Electronics > Projects, Designs, and Technical Stuff

High Speed Transimpedance Amplifier

<< < (5/6) > >>

StillTrying:
"Also, the lab at present uses the amplifier 6950ds(attached) from Phillips Scientific. I need a better performance than this..."
"Input : One; 50 ohms"
"Wideband Noise : Less than 25 μVolts RMS; Referred to the input. (1.5nV/ÖHz.)"

"Even after this, the current is between 1 to 5 pA"

I can't see how your 1 to 5pA could be correct as input to that amp.

arivalagan13:
The stage previous to the amplifier is the microchannel plate which generates a current..as shown in the graph...and I need a TIA...The microchannel plate generates current in the range of 1 to 5pA....

Did I answered your question or I'm mistaken?

Marco:

--- Quote from: arivalagan13 on June 16, 2019, 04:22:35 pm ---Did I answered your question or I'm mistaken?

--- End quote ---

You didn't address Alex Nikitin's post ... you will never see 5 pA unless you average over some huge period like a second. With a high bandwidth amplifier you will always see massively higher currents because the electrons arrive in bunches, so worrying about noise is silly when working at high bandwidth. You just need enough SNR to distinguish one event from multiple near simultaneous events.

StillTrying:

--- Quote from: arivalagan13 on June 16, 2019, 04:22:35 pm ---Did I answered your question or I'm mistaken?
--- End quote ---

Yes, you confirmed only 1-5pA.

It looks to me that software records the time that the a 1-5pA pulses arrive, and their amplitude is not that important, which woud help a little bit but not much.
But, I don't much about currents that low, or how these types of things work, or how the signal is extracted from the noise.
https://en.wikipedia.org/wiki/Time-of-flight_mass_spectrometry

I can't see how the OPA656's 1*500MHz could ever improve on the Phillips's 10*300MHz.

But hopefully someone will have a much better idea than me. :)

Marco:

--- Quote from: arivalagan13 on June 16, 2019, 03:15:04 pm ---Also, the lab at present uses the amplifier 6950ds

--- End quote ---

This isn't a transimpedance amplifier by the way and I don't think you will be able to use one at the moment (ie. with the coax connector). Where you do you even measure 5pA? You're never going to be dealing with currents at the end of a piece of coax at these frequencies ... somewhere deep inside the measurement device there's current, then it goes into coax and from that point on all you have to worry about is the voltage on the 50 Ohm termination.

Lets say you get 10000 electrons in a single event from a single stage MCP, lets say the time domain response of your digitizer is 5 ns, so the average current is 10000*1.6e-19/5e-9 =~ 0.3 uA. Times 50 Ohm is 15 uV. The Johnson noise for 50 Ohm over 100 MHz is ~ 6 uV ... so this is kinda pushing it. The Philips amplifier is indeed not good enough to detect a single event within these specifications. A TIA would help, but you can't connect that through coax in this case, because the necessary bandwidth is way too high for a piece of improperly terminated coax. A TIA needs to be at mm distance to the MCP.

How fast is your digitizer BTW? It's rather relevant ...

PS. when I say TIA I actually mean 10k resistor and voltage buffer, I doubt the MCP cares much about the load resistance.
PPS. oh that's not right, I underestimated the capacitance of the MCP ... guess you'll need a proper TIA.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod