Preamplifier for photomultiplier

[micard]

Dear friends,

I was given the task to design the amplifier for the signal from photomultiplier. I need to detect 1 uA amplitude, 6 us long current pulses of gaussian shape, on top of the 40uA dark current (due to the ambient light, that I cannot get rid of at the moment) and the output needs to be in range +/- 5V.

The requirements sets the gain to 50 V/mA. I used the topology of transimpedance amplifier (see attached schematics). Unfortunately my design still suffer from high noise.

Could you help me to improve the design to minimize the noise?

I am using Multisim to simulate the behavior of my design, but I am completely new to it. Could you recommend me, how could I simulate such a noise as I have in it, to investigate the ways how to reduce it?

Best regards,

Micard

 

[keith1200rs]

It would be useful to know what noise you are measuring. Also, is this measured or simulated results you are concerned about?

A few comments:

1. Ground the non-inverting input. Adding a resistor simply increases the noise

2. 47k and 1pF gives you a far higher bandwidth than you need and hence a far higher noise.

You might want to add some protection diodes for the opamp input.

Keith

 

[micard]

Thanks for suggestions.
The noise i have is in measured system - therefore there was my question as well - how to simulate it.

I've made a mistake in schematics putting there 1pF - actually it has 10p - as i was advised. I was said, that feedback capacitance will interfere with the capacitance of PMT, therefore 10p should be "fairly good" value.

Would you have some other suggestions?

Micard

 

[keith1200rs]

I can look in detail at the problem tomorrow. What is the PMT capacitance? The is a point where the noise increases with frequency due to the PMT capacitance and feedback capacitance & input capacitance.

Also, have you some specific numbers you have measured? E.g. total output noise?

I don't use Multisim but you should be able to do a noise analysis. Model the PMT as a current source with capacitance across it and a leakage resistance.

Keith

 

[keith1200rs]

I have done a quick simulation of your circuit (without knowing the PMT you are using) and I think if the circuit is noisy then it is unlikely to be the amplifier design that is the cause. It will depend on the PMT capacitance, but even at 100pF it only simulates at 52uV over 1Hz to 1MHz. The noise from the 47k will be 28uV at room temperature over the same bandwidth.

I would suggest that you disconnect the PMT and see what your noise is. If it drops to very low levels then the noise is coming from the PMT. Otherwise you have a construction problem.

One problem with the ambient light is it will create noise. I have not used PMTs, just avalanche photodiodes, but with APDs there ambient light adds to the noise due to avalanche effects. I assume there is a similar effect in PMTs. It should be easy to find out if that is the cause.

Keith.

 

[micard]

Thank you very much!
I will just check how it behaves without PMT.