3
\$\begingroup\$

I have made an RF amplifier based on the MiniCircuits PSA4-5043+. The circuit is very similar to the one in the evaluation board by MiniCircuits. The layout was changed a bit but not much, and I am using the suggested bias T. The main difference I would say is that in the input I don't have an RF line but rather a radiation detector installed very close to the amplifier, less than 1 cm away, like this:

enter image description here

which produces fast signals pulses. These radiation detectors I know very well because I have used a lot with a different amplifier.

Now, when I look at the signals in the output with the scope I get this:

enter image description here

The peak looks good, but there is this change in the baseline that I have indicated in red. Zooming out it looks like this:

enter image description here

i.e. a damped low frequency oscillation.The signals are all with different amplitude because of the different energies deposited by the particles in the sensor, but over time it ends up looking like this:

enter image description here

What could be the reason for this?

I am trying to simulate this behavior but I don't think it can be replicated with a linear circuit like this

enter image description here

because the kink in the signal at ~4 ns (see oscilloscope) looks non linear.

\$\endgroup\$
10
  • \$\begingroup\$ Where's your 50 ohm output load resistor? This will be a significant issue to signal shape and quite possibly the wavy line that follows. \$\endgroup\$
    – Andy aka
    Jan 13 at 19:40
  • \$\begingroup\$ It is the oscilloscope. \$\endgroup\$
    – user171780
    Jan 13 at 19:55
  • \$\begingroup\$ Is that a large-area diode? How much capacitance at zero DC bias? \$\endgroup\$
    – glen_geek
    Jan 13 at 20:02
  • \$\begingroup\$ Yes, it is a large area diode. The capacitance varies from device to device, at 0 V it should be around 300 pF and at working voltage around 3 pF roughly. \$\endgroup\$
    – user171780
    Jan 13 at 20:33
  • 2
    \$\begingroup\$ NB: I don't think that minicircuits are for "transient" inputs. You should "model" your detector. 10nH wiring should be "enough" for this behavior. \$\endgroup\$
    – Antonio51
    Jan 14 at 10:12

2 Answers 2

Reset to default
1
\$\begingroup\$

Here is a simulation (MiniCircuit modeling guess).
For "reference" ...

To be complete here is also the frequency response (not as good as minicircuits ... :-)

enter image description here

enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ Thank you, this looks promising. I added a new screenshot from the oscilloscope in which the two oscillatory components that you get in your simulation are visible (the high frequency and the low). I will try to implement your circuit here to experiment. The high frequency component I think is related to your L2. \$\endgroup\$
    – user171780
    Jan 14 at 13:22
0
\$\begingroup\$

It turned out to be a very large inductance in the bias tee (unknown value as it is not specified in the datasheet). I changed the bias tee by a discrete one made with a single inductor and a capacitor and I found that id the inductor is larger than 300 nH, this problem appears and is somehow proportional to the inductance, lower values than 300 nH make the amplifier to undershoot. With a 300 nH inductor it is the sweet spot where the pulses are amplified properly.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.