Hi all,
Just had word from Marek yesterday that PRA version 22 is out and there seems to be a few changes..
The biggest change is the increase from 16 to 60 sample points, yes first I thought it was crazy, because at 48 kHz sampling, 60 samples would take 1250 µs and the dead time would go through the roof, but you simply increase the software sample rate to 384 kHz to bring it down to a more reasonable 125 µs. Marek claims that this change gives better results, I have yet to confirm this.
I find the pulse shape editor a little bit clumsy with all the tick boxes, but I guess it offers more flexibility.
I was also surprised to see that Marek included my recent suggestion to add energy per channel, it is called "Sum Quantity", I'm not sure about the name, but it works well.
Please give it some testing and provide your feedback for Marek below.
Steven
PRA 22 Hold on to your seats..
- Sesselmann
- Posts: 926
- Joined: 27 Apr 2015, 11:40
- Location: Sydney
- Contact:
PRA 22 Hold on to your seats..
Steven Sesselmann | Sydney | Australia | https://gammaspectacular.com | https://beejewel.com.au | https://www.researchgate.net/profile/Steven-Sesselmann
Re: PRA 22 Hold on to your seats..
Interesting, I'm excited to try the sum quantity setting.
Geoff Van Horn
Former Alaskan living in rural Wisconsin
Former Alaskan living in rural Wisconsin
Re: PRA 22 Hold on to your seats..
Hi,
Can someone explain what is the best metod for pulse shape acquisition?
Next biggest step imho could be the centroid peak tracking to compensate for temperature drift with very long integration time.
I'll try ASAP this release.
Thanks to Marek!
Can someone explain what is the best metod for pulse shape acquisition?
Next biggest step imho could be the centroid peak tracking to compensate for temperature drift with very long integration time.
I'll try ASAP this release.
Thanks to Marek!
Ciro
- Sesselmann
- Posts: 926
- Joined: 27 Apr 2015, 11:40
- Location: Sydney
- Contact:
Re: PRA 22 Hold on to your seats..
Ciro,
The first thing I suggest, is going to Settings >> Audio interphase and setting up your sampling parameters.
Suggested: 384 kHz - 16 Bits - left Channel
In the latest version of PRA 22 you may as well go for the highest sample rate 384 kHz even if your sound card is slower, the software does a good job of interpolating the samples.
Next, I think it is important to know the pulse direction, pulse length and pulse height from your detector and spectrometer setup.
The pulse direction is easily identified from the View >> Audio Input window (it is like a simple oscilloscope).
Then run the software with "Shape Tolerance method" un-ticked. and look at View >> Pulse Width Histogram to identify and adjust your mean pulse width.
Once you have determined the pulse width, say for example 75 µs. you can open the Settings >> Pulse shape editor and adjust the number of samples you want to use.
i.e. say you are sampling at 384 kHz then;
\[ (1 sec /384000 kHz = 2.6 µs) \]
So there will be 2.6 µs between each sample, so to capture the 75 µs pulse you will need a minimum of 30 samples to capture your pulse , and I would allow for a bit of space either side and round it up to 40 samples.
PRA now has up to 60 samples and the pulse peak always sits in the middle at around the 30th. sample, so if you want to sample with 40 samples, you need to untick samples 1 to 10 and 50 to 60, leaving the samples 10 to 50 ticked, giving you a window of 2.6 * 40 = 104 µs. wide.
Keep in mind that the sample window size affects your dead time, so with a sample window of 100 µs the maximum count rate will be around 3000 cps.
There may be a more scientific way to calculate this, but as a rough estimate I use;
\[ Max Count Rate = (1 sec) / (window size) * 3 \]
If you need to push your count rate higher, you can narrow down the sample window.
Once you are happy with the settings you can tick the Use shape tolerance box and start recording.
Hope that makes sense..
Steven
PS: Latex Math was used in this post.
It's easy just start and end your expressions
with backslash squarebracket like this;
(ignore the Latex button in the editor.) looks like this..
\[ a^2 + b^2 = c^2 \]
The first thing I suggest, is going to Settings >> Audio interphase and setting up your sampling parameters.
Suggested: 384 kHz - 16 Bits - left Channel
In the latest version of PRA 22 you may as well go for the highest sample rate 384 kHz even if your sound card is slower, the software does a good job of interpolating the samples.
Next, I think it is important to know the pulse direction, pulse length and pulse height from your detector and spectrometer setup.
The pulse direction is easily identified from the View >> Audio Input window (it is like a simple oscilloscope).
Then run the software with "Shape Tolerance method" un-ticked. and look at View >> Pulse Width Histogram to identify and adjust your mean pulse width.
Once you have determined the pulse width, say for example 75 µs. you can open the Settings >> Pulse shape editor and adjust the number of samples you want to use.
i.e. say you are sampling at 384 kHz then;
\[ (1 sec /384000 kHz = 2.6 µs) \]
So there will be 2.6 µs between each sample, so to capture the 75 µs pulse you will need a minimum of 30 samples to capture your pulse , and I would allow for a bit of space either side and round it up to 40 samples.
PRA now has up to 60 samples and the pulse peak always sits in the middle at around the 30th. sample, so if you want to sample with 40 samples, you need to untick samples 1 to 10 and 50 to 60, leaving the samples 10 to 50 ticked, giving you a window of 2.6 * 40 = 104 µs. wide.
Keep in mind that the sample window size affects your dead time, so with a sample window of 100 µs the maximum count rate will be around 3000 cps.
There may be a more scientific way to calculate this, but as a rough estimate I use;
\[ Max Count Rate = (1 sec) / (window size) * 3 \]
If you need to push your count rate higher, you can narrow down the sample window.
Once you are happy with the settings you can tick the Use shape tolerance box and start recording.
Hope that makes sense..
Steven
PS: Latex Math was used in this post.
It's easy just start and end your expressions
with backslash squarebracket like this;
(ignore the Latex button in the editor.) looks like this..
\[ a^2 + b^2 = c^2 \]
Steven Sesselmann | Sydney | Australia | https://gammaspectacular.com | https://beejewel.com.au | https://www.researchgate.net/profile/Steven-Sesselmann
Re: PRA 22 Hold on to your seats..
Steven,
I've just tried the new version, but there's something I don't understand.
I followed your suggestions, I put 384 kHz - 16 Bits - left Channel, then I unticked samples 1 to 10 and 50 to 60, but look at the spectrum attached, it's quite strange isn't it? The source is fukushima's soil. Moreover, is it normal that during the counting, at every step (I used 2 seconds), PRA writes a short log on the comments field?
Thank you
I've just tried the new version, but there's something I don't understand.
I followed your suggestions, I put 384 kHz - 16 Bits - left Channel, then I unticked samples 1 to 10 and 50 to 60, but look at the spectrum attached, it's quite strange isn't it? The source is fukushima's soil. Moreover, is it normal that during the counting, at every step (I used 2 seconds), PRA writes a short log on the comments field?
Thank you
Daniel, Italy
- Sesselmann
- Posts: 926
- Joined: 27 Apr 2015, 11:40
- Location: Sydney
- Contact:
Re: PRA 22 Hold on to your seats..
Ciro,
Wow, that's fantastic... what settings did you use?
Steven
Wow, that's fantastic... what settings did you use?
Steven
Steven Sesselmann | Sydney | Australia | https://gammaspectacular.com | https://beejewel.com.au | https://www.researchgate.net/profile/Steven-Sesselmann
Re: PRA 22 Hold on to your seats..
Hi Steven,
nothing special about the settings, i run normal pulse shape acquisition and then narrowed a little bit the samples window as per your's suggestion, only difference i'm running a 24bit sound card so i leaved 384Khz 24Bits instead of 16Bits.
nothing special about the settings, i run normal pulse shape acquisition and then narrowed a little bit the samples window as per your's suggestion, only difference i'm running a 24bit sound card so i leaved 384Khz 24Bits instead of 16Bits.
Ciro
Who is online
Users browsing this forum: No registered users and 1 guest