Redpitaya Forum

the quoted 125Msps is only possible under what conditions?

The ADC on the Red Pitaya 125-14 board is always running at 125 MSPS. However, the bandwidth of the input signals is limited by other factors:

• Whittaker–Nyquist–Kotelnikov–Shannon sampling theorem
• bandwidth of circuits in front of ADC input
• application's requirements on the noise performance

I'll have to pull data from the Rpitaya somehow.

The control programs that I mentioned in one of my previous comments all work on a remote PC and communicate with the MCPHA application running on the Red Pitaya board via Ethernet or Wi-Fi. These control programs pull the data from the Red Pitaya board and save them to a file on a remote PC.

I think that both the MCPHA application running on the Red Pitaya board and the control programs can be modified to communicate with other devices via other interfaces.

I will probably look at the Micod PMT charge sensitive-preamp

If you're going to buy Micod PMT u-CSA, then maybe you could also buy their shaping amplifier (MS-25)?

I've just checked the MS-25 specifications. The output pulse duration is around 25us. It seems that the duration and shape of the pulses correspond perfectly to the pulse processing capabilities of the MCPHA application.

The MS-25 seems to be for SiPMs --- at least that is what they are being used for...
I have forwarded an email to the company to get further info on charge shaping amplifiers specific to PMT use. It seems to me that the PMT-uCSA combines both functions but I am not sure how much pulse stretching is possible on the this particular unit.

I am aware of the Whittaker–Nyquist–Kotelnikov–Shannon sampling theorem as well as the
bandwidth of circuits in front of ADC input.

My misunderstanding is the sampling rate in regards to the circuitry in front of the ADC--I was unclear in that regards.
Despite the limitations I still think there is much potential in creating a PMT based gamma spectrometer that could use multiple scintillation mediums.

The MS-25 seems to be for SiPMs

Since MS-25 isn't directly connected to the detector output but to the output of a charge-sensitive amplifier, I don't see why it won't work with PMT u-CSA.

It seems to me that the PMT-uCSA combines both functions

According to the schematics in the PMT u-CSA datasheet, it's just a charge-sensitive amplifier and there is no shaping amplifier inside. In one of the posts linked from one of my previous comments (here is the link), PMT u-CSA is used with a shaping amplifier and all the oscillograms look OK.

BTW. Since you previously mentioned T-bias splitter. Figure 5 in the PMT u-CSA datasheet shows how to power a PMT with a negative HV power supply and without T-bias splitter. I'd say it's a safer way to power a PMT and you won't need a HV capacitor.

The CSA-25 is available as the Charge-sensitive amplifier (a-CSA) & Shaper amplifier KIT for SiPM detector, which they have confirmed however,

The PM-uCSA needs to be used in this case and the company has sent me a response followup regarding my question. The key is to have pulse processing as you've suggested in combo with the amplifier.

They may have a product that would work...

Thing is that I already have a HV power supply 0-2Kv with extremely low ripple that is geared towards Gamma spectroscopy. I'd like to use this first. IF I need to spend more I'll do so later.

pavel wrote: ↑

Mon May 11, 2020 12:05 pm

The MS-25 seems to be for SiPMs

Since MS-25 isn't directly connected to the detector output but to the output of a charge-sensitive amplifier, I don't see why it won't work with PMT u-CSA.

It seems to me that the PMT-uCSA combines both functions

According to the schematics in the PMT u-CSA datasheet, it's just a charge-sensitive amplifier and there is no shaping amplifier inside. In one of the posts linked from one of my previous comments (here is the link), PMT u-CSA is used with a shaping amplifier and all the oscillograms look OK.

BTW. Since you previously mentioned T-bias splitter. Figure 5 in the PMT u-CSA datasheet shows how to power a PMT with a negative HV power supply and without T-bias splitter. I'd say it's a safer way to power a PMT and you won't need a HV capacitor.

I just got info back from the company regarding the PMT-uCSA ( PMT pulse amplifier) and MS-25 (shaping amplifier).... while you are correct both of these units can work together, both units need to ideally be on one PCB.. The company has an EVA board (not available through their website) that fortunately combines both the uCSA and MS25 onto a single board. They cautioned that for efficient operation, it is necessary to select the external feedback, experimentally choose the value of the capacitor and resistor otherwise the gain of the uCSA will be too high for the MS25.

So... there we go.


btw, thank you for your input. It has helped me a great deal!

Here are the negative pulses generated from my Scionix NaI/Tl scintillator.
Graph 1 and 2 show the pulses from a 5uCi Cs137 disk source and the third graph showing Cs137 (5uCi) and Mn54 (4.5uCi)

The time scale is also different as well.. combined isotopes is at 200uS per division.



Next will be to try out the MCA app.

Here is the experiment utilizing one of 2 MCA applications for the Red Pitaya.

HV was obtained from a Gamma Spectacular at 550V and the signal was split using a Bias T.
Signal was fed into input 1 of the Redpitaya 125-14.

Notice the blue logarithmic spectrum forming in the browser window. This spectrum is of 2 isotopes
Cs137 and Mn54 both are nearly 5uCi each. The complete spectrum was obtained in roughly 20-30 seconds. The basic shape of both gamma energies was resolved in 5 seconds.

While the acquisition of the signal was pretty straight forward, there was some odd behaviour from this particular app. The app appears only to display pulse heights logarithmically and will not display exponentially. There is also no energy calibration field nor temperature adjustment... Bins are limited to either 1024 or 16656 samples. -- more typical is 2048 and 4096bins



Application settings :

Pulse IN1 = negative
32nS sampling rate.
1024bins sample
Deadtime: 0
30minute sample time for Th232 (12g of 99.98% Thorium Dioxide)
vs 1-5 minutes for Cs137 and Mn54

Alright, while the web browser application might not have displayed the exponential scale at all,
here is the exported data. As you can see the detail of the spectrum is quite excellent.



the .CSV file generated was put into Interspec for processing gamma energies.

The spectrum shown here is typical of Th232 progeny (99.98% Thorium Dioxide)