I've been working on a project recently, where I need to build a device which detects gamma radiation and can be used for gamma spectroscopy. I've done some research and found many interesting ways of achieving that but at the end I chose to go with X100-7 PIN Photodiode.

I'm more of a software guy, without much expertise in electronics so as a first step, I've decided to build&test a prototype of the measurement circuit. I've copied 95% of the design proposed by the diode's manufacturer - it can be found here (last page) and my design looks like that: PIN photodiode circuit

The only differences between my project and the design from the document are:

  • C5 in the feedback loop of the second OpAmp, someone told me it's a good idea to limit the bandwidth,
  • Used AD8639 instead of the ICs proposed by the manufacturer, simply because I've had them in my junk.

So, as you can see on the picture, I'm connecting the diode to J3. I'm powering the whole board with 15V from my regulated power supply - Zhaoxin RXN-1505D. The Jumper J4 is only there to be able to disconnect the second stage amplifier and easily measure the output of the first stage. Normally it stays closed.

So at the moment, all I'm getting on the output is a rectangular wave with Vmax equal to the voltage I'm powering the circuit with. What's more funny, the output stays the same even when there is no diode connected to the circuit.

enter image description here

I tried powering it up with a 9V battery but it stayed the same (wave had an amplitude of 9V).

Normal behavior of the circuit should result in a single voltage impulse every now and then when a gamma particle hits the photodiode, with an amplitude far below the Vcc.

I guess this behavior has something to do with noise - that's why I haven't made this prototype on a solderless breadboard but on a homemade PCB. I know it looks like a mess but my soldering skills aren't something I'm proud of and my hands tend to get shaky sometimes.

enter image description here

So my questions are:

  • Does any of you have an idea of what might be wrong here?
  • Is there anything I can do to further investigate the problem? If you'd like me to provide some waveforms from other parts of the circuit I'd be happy to oblige.
  • Is there any possible hot-fix that I can apply here to get rid of this behavior or should I use some completely different design?
  • \$\begingroup\$ It looks like things are unstable which could be due to bad layout, the op-amp replacement or any number of things. Try increasing your C4 and C5 to 10 or 20pF. Also, replacing the photodiode with a visible light one so you can test with lower bandwidth continuous signals for testing. Increasing the feedback capacitors makes things more stable but decreases bandwidth and if you're only occasionally getting gamma hits then the bandwidth might be decreased so low that it's too slow to react to the gamma which won't enlighten you to the issue. Maybe bypass C6 for testing for the same reason. \$\endgroup\$ – DKNguyen Jun 11 at 0:28
  • 1
    \$\begingroup\$ What I see is power line 50 Hz noise. So what you need to show is all the sources of stray 50 Hz in a photograph and method of earth grounding. short PD connections are critical and high impedance thus prone to antenna reception of electric fields. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jun 11 at 1:22
  • \$\begingroup\$ @DKNguyen "Increasing the feedback capacitors makes things more stable" - that is not generally correct - it's usually the other way round for a lot of op-amps unless the feedback is so swamped with capacitance that you form an integrator. \$\endgroup\$ – Andy aka Jun 11 at 8:07
  • \$\begingroup\$ @Andyaka I guess I should clarify that I wasn't speaking generally. I was more speaking for this very specific case where feedback capacitance is needed to stabilize the TIA. It's unstable with no feedback capacitance, and 1pF is next to nothing, so a little more should bump it towards being more stable if it's already unstable from not having enough capacitance. \$\endgroup\$ – DKNguyen Jun 11 at 21:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.