Foss Infraxact Manual
ISIscan Version 2.71 Details August 12, 2005 ISIscan version 2.70 contained two bugs which were fixed in this release. Thus 2.71 contains the same improvements listed below with the following 2 additions:. Abort crash.
FOSS E-shop for Chemical Analysis parts and consumables for kjeldahl, kjeltec, fibertec, soxtec, digestion blocks, fiastar, distillation, sample mills.
When analyzing samples with System II based instruments, using the Abort button at certain times would cause ISIscan™ to lock up. This has been fixed.
InfraXact™ Check cell. Version 2.7 uses a more rigorous identification of InfraXact™ cup types as identified by the RFID. This was done to prevent incorrect samples from being analyzed as a check cell. 2.7 identified the Check cell as an incorrect cup type.
The Check cell cup is now exclusively recognized as the correct cup type. Complete implemenation of the Check cell fix requires the use of a utility program to update the current check cell product after ISIscan™ version 2.71 is installed. Please see for details. ISIscan Version 2.70 Details ISIscan™ version 2.70 adds support for the InfraXact™ Pro at-line instrument as well as continuing support for the InfraXact™, FOSS XDS bench instruments and FOSS System II based instruments.
ISIscan 2.70 fixed bugs relative to version 2.31 The bugs below were present in ISIscan™ version 2.31 or before and are fixed the in the 2.70 update:. User access levels. Any changes made to the edit and/or delete for history tab do not take effect until after the user logs out or exits program and then comes back in. Prediction model profile, parameter tab, moisture correction button. For “Moisture correcting parameter”, program shows and allows operator to pick “Moisture”, even though there are not any prediction models that have Moisture in them. If using the hide results if outlier option, the view sample tab, sample details button, sample prediction tab still shows the results of any calculated or user entered variables.
(The NIR predictions are correctly hidden, though). Using an InfraXact instrument, if the operator tries to scan a cup with a product profile linked back to a different cup type, an error message will correctly be displayed of: 'The sample container loaded in the instrument does not match cup type specified for the selected product!' After the operator responds to this error message, another error message will be displayed of: 'Sample cup RFID tag not found. Scan aborted!' This second error message is not needed.
History tab requires that check cell product be named “CheckCell”. If it is not, the results will not be displayed on the History tab (results, control chart, and spectrum tabs are ok). When exporting the check cell spectra file, the created file is only a NIR file. It should be a.CHK file with the predicted results as lab data. Storing spectra on command.
Scan a sample with the history tab open. When finished, click “store spectra”. The History tab does not get the “stored spectra” icon until you close the history tab, go to another tab, and then come back to the history tab. Standardization wizard with an InfraXact instrument.
On page 2 of 5, click option 3. If click next and then back, radio buttons 1 and 3 are both highlighted. This is only a cosmetic problem, as option 3 is used if the operator continues. At the time of scanning, if the selected product profile is linked back to an operations profile that uses any of the continuous scanning optionsm the program should check to see if any prediction models linked to the selected product profile have sub-sample averaging turned on. If they do, the program should display an error message that states “Cannot do continuous scanning on a prediction model that uses sub-sample averaging.” and do not initiate the scan. Performance test with an InfraXact instrument, Peaks 1 and 2 delta calculation of average. If test is aborted, the program still used the total number of cycles to calculate the average instead of the actual number of cycles completed.
The Auto archiving routine is called even if the option is not turned on in the system profile. Parameter profile in new mode, the “Calculated formula:” field is grayed out because default “Parameter type:” is “Predicted correctable”. However, if open the Parameter tab and make a new parameter that is “Parameter type” “Calculated variable” and save it, if then open parameter tab again in New mode, default “Parameter type:” again will be “Predicted correctable”, but the “Calculated formula:” field is not grayed out. Problem does not happen if parameter tab is closed and reopened in between variables. If a Product profile has a mixture of NIR and calculated parameters, then the order of the parameters on the Product limits tab does not match the order specified in the parameter profile, display order tab. If a Product profile has a mixture of NIR and/or calculated parameters that total 20 or more, the Product limits tab will only display 16 parameters and the Outlier limits tab will only display 3 parameters.
With an InfraXact instrument, if the program fails to read/write the RFID as last step of the scan sequence the program will crash. With an InfraXact instrument, after a SPU reset an error message of “Scan failed. Please rescan sample” will be displayed. The program recovers fine from this hardware reset, but the instrument is left with a blinking red light. Initiating another scan resets the light, but ideally, after the operator clicks OK on the error message, the blinking red light should be turned off.
With an InfraXact instrument, if a scan was started when there was no sample cup on the instrument, the SPU would spin and eventually the program would display an error message of “RFID tag not found. Scan aborted.” At this point, to scan a sample the program must be exited and re-started. If not, the “RFID tag not found. Scan aborted” error message came up immediately after the scan was initiated for any subsequent scans and the SPU would no longer spin so the RFID tag could not be found. Using an ANP model but no encryption key on computer, when the monitor routine was run an error message of “You are not authorized to use this equation.” which was correct. However, afterward another error message of “The following errors were encountered: There were no samples predicted.
Foss Foodscan Lab
Insert lab values and try again.” would be displayed and this second message was not needed. When trying to use encrypted models (EQP, ANP, or RPP) with no encryption key on the computer, an error message of “You are not authorized to use this equation.” Would be displayed. This message is correct, but it would be displayed for every sample in the file. If an InfraXact RED file (developed with 1104-1840,4, snv & detrend, 1,4,4,1 that came from a full range InfraXact (570-1848)) was used in prediction on a 6500, the predicted values, GH, and NH were not correct. Auto-archive did not work properly if the archive method was set to “number of days to keep samples”.
In the System profile, any change to the 'Automatic sample import from cup or file' option would not take effect until the program was logged out of or exited, and then re-started. If InfraXact data was collected with the “System II bandpass normalization” standardization option turned on, the Monitor option would not use it when re-predicting the samples. If the Auto archive routine was run with options set to delete all samples from the database (regardless of which selection method was used), then an error message of 'error size!
= nod, dat file', followed by another error message of 'Attempt to scroll past end or before beginning of data.' Would be displayed, followed by a program crash.
If the Auto archive routine was run on a database where none of the samples had stored spectra, an error message of 'ERROR: No dat file present. Aborting Archive' would be displayed and the auto archive would not be performed. The Windows Operating System will not allow a folder name to contain the following characters: /:.? ' However, the ISIscan product tree does allow the use of these characters in Product group and Product profile names. So if the user used any of these characters in the names, the Auto archiving routine would not be able to create the archive folders to write the spectra files to, but would still purge the samples from the ISIscan database. This was fixed by having the Auto archive routine substitute a “-“ in place of any of the characters that cannot be used in folder names, if encountered. In the System profile, have the “Automatic sample import from cup or file.” option turned on because the operator was doing LIMS sample import.
While doing this, if the InfraXact lid was closed at the beginning of the sample scan, the program would try to get sample information from both the LIMS and the RFID tag, since the closing of the lid was the trigger for automatic import of RFID tag information, and the program would freeze. The program has been changed so that there are now radio buttons in the System profile where the operator must choose either importing from a file (LIMS) or from the cup (RFID tag). During continuous scanning, the Scan button was not grayed out. The Operator could click 'scan' multiple times and get multiple instances of continuous scanning during the same session. If done enough times, the program would crash.
The program was changed so that with continuous scanning during the pause period, the Scan button is changed to an Abort button. This allows the continuous scanning to be aborted during the pause period and the operator does not have to wait until the next scan occurs to press the Abort button. With InfraXact instrument spectra, the Auto archive routine did not do the InfraXact bandpass only standardization. The routine would display an error message during the archive and then archive the spectra unstandardized. This would happen whether the bandpass only standardization was used alone or in combination with product standardization.
Local prediction with a repeatability file frequently crashed the program. An option was added to the System profile to direct all error messages to an error log file instead of the screen.
This is especially useful when scanning in an online environment. Certain ANP models would cause the Monitor routine to crash. System II and InfraXact spectra exported 'Standardized' whenever the operator had logged in 'offline' were actually exported 'unstandardized'. Further complicating the problem, the spectra file had the Master serial # changed and the standardized flag set to 'YES', giving the outward appearance that the file was actually standardized when in fact it was not. If logged in “Online” with a device, then standardized spectra was exported correctly. In the Parameter profile, highlight a previously created parameter with a parameter type of calculated variable.
Next, press Edit, then New. Enter in a name (JUNK in this example), description, and check 'Force negatives to Zero', but do not highlight or change the parameter type (it should be Predictable correctable by default). Press Save and then highlight other parameters by simply selecting them, going back and forth between parameters.
Close the Parameter profile and open the Prediction model profile, browse in an equation file, open the Parameter tab and try to find the JUNK variable. It is not available because the Parameter type has been changed to 'calculated variable'. On the product tree, if a search was done on one of the five UDF fields an error message would be displayed and all samples from the product were moved to the selected samples folder instead of just those matching the search criteria.
If two ANP models were imported into ISIscan that used two different equation protection keys, ISIscan would only recognize one of the two keys so only one model could be used. The program could not use a calculated variable if there was another calculated variable in the formula.
The Prediction model was changed so that if a model is updated that had a slope and/or intercept value on any parameter, the operator must now choose whether they want to keep the previous slope and/or intercept values or have them reset to slope = 1 and intercept = 0. When and equation was updated in the Prediction model profile, if the input moisture basis and the output moisture basis were different, the profile would automatically change the output moisture basis to match the input moisture basis. This was done without any warning message being displayed to the operator. The profile has been changed so that no changes are made to the output moisture basis during an equation update. If wet weight / dry weight analysis was being used on an equation or local RED file, the DM variable could not be the first parameter in the external file or the program would display zero for the dry matter variable and would not do any moisture correction to any other variable(s). © FOSS 2005 All rights reserved.