Suche Hauptanteil des Rauschens im Vorverstärker

[mabra]

Hallo !

Weiß jemand, wo dieser Haupanteil liegt?
- Dunkelstrom der Fotdiode [oder diese allgemein]
- 2x 10 MOhm Widerstände
- FET

Ich suche noch 'nen Weg, evt. durch höherwertige
Bauteile den Rauschanteil zu reduzieren.

Danke!
++mabra

[roeliepoelie]

Dear Mabra,

Experiments show that the 20Mohm resistor is the main cause of noise. The presence of diodes only decrease the noise level. This can be seen in the following diagrams. They show the output signals of the second opamp. If the 20 Mohm is shorted, there is almost no noise (first diagram). If the 20 Mohm is not shorted, the noise level decreases if the number of diodes increases (the next three diagrams). This is due to their capacity.

The following calculations attempt to calculate the noise level without the presence of diodes.

The effective voltage U of the noise of resistor R is:
U = sqrt (4 k T df R).
In this equation, df is the band width of the noise. This is determined by the opamps and is approximately 10 kHz.
Thus we get
U = sqrt(4 x 1,4x10exp(-23) x 293 x 10exp(4) x 20x10exp(6) )
U = 65 mV
This is almost twice as high as the signal in the second diagram. At this moment, I cannot explain why.

Regards,
Roeliepoelie

[mabra]

Hi !

Much thanks for your investigation!
And sorry for my late reply, I was
unforeseen a little absent ...

To your analyzis: There migth be even
noise induced by current ... ?? I am not
clear from my reading in the internet.
But because of this, I'll buy a higher
quality resistor. Additionally, I'll see,
if a better transistor and OpAmp
may help in reducement.

I am mostly interested to measure
alpha radiation and I'll use the diode
SLCD-61N1, like Jekket recomends.

I'll need some time, but I'll come
back with my results.

Best regards,
++mabra