Page 1 of 1

new version of dynamic time over threshold (work in progress)

Posted: 18 Feb 2019, 05:54
by miree
Hello everybody,

week-end is ending and I wanted to show my progress in improving my dynamic time over threshold device that I've shown previously: ... =366#p1735

Some points that I'm trying to improve:
The old device needs a third party FPGA board (Mojo FPGA board), but I want everything on one PCB.
The USB connection of that FPGA board is a bit slow and the event rate is limited to ~2000/s.
The old inputs have 33kohms termination resistors on the inputs to make the signals longer.
The old TDC resolution was 5ns, which is not too bad but I'm trying to get something around 1ns.

Here is a picture of the functional PCB with just one of four front ends soldered
It can send up to ~200000 events per second, has 50 ohm input impedance and ~1.5ns TDC resolution.
It is all on one PCB (made this with toner transfer method), and the values for the threshold are adjustable over USB.

Here is the first spectrum from a soil sample.
SoilSample.png (18.68 KiB) Viewed 741 times
The resolution is not perfect. The detector has 6% and the spectrum here shows only ~8.5%.
It also looks a bit noisy... anyway this is just the first working version, and I thought it is worth sharing.


Re: new version of dynamic time over threshold (work in progress)

Posted: 18 Feb 2019, 16:44
by Steven Sesselmann

Nice effort, great to see that you are making progress.

I think for your TOT method it is extra important to reduce any ripple and noise, so you can lower the threshold close to the baseline. You might also need some baseline stabilisation unless you can somehow float your threshold.

With our software MCA we do not need to worry too much about baseline drift as the pulse rise/fall is measured relative to the baseline, wherever it is.