you tried using a series capacitor to block the dc?

Moreover I came accross the next problem, while I was trying on simpler light and PMT output was within a range. When I connected PMT to lock-in amplifier (input parameters 10MOhm/25pF) it had problems with detecting signal and phase lock it. But if I connected PMT through simple amplifier (Philips 1MOhm/20pF; output: 50 Ohm) with amplification 0dB the signal was OK. So, in principle, it works but I have to use amplifier, which does nothing. It's stupid. Can someone explain me the reason for that?

Even though the amplifier was set at 0dB gain the actual gain would be 6dB if the 50 ohm output was driving a 10MOhm load. Generally 0dB gain is for a 50 ohm load. The amp actually has an internal gain of 2 (6dB). This is reduced by 1/2 for a 50 ohm load because of the voltage divider action between the 50 ohm output impedance and the 50 ohm load.

 

[Krzysztow]

I didn't want to use capacitor, since my mogulation signal changes sometimes not so fast, and then it would be also cut.

And to the second advice and explanation, when I saw two voltages (with amp and without) on oscilloscope (1MOhm and 10pF), there was no significant difference in them. So I'm not sure if this was a reason. Anyway thanks and maybe You have other explanations. (Today one person tried to tell me that this was because power supplier and PMT have their own ground levels. But I don't know if this is a reason, since they give voltage difference on the output. And also I was told, PMT should be considered as an ideal current source, not voltage)

 

[crutschow]

I didn't want to use capacitor, since my mogulation signal changes sometimes not so fast, and then it would be also cut.

What's not so fast? The lock-in amp has an input resistance of 10Mohms so you can get a very low frequency response with reasonable sized capacitors. For example 1uF will give a time constant of 10 seconds and a low frequency cutoff of .016Hz.

 

[Krzysztow]

OK, thank You. You are right. I'll try that this way. Probably it'll work. Thanks for help.