0

Calibrator Drift - How to Incorporate it into uncertainty calculations in Met/Cal?

I don't know if this has been discussed previously, but I would like to know how to incorporate calibrator drift for uncertainty calculations into a Met/Cal procedure. (Calibrator: reference or standard; it can be a calibrator or reference multimeter. Drift: the change in specification over time)

The 5522A specifications reads under Absolute Uncertainty Definition as follows "The 5522A specifications include stability, temperature coefficient,linearity, line and load regulation, and the traceability of the external standards used for calibration."

Does "stability" means drift? If it doesn't, then how do I incorporate drift in the uncertainty calculations? Can it be added yearly to the calibrator's accuracy file and if so, how?

8 comments

Date Votes
0
Avatar
Patrick Butler

First of all you need to use the VSET FSC in order to populate additional uncertainty parameters.These can be populated directly by the procedure writer if the value is known at the time the procedure is written. Otherwise, you may prefer to use variables to populate them dynamicaly at the time of execution by simply asking the operator for input; using the LIB FSC to grab data from spreadsheetsuse; or using the MATH function "INV" or "CAL" to grab data from the database. You can also modify ACC files. MATH functions allow you to read the ACC files too.

0
Avatar
James Smith

We primarily account for similar occurances through the accuracy files. You can generate an accuracy file per asset number and compensate for what factors you need for that particular asset only. For example we have a very long history on some equipment and justify altering the specification for those particular assets. Here is an example of how to edit the accuracy file so that when metcal pulls data from the accuracy fiels, it pulls from that asseted file. If an asseted file does not exist it pulls the typical file. So it is important to make a new file for each asset and maintain them manually. In this example we have 3 accuracy files for 732A. Two for the assets with a long history and one original for all other 732A. 

 

Begin Header                                                                                                                                                instrument = Fluke 732A                                                                                                                    interval   = 365 days                                                                                                                            confidence = 2.00 sigma                                                                                                                   asset = 9084                                                                                                                            End Header 

 

I suppose it is possible to account for drift in many ways, this is just one way. This route your either going to have to update alot of accuracy files manually, or insert a section per file for date and edit your procedure to do what ever math function based of that date value. I am not versed in the LIB FSC, so I am not currently able to provide any examples. If the date value is to be used within the procedure you would have to either use the VSET (F1 to go read the help file) as Pat has suggested or do the math manually and write in an ACC line.Hope this helps.

0
Avatar
Khaldoun Al-Jazzazi

Dear Mr.James Smith,

Can you explain how to create/write/generate/Mapping accuracy file in to the MET/CAL for the particular asset no. and How to address the traceability information and other contributors such as (U3, U4...U10) in to accuracy files.

If possible can you send a sample accuracy file to understand the protocol.

Thank you,

 

0
Avatar
James Smith

Your accuracy files are located in the location they are mapped to. Locate this place by going to windows explorer and addressing %programdata% then go to the fluke folder. Inside that folder should be a metcal.ini. Inside the INI file there is a line named accdir. This is the location your MetCal is pulling the accuracy files from. 

 

As far as creating an accuracy file, it is not very difficult. Open an accuracy file that already exists. Open it in Excel. This will give you a basic format of how the file should look. It will take some tweaking to align everything and possibly make the file function correctly. If you read through the accuracy file you should be able to see how MetCal is using this file. 

 

Inside the header information you can see there are a couple line of information that direct MetCal to the file. There should be an instrument line, interval, and a confidence line. Before the end header statement, enter a new line and insert the line "Asset = x" x being whatever your asset number is. It will compare this asset number to your database. If during runtime you have this particular asset number, then MetCal will pull and use this ACC file. If you do not then it will pull the ACC file that has no asset line entered. When you are done, you should have multiple accuracy files for the same model if you are using the asset line. I do not suggest getting rid of the original accuracy file that does not contain an asset line. You can then go through and modify the file however you want. You can delete all the information and save it as a blank acc file that you can use later on. 

 

Once you have your new accuracy file as you like it, save it as a tab delimited text file. Open it in Notepad ++. Review the file and ensure that it looks correct and makes sense. If everything is the way you like it, then save it as X.ACC. X being whatever you want to name it. It is highly suggested to test and debug this file. Make sure everything works correct and calculates as you want it. If you do not do this you will probably find yourself in an odd predicament later on. 

 

I am not able to give you any copies of any files that we may have. Hopefully this serves as a rough enough outline for you to figure out how to do what you are looking for.

 

As for your additional contributors, you can manage this is a few different ways. You can write in additional sections of your accuracy file and then do math functions within the procedure to use these additional parameters you put in. Or you can directly alter the values of the accruacy file to be current to the values you are looking for. 

 

Another method is to save them as a secondary file that is not an accuracy file, but rather just a generic text file. You can then pull them from that file using math and file functions. If you use this method you can change a single file and have it directly affect the procedure you want.  To do this I would do something similar as to how I explained accuracy file creation. You can manage the data via excel then save it to text files. If you want to see a good example of how fluke already uses this method, look into the user configured instruments. They do a pretty good job of using this file structure and call method.

 

Another method is to directly write the information into the procedure. Personally I am not a big fan of hardcoding everything into the procedure. If you do this, you will have to go back and change the code of the procedure every time you want to update the information. If you use a file look up function method instead, you can essentially write the code one time and then pull the information from the files that you maintain separately. It’s up to you to choose the most appropriate method for you. 

 

Hopefully this helps. 

0
Avatar
Khaldoun Al-Jazzazi

Thanks Mr.James Smith,.

Let me workout and come back to you.

 

 

0
Avatar
Khaldoun Al-Jazzazi

Dear Mr.James,

Can you help me how to map the file Asset.acc to the procedure in order to populate the contributors.

We have modified the existing accuracy file by adding the rest of the contributors such as U2, U3...U10 and assigned with the unique asset no.

Now we don't know how to map the accuracy file to procedure and what is the procedure to be followed thereafter

please explain me with the sample coding using the math and file function.

Thank you for understanding and cooperation.

0
Avatar
James Smith

The file location will be mapped through MetCal.INI. That would be the file location I explained earlier. You can have multiples of accuracy files for the same model both asset specific, and general both with interval specific. When MetCal runs it will pull the accuracy file from the location in your INI file. If there are multiples with the same instrument tittle and interval it will look at the asset number. If there is not an asset number that matches in one of the files, it will pull the generic file. The generic file is the one that does not have an asset number in it. 

 

There are a couple different ways you can have MetCal use the data in these files. One method being the built in FSC. Example, 5520 FSC. This is flukes built in method. It will only do what it is programmed to do and you cannot really alter the way that it works. You can use the values it gives you and work with them via Math FSC. You can see what value it is using by doing something like 

MATH L[1] = REFACC(). Type that in immediately after 5520 FSC and go read the help file by pressing F1.

 

A different method that is much more user controlled is to use file look up and math functions. If you are not familiar with what you are doing it will be very easy to make a mistake. If you know what you are doing it gives you greater control of the calculated values. Example, ACCV math function. This will go into the accuracy file and extract the specific information you want.

MATH L[1] = ACCV("instrument","mode",value)

This may not be the best example, or something you would actually do, it is only an example. If you copy and paste it may or may not work, I have not debugged it. This is just to get you pointed in the correct direction.

math L[1] = ACCV("Fluke 5520A","33V",24)

math L[2] = ACCV("Fluke 5520A","3A",2.6)

math L[3] = RSS_L(1,2)

Again, if you do not know what you are doing this may not be the best route to go. It is easy to miscaluculate the values. 

0
Avatar
James Smith

That and if you are trying to populate additional parameters that MetCal uses such as U2 ... Or any other parameters, you can set them by using the VSET or TSET FSC. Think V for permenant or T for temporary. the help file will give you pretty decent guidance on how they work. 

Please sign in to leave a comment.