Cannot Dump Spectra from SciAps Analyzers #1
Labels
No Label
bug
duplicate
enhancement
help wanted
invalid
question
wontfix
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: spacecruft/pysalx#1
Loading…
Reference in New Issue
There is no content yet.
Delete Branch "%!s(<nil>)"
Deleting a branch is permanent. Although the deleted branch may exist for a short time before cleaning up, in most cases it CANNOT be undone. Continue?
Procedure:
Power on, log in, main menu.
Results
Three dots --> Templates Setup
New Template: "Raw"
This may be where I'm failing. Setup Fields (Checked):
Date
Test #
Sample ID
Grade Match <--- Probably can uncheck that, but doesn't matter (?)
Then a number of unchecked boxes below (e.g. lat/lon).
Chem Elements(32)
Formulas(0) Left at default, unchecked.
Arrange order. Default.
Spectra Format: Raw <--- This is what I think is the important one to set, but it doesn't appear to have an effect.
Uncertainty, Default unchecked.
Test Info, Default is unchecked. I have it checked. OK?
Save that template.
Then back at main menu, go to Analyze.
I've tried both Geochem/Element Pro, for this example I'll do Geochem. Does it matter?
I guess here I should be telling Geochem to use the template? Or is that after acquisition?
Take a sample. After sampling, it displays the table of elements' %.
Click the round graph icon and I can see the spectra display. That is the raw dump I'm trying to get.
Go back to top main menu --> Results.
At this point, I think I could just select the one I want to dump, but following your example from yesterday, go to the three dots "Filter by Date" and select today's date.
"Show 5 results" (or however many)
Click export/arrow icon.
Format: Excel.
Use default filename, overwrite file.
Choose Template: Raw <--- This seems like the important one, matching to template above.
Start Export: "Files have exported successfully"
View the exports files, and see there is the file:
So that appears to export ok.
That directory is empty. That is where I was expecting to find the exported file, but I've done something wrong. I've tried a few variations of the above.
One thing of note, if you do this:
Top main menu --> Results
Select any of the sample results.
Click the round spectra icon.
There is no spectra data displayed.
When this is viewed immediately after the sample is taken, the sample data is there. When this is viewed later, it isn't there. So it can't export it because it doesn't have it available for export.
This directory is also empty. It may store the spectra, pre-export (?).
Doing a sample finds the peaks ok, as can be seen by hitting the spectra icon immediately after taking a sample (but not later when viewing the sample). This is from the logs while taking a sample:
As a test, I set the Defaut template to
Raw
forSpectra Format
. This was similar to above. I could view the spectra immediately after taking a sample. But if I went to the top main menu, then toResults
and viewed it there, the sample was available, but the spectra wasn't. Doing an export showed this in the logs:I did another test sample to export spectra using the Element Pro application. I deleted all of the days samples and did a single sample for export. It exported the CSV, but not the spectra. The spectra also isn't visible on the device under
Results
when viewing the sample. From logs:This paper has some info about it:
Analyses of Li-Rich Minerals Using Handheld LIBS Tool
https://www.mdpi.com/2306-5729/6/6/68
It looks like the raw files are available here:
https://ordar.otelo.univ-lorraine.fr/record?id=10.24396/ORDAR-65
This is an excerpt of the CSV file:
Each file starts at one (e.g. isn't one huge multi-part file):
Q: What does each file represent? 16 sample parts?
NIST LIBS Database
https://physics.nist.gov/PhysRefData/ASD/LIBS/libs-form.html
This is one sample from the free dump linked above:
Each file has the same header:
16x
files by8264
lines is132,224
.What are the YUV files? Are they just for sample presence detection or are they the sample used for the creation of the CSV (e.g. YUV is pre-processed raw).
?
Edit: these are probably for sample presence detection.
This is a sample from all 16 CSV files combined, with one file per column. The first column,
Pixel Index
is the same in every file. The second column,wavelength
, is the same in every file. So just theintensity
column is different for each file.Calibration Curves for Commercial Copper and Aluminum Alloys Using Handheld LaserInduced Breakdown Spectroscopy
https://www.osti.gov/pages/servlets/purl/1427700
New API repo (?).
https://github.com/SciAps/SciApsApi
Links to:
https://github.com/SciAps/SciApsApi/blob/master/api/src/main/java/com/sciaps/ZAcquisitionResult.java
Contents:
To get remote web listening:
Go to top main menu
Swipe right
Remote Service
Should show IP/port.
It looks like the API isn't implemented though? Basic queries return
not implemented
.The device gives itself the static IP
192.168.42.129
. It still listens with the webserver on all ports. The device itself hascurl
on it. So you can run that directly on the device, using that static IP as a test as well (e.g. there's nothing else in between could cause an issue). That also gives501 Not Implemented
.The URL endpoint appears correct:
https://github.com/SciAps/SciApsApi/blob/master/client/app/src/main/java/com/sciaps/apidemo/SciApsClient.java#L59
I have the X-555 XRF model available now.
It works in a similar way to the LIBS model, but it slightly different. For setting up an export, it just has a checkbox to
Export Spectra
, it doesn't give an option for which type (at least not in the same spot as on the LIBS model).The X-555 does export the spectra as it should though. It is in
mca
format:Gah, it looks like that spectrum file is zero bytes though, so no data. :(
This is the error in the logs when the X-555 tries to export:
The default profile on the X-555 has
Export Spectra
enabled by default. It is exporting files of zero bytes:This is true when using both
Alloy
andSoil
applications. TheSoil
application has it's own template, which is also set toExport Spectra
by default. But it too exports zero byte files. Examples so far:Note, with the X-555 it does dispay the spectra properly both after acquiring the sample, and when viewing it in
Results
. The Z-903 won't display the spectra correctly inResults
nor in export. So they have slightly different problems.The X-555 gives the same
501 Not Implemented
when using the API. It is able to get an IP via wifi and start the service on port8080
, which answers, but always501
.The X-555 also apparently statically assigns itself the address
192.168.42.129
. This could be a potential conflict on that IP if both units/models are used at the same time on the same network.Looks like the
192.168.42.129
is inAirdroid.odex
.In the file
extsdcard/db/maindb.p
in theSoil
and theEmpirical
analysisMode
thesaveAcquisition
field is set tofalse
.In that same directory there are other main files, and a sqlite database:
The
UsersAndPermissions.db
is standard sqlite3 files. I'm not sure the format of the threemaindb
files, but they have the settings above.The
ExportBatchMCA.java
andExportMCA.java
functions are in/system/app/XRFAndroid/arm64/XRFAndroid.odex
, which looks to be the main app. There are also a fewExportCsv*
functions. On the X-555 nothing forraw
, so that must just be Z-903.Other SciAps customers see same issue.
No word from Sciaps.
New tech support agent from SciAps called and said the export needs to be enabled in the
Factory Settings
. Those are greyed out in the menu and "inaccessible". I need to sign a waver before getting the code to enable changingFactory Settings
to allow export. Awaiting document from SciAps.SciAps:
Sent ACK.
I have user/PIN.
Logged in with the factory PIN provided to the SciAps X-555.
Go to
Settings
.There's many more options. There are no new options under
Safety Settings
. Nothing related to exports nor spectra.Under
Factory Settings
, there's temp & time settings, nothing related to exports nor spectra.Database Management
, nothing related to exports nor spectra.Application Config
, which is new, has a few options, nothing related to exports nor spectra.Camera Settings
, nothing related to exports nor spectra.GPS Connection
, nothing related to exports nor spectra.Test Stand Settings
, nothing related to exports nor spectra.So it doesn't appear to fix the SciAps X-555.
With the factory user/pin, the SciAps Z-903 has under
Settings
,Factory Settings
,Save Spectra
. It was unchecked. Checked now. Will log out and test as regular user.It looks the Z-903 is now exporting since
Save Spectra
fromFactory Settings
was enabled.Cannot Dump Spectrato Cannot Dump Spectra from SciAps AnalyzersThis was sent to me March 11th, 2022, but there was never any followup.
I spoke with Jonathan today. He's going to read up, talk to software, and give me a call tomorrow.
From SciAps:
SciAps requested the
xrf.log
file, which I sent. They said they need more time to analyze it and will call/email.From SciAps:
Running:
It did the update and rebooted. Calibration now appears to work.
It still exports zero byte file:
SciAps writes:
Running:
Pushed hardware.cfg to sciap's forlder on SD card.
Power off.
Power on.
Click Analyze.
Click Alloy.
Gives error:
Options from SciAps to consider:
SciAps has the device, but never sent an email/call to confirm they received it. I called Jonathan Moller today. He said they had received the device. I spoke with Keith Bowman who said it just required their proprietary Windows software to do the update and it should take 30 minutes, but he also said it was more complicated than that. Bowman also said other customers were complaining about the lack of response from their customer service.
Mike Fuller emails just now saying it will be shipped overnight today or Monday.
David Bowman says device is shipping today for delivery tomorrow.
Received X555 unit. Looks ok. I haven't plugged it it. The device is cold, waiting for room temp & time.
I received the unit.
I did "Analyze". I did an export. It exported an
.mca
file. The file just had one column, I thought it was supposed to have two.Excerpt from
.mca
The device takes 2048 samples each analysis. See attached spreadsheet column provided by SciAps for static values to match with MCA dumps.