Can anyone give example of uncertainty budget calculation in microsoft excel? Thank you. mail:vadim.popko@gmail.com
www.nrc-cnrc.gc.ca/obj/doc/solutions-solutions/advisory-consultatifs/clas-clas/uncertainty_budget_template.xls
Here is a paper we wrote for Quality Magazine.
http://digital.bnpmedia.com/publication/?i=273926&p=36
This is how we are going to start building our uncertainty budgets.
The problem is once you get a budget built.. How to you use it and call it from all your software packages.
Mike.
Quote from: CalLabSolutions on 10-29-2015 -- 20:52:33
Here is a paper we wrote for Quality Magazine.
http://digital.bnpmedia.com/publication/?i=273926&p=36
This is how we are going to start building our uncertainty budgets.
The problem is once you get a budget built.. How to you use it and call it from all your software packages.
Mike.
and how i can use this program?
Ill paraphrase the question to clarify
"Guys I have no idea how to calculate uncertainties, I dont even know where to begin...as far as that goes I dont even understand the basics.
Will someone please give me an easy fix so I dont have to be bothered to learn anything new...."
Ill answer you honestly here.
If you had enough understanding of what you needed to do then you would already have your own working excel uncertainty calculator that works...
The book that helped me understand measurement uncertainty is "Measurement Uncertainty, Fourth Edition: Methods and Applications". It a good one to read.
If you do not have the training for Uncertainty Budgets, the excell program will not help you.
Good point. You can not just plug and play an Excel sheet. I use numerous excel sheets. But there are a lot of different math formulas, knowing which to apply, which curve shapes/divisors to use for your uncertainties, and many more details. Matter of fact, if this is for ISO17025 Accreditation purposes, one of the requirements is to have someone competent to know how to do the calculations. Having a spreadsheet may be a nice tool as an example for while you are learning. But training is a must (even if it is getting things online and self-training).
Seems a little harsh. I sat through that Class A Uncertainty NCSLI class with you earlier this year and that's some complicated material to wrap your head around!!
It is difficult material. I definitely struggle with it. Hope I wasn't being harsh. Just trying to be real about how much it takes to do uncertainty budgets.
You're good. I was responding to a Marine buddy of mine that commented a few posts above yours.
Quote from: RFCAL on 10-30-2015 -- 09:09:43
If you do not have the training for Uncertainty Budgets, the excell program will not help you.
I have big experience in uncertainty calculation but i want to se experience in excel calculation of my colleagues.
Its not intentionally harsh, it is the truth though.
There are two problems with these types of requests
1) (as stated above) If you know how to do uncertainty calculations you could easily write an excel sheet to calculate them. If you do not know how to calculate uncertainties then an excel sheet, uncert software, etc isnt going to help.
2) Asking for someone to "give" you professional tools they (or someone else) developed is wrong. Asking for assistance understanding the problem or guidance in creating an excel is far better
A better request given the current information in the post might be
Guys,
I have a good understanding of uncertainties and a terrible understanding of excel.
Could someone provide guidance, (or review the attached sheet) on proper formatting of the math in excel"
Just my 1/50th of a dollar
Okey, may be anyone give excel template calculating with uncertainty toolbox?
You cannot expect people to just give you spreadsheets and software.
You will need to contact Quametek for Uncertainty Toolbox. You need the training and a license in order to operate / use that software.
One suggestion I would make....
I do most of my uncertainty work in Excel. I would highly recommend that unless you go get training in one of the freeware or paid tools, that you take some classes in Excel. There is a lot of free information about Excel on the internet. The first prerequisite I think is that you need to understand the math under uncertainty calculations before learning Excel. You need to learn what the formulas are, how to do the statistics and other functions. Then do some searching for tutorials for Excel on how to do those formulas. You will find, I think, that if you know how to do the math, Excel makes it a lot easier. I was self-taught in both Excel and uncertainty. I did tons of web searches and found many power points and pdf's from various organizations with breakdowns of what math is involved.
There is no shortcut. You will have to learn all the math (most of it can be done with Algebra. The statistics can be done by Excel using simplified formulas).
We use the Quametec uncertainty spreadsheets and it is easy to use and presents everything for an auditor. I am starting to wonder if all of this uncertainty calculation is really worth the investment, though. We are making the exact same measurements using the exact same standards. The only difference is I get a little better idea of how good the measurement is. With all of the investment in software and time to perform uncertainty budgets, where is my ROI? Sure I can charge my customer more for the additional calculations, but how does that serve them better? I am usually giving them more data on top of data that many of them don't understand anyway. I can understand providing uncertainty from a higher tier lab to a secondary lab, but for the end user, I don't get it.
I agree. We have customers doing work in contexts where they need accredited cals on everything, in many cases where they don't need, use or even understand the uncertainty calculations. They just know they need to have them. Costs a lot of effort doing calculations that aren't used.
I think the same goes for providing data most of the time to customers. Some auditor told them they needed to have it and they never end up even utilizing it! Unless its a fixed value and/or primary standard why would you care to know exact values for all the ranges and functions for general purpose equipment?
That's where your auditor would be concerned.
Now, to speak out both sides of my mouth...
I have been in the cal world for a very long time, since the MIL-STD-45662 days (the version without the A at the end).
It has taken me a long time to come around. But I am gradually getting "brainwashed" in this. After going kicking and screaming for many years, I do get it. A NIST traceable (or other traceability) measured value is only traceable if there is an uncertainty associated with it. From a basic calibration technician perspective, or a user perspective, it doesn't seem to make sense. You use a good calibrated standard with a "4:1 TAR" (I know it's an outdated term - here for illustration). But from a metrologist/metrological perspective, if you measure a UUT, by good metrological definition, you need to know how good the measurement is. If you don't, it's not really a valid measurand. So although posting MU on certificates is of little use to many lower echelon users, it is by definition, a part of what makes it a valid value.
There is, after all, in any context where the value and how quantitatively accurate that value is, the by definition need to know that it validly meets the measurement requirement. Even an end user of a Fluke 87 measuring 115 VAC at 60 Hz needs to know that the accuracy of that measurement meets their needs. If for example you "calibrated" it by plugging a light bulb into an outlet and the tolerances for the lightbulb were UTL=blow out the bulb, and LTL=bulb does not light up (disgustingly loose for illustrative purposes), and you then used that as your standard for "calibrate the DMM, that would not be adequate. Even for the least accurate application of that DMM, there is some uncertainty of measurement needed, even if the user doesn't recognize or use GUM principles.
So, I give in. Calibrations need uncertainties, period (in some form). Because a measurand is not a measurand without them. There, I spoke out of the other side of my mouth.
wow!! half of that is so wrong!
There is no need to report them to people who won't use them. But there is the need for them to exist, reported or not. This is speaking philosophically/theoretically. That is my point, that when you make any quantitative measurement, it is only of value when there is at the least, some implicit understanding that the measurement is good enough for its intended purpose. That is what I think is misunderstood about the entire topic of MU. When you use a Fluke 87 to measure a voltage, there is the implicit need to know that it is a "good" value. My illustration was that if I am measuring 115 VAC and the extent to which I know it is "good" is that even if the DMM reads 115, if I only know that the accuracy of the measurement is between 0 and infinity, then that 115 V reading is meaningless. For every tolerance, for traceability, there must be an uncertainty. There always is an uncertainty. Whether someone calculates it in accordance with GUM is a separate topic. But no measurement is traceable unless there is an associated uncertainty (reported/calculated or not).
One of my favorite papers that discusses the need for measurement uncertainty is NPL Good Practice Guide no. 131, "Beginner's Guide to Measurement in Mechanical Engineering" from the National Physical Laboratory in the UK. It treats the subject in layman's terms, and even has the famous exercise where you build an uncertainty budget for the length of a piece of string.
http://www.npl.co.uk/publications/guides/guides-by-number/
I use the Quametec Excel add-in from time to time, but it can get a bit complicated (after years of NCSLI conference tutorials on the subject, things like Student's T and Welch-Satterthwaite Formula still fly over my head). I finally came up with a simplified version of my own, based on the Excel sheet from NRC-CA, mentioned in the first reply:
Quote from: OlDave on 10-29-2015 -- 16:44:41
www.nrc-cnrc.gc.ca/obj/doc/solutions-solutions/advisory-consultatifs/clas-clas/uncertainty_budget_template.xls
I validated it against the above mentioned programs, as a sanity check, and even our 17025 accreditation assessors have reviewed my uncertainty budgets and have given a nod to them.
Quote from: Hawaii596 on 11-06-2015 -- 09:17:40
For every tolerance, for traceability, there must be an uncertainty. There always is an uncertainty. Whether someone calculates it in accordance with GUM is a separate topic. But no measurement is traceable unless there is an associated uncertainty (reported/calculated or not).
Well said. I would only add that there also needs to be an unbroken chain of measurements, and a quality system implemented to ensure the associated measurement equipment maintains their measurement uncertainty.
whatever happened to the 4 to 1 for instruments and 10 to 1 for standards rule. This still works, but there was no money to be made. Hence, uncertainty budgets.It's all a brainwash to me. If I calibrate a sig gen with an 8902A system that was calibrated at Keysight, there shouldn't be any uncertainty budgets needed. It's all a ruse to get you to shell out bucks.Nothing but BS!!
The other problem is nothing is standardized. It depends on who is your accrediting body. That will determine how you do things.
I have just completed my 5th inspection by A2LA and all was in order.Some companies, such as K---, shop around for the least intrusive inspection body so they can do things their way.They were with A2LA, then went to A-Class and now LAB. Using the same equipment,their Uncertainty Budgets are way better than mine.Just a bunch of BS!
Ok. Сan you explain me, how calibrate Fluke 8508a with Fluke 5720 on DCV mode. TUR ~1:1. And when i can use 24h, 90 day specification?
We just finished an Fluke 8508A Procedure for a customer.
You don't calibrate the 8508A with the 5720A. You have to characterize it with your labs best standards, and you have to have solid data behind your standards. Then the Fluke 5720 is used as a transfer standard.
For DC Voltage you would have to use a 10V cell, an accurate / known divider network and a null meter. Then you can use the 5720's 24 hr Specs.
Mike
Quote from: CalLabSolutions on 11-08-2015 -- 22:05:05
We just finished an Fluke 8508A Procedure for a customer.
You don't calibrate the 8508A with the 5720A. You have to characterize it with your labs best standards, and you have to have solid data behind your standards. Then the Fluke 5720 is used as a transfer standard.
For DC Voltage you would have to use a 10V cell, an accurate / known divider network and a null meter. Then you can use the 5720's 24 hr Specs.
Mike
We have DC Fluke 732b-10V, divider Fluke 752a, and null indicator Keithley 2182. With this units i calibrate 5720 (100 mV, 1V, 10V,100V, and 1kV range), and than i can use specification 24hr+this U95 calibration certificate in uncertainty budget?
Quote from: Irv1n on 11-08-2015 -- 22:25:36
We have DC Fluke 732b-10V, divider Fluke 752a, and null indicator Keithley 2182. With this units i calibrate 5720 (100 mV, 1V, 10V,100V, and 1kV range), and than i can use specification 24hr+this U95 calibration certificate in uncertainty budget?
Irv1n sounds like you guys have it all put together right. In your uncertainty budget you may have to add the resolution and repeatability of the UUT / Measurement. But that should be small contributor if you have a good set of test leads. When I was working on the procedure we had to dig around for some good cables. And in the end the customer decided to order the 8508A-Leads.
Mike
MetVet I agree about the unbroken chain. I was just simplifying.
Regarding the 4 to 1 and 10 to 1, those were simpler times. I miss them. But even the 4 to 1 and 10 to 1 were simplified means to deal with those uncertainties that were present even back then. The concept of uncertainty has been around a very long time and was used way back into the MIL-STD-45662 days. We just didn't use them at Cal Lab levels. I believe in terms of quality, the application of uncertainty has been an improvement needed as technologies push the envelope. The demand for accuracy is ever increasing. 4 to 1 and 10 to 1 were a little more "qualitative" whereas the MU method (if I may be so loose with my language) is turning it more "quantitative." Like it or not, I believe the day is coming when uncertainties will need to be calculated for every calibration. Hopefully later than sooner, as there are many calibrations where those calculations don't make much sense. Not to mention that easy application of methods to calculate and report uncertainties in accordance with ILAC P14 are not universally available (many of us spend a lot of hours on uncertainties).
Quote from: Irv1n on 11-05-2015 -- 08:49:53
Okey, may be anyone give excel template calculating with uncertainty toolbox?
There was book by ASQ (Dilip Shaw and some others) a a few years back that came with a CD with spreadsheet examples for several chapters, could not find on Amazon, check with NCLSI, ASQ and ask for Dilips book..... everyone knows Diiip!
Quote from: CalLabSolutions on 11-09-2015 -- 08:51:27
Quote from: Irv1n on 11-08-2015 -- 22:25:36
We have DC Fluke 732b-10V, divider Fluke 752a, and null indicator Keithley 2182. With this units i calibrate 5720 (100 mV, 1V, 10V,100V, and 1kV range), and than i can use specification 24hr+this U95 calibration certificate in uncertainty budget?
Irv1n sounds like you guys have it all put together right. In your uncertainty budget you may have to add the resolution and repeatability of the UUT / Measurement. But that should be small contributor if you have a good set of test leads. When I was working on the procedure we had to dig around for some good cables. And in the end the customer decided to order the 8508A-Leads.
Mike
I worked out a way to calibrate our 5720A by using the 8508A in ratio mode. The process goes basically like this:
1. Connect the standard cell output to the rear terminals of the 8508A and the 5720A to the front terminals of the 8508A.
2. Set the 8508A on the 20V range, then set the output of the 5720A to 10 volts, and operate.
3. Set the 8508A for ratio measurement by pushing the Input button, then the Scan button, then F÷R.
4. Next, dial the 5720A to the predicted value of the standard cell from the control chart (e.g. 9.999893 V), then press the "NEW REF" button.
5. Then dial the knob until the 8508A reads as close to 1.0000000 as possible.
6. The actual output of the 5720A can then be calculated from the error displayed on the screen (e.g. if the error is -0.7 ppm, then the actual output of the 5720A at 10 V is 9.999993 V).
You theoretically could then turn off the ratio mode of the 8508A, return the 5720A to nominal 10V and read the output displayed on the 8508A, then calculate the difference from the actual 5720A output found in step 6 above. That way you have calibrated your 5720A and 8508A at the same time.
For 100 and 1000 volt ranges, the uncertainty is too large across the ranges of the 8508A to use it by itself for a 10:1 and 100:1 ratio; so instead you would use a (properly balanced) 752A divider to divide down the 5720A output to stay within same 20 V range as the standard cell, so you can still take a 1:1 ratio measurement.
For 1 V and 100 mV ranges, instead of dividing the 5720A output with the 752A, you would divide down the standard cell output, and connect the 5720A directly to the 8508A. The 8508A would still be doing a 1:1 ratio measurement, except this time on the 2 V and 200 mV ranges.
BTW- just to keep on topic- another great uncertainty budget excel template can be found at the NIST website at http://www.nist.gov/pml/wmd/labmetrology/upload/Uncertainty_Budget_Table_Template_16Jan2013.xlsx
I'll have to try it. I do have a 752A.
Scott,
I pretty much did the same thing only with a 3458A.. We all know the adjustment of the 3458A is done with a 10V cell and two resistors. So I used the HP 3458A because of its design. Its internal circuits take that 10V and transfer it to all the ranges. That leaves the question, what dividers are better, but for me I like the 3458A because I can compare the 10% of one range to 100% of the other.
Mike
Quote from: scottbp on 11-10-2015 -- 17:14:38
Quote from: CalLabSolutions on 11-09-2015 -- 08:51:27
Quote from: Irv1n on 11-08-2015 -- 22:25:36
We have DC Fluke 732b-10V, divider Fluke 752a, and null indicator Keithley 2182. With this units i calibrate 5720 (100 mV, 1V, 10V,100V, and 1kV range), and than i can use specification 24hr+this U95 calibration certificate in uncertainty budget?
Irv1n sounds like you guys have it all put together right. In your uncertainty budget you may have to add the resolution and repeatability of the UUT / Measurement. But that should be small contributor if you have a good set of test leads. When I was working on the procedure we had to dig around for some good cables. And in the end the customer decided to order the 8508A-Leads.
Mike
I worked out a way to calibrate our 5720A by using the 8508A in ratio mode. The process goes basically like this:
1. Connect the standard cell output to the rear terminals of the 8508A and the 5720A to the front terminals of the 8508A.
2. Set the 8508A on the 20V range, then set the output of the 5720A to 10 volts, and operate.
3. Set the 8508A for ratio measurement by pushing the Input button, then the Scan button, then F÷R.
4. Next, dial the 5720A to the predicted value of the standard cell from the control chart (e.g. 9.999893 V), then press the "NEW REF" button.
5. Then dial the knob until the 8508A reads as close to 1.0000000 as possible.
6. The actual output of the 5720A can then be calculated from the error displayed on the screen (e.g. if the error is -0.7 ppm, then the actual output of the 5720A at 10 V is 9.999993 V).
You theoretically could then turn off the ratio mode of the 8508A, return the 5720A to nominal 10V and read the output displayed on the 8508A, then calculate the difference from the actual 5720A output found in step 6 above. That way you have calibrated your 5720A and 8508A at the same time.
For 100 and 1000 volt ranges, the uncertainty is too large across the ranges of the 8508A to use it by itself for a 10:1 and 100:1 ratio; so instead you would use a (properly balanced) 752A divider to divide down the 5720A output to stay within same 20 V range as the standard cell, so you can still take a 1:1 ratio measurement.
For 1 V and 100 mV ranges, instead of dividing the 5720A output with the 752A, you would divide down the standard cell output, and connect the 5720A directly to the 8508A. The 8508A would still be doing a 1:1 ratio measurement, except this time on the 2 V and 200 mV ranges.
BTW- just to keep on topic- another great uncertainty budget excel template can be found at the NIST website at http://www.nist.gov/pml/wmd/labmetrology/upload/Uncertainty_Budget_Table_Template_16Jan2013.xlsx
and what uncertainty model (mathematical model) of this measurements?
UGH.. I spent as much time on the Unc calculations as I did writing the software.
It is pretty straightforward to do calculations for ScottBP's method. But not simple. You have to go through each step, enumerate all the contributors, and run the various RSS calculations, etc.
Just lots of work in Excel.
I worked up a quick and dirty budget as follows:
10 V standard cell: 0.7 µV/V
8508A ratio uncert. on 20V range: 0.12 ppm rdg+0.1 ppm rng
This works out to be 1.43 µV/V, which is ±14.3 µV at 10V. The 1 year, 95% spec of the 5720 at 10V is 37.5 µV, so it's about a 2.61 t.u.r. The largest uncertainty contributor is the standard cell; if we could hold it down to about half, we could bring the uncertainty down to 7.66 µV and the t.u.r. up to 4.89.
The divider adds 0.2 ppm on the 10:1 ratio, and 0.5 ppm on the 100:1 ratio.
The 8508A is 0.4 ppm rdg + 0.3 ppm rng on the 200 mV range, and 0.12 ppm rdg+0.1 ppm rng on the 2 V range.
Like you say... Straightforward, but not simple; a lot of Excel calculations.
With the standard cell on the back of the 8508, 5720 on the front doing F/R ratio you could just set the 5720 to the nominal value and then multiply the 8508 measured ratio times the standard cells value to determine the 5720 actual. At least that's what I'm working on, hoping to shelve the 752 and come up with a setup that lessens operator influence/interpolation
You are not making absolute measurements with the meter, so you don't have to use the absolute spec. Because we made the measurements, one after the other at 10V, we only used the meter's floor value. Then we RSS-ed in the 5720's linearity spec. Because our ultimate goal was to generate exactly 10.000000000V.
This also allowed us to test our methodology. We can then cal a 3458A to the 10V cell and see just how close we came to our 10V.
Mike
What resolution uncertainty i can use for analog devices, how calculate?
Tom Morgon did a great article on this..
http://www.callabmag.com/?p=4201
Quote from: scottbp on 11-12-2015 -- 17:25:33
8508A ratio uncert. on 20V range: 0.12 ppm rdg+0.1 ppm rng
This works out to be 1.43 µV/V, which is ±14.3 µV at 10V.
how u receive 14.3 µV at 10V? what rdg was in measurements?
can u help me with 1 question. I calibrate dmm (fluke 8508, agilent 3458) from 100 mV to 10V on my josephson voltage standard. And have result:
gain factor (example 1.00035 from -10 to 10V) and mean square error (example 23 nV). How i can use this results? how i can calcultae voltage for point like 3 V? How calculate uncertainty with this values?
Wow!! You need more help than I can give you!
We have MET/CAL procedures available for both the HP 3458A and Fluke 8508A each with full uncertainty budgets based on the customer's standards.
Irv1n, I'm sure we can help you with your uncertainty problem, but I need to know all the details about your test setup methodology.
You can shoot me an E-mail MSchwartz@CalLabSolutions.com
Mike
Quote from: CalLabSolutions on 11-20-2015 -- 12:55:42
We have MET/CAL procedures available for both the HP 3458A and Fluke 8508A each with full uncertainty budgets based on the customer's standards.
Irv1n, I'm sure we can help you with your uncertainty problem, but I need to know all the details about your test setup methodology.
You can shoot me an E-mail MSchwartz@CalLabSolutions.com
Mike
Okey tomorrow i give more info about this calibration.
Can anyone say, what program use Fluke for creating Calibration certificate? Metcal or excel?
If u know or have example of certificate can u say what fonts and size they used?
Can anyone say, how i can verify 3458a for 90day option 002?
For what it's worth I have been using Integrated Sciences Group "Uncertainty Analyzer" software. It took me about 4 months to go through the working exercises and I used actual instruments with their certificates of calibration. This worked. I now understand the basics and recognize the scope of what I do not.
For proficiency testing, I asked one of the proficiency testing providers if he really needed a "full blown" estimate of uncertainty and he gave me an excellent tool for on-the-spot uncertainty estimates.
1 - Square the tolerance of the UUT.
2 - Square the tolerance of the standard(s)
3 - Take 5 repeatability measurements, find the standard deviation, find the mean and square that.
4 - Add these together.
5 - Take the square root of that.
This estimate of uncertainty will be in the ballpark and will be useful but of course will not be what an ISO auditor is looking for.
Statistics and math are not my domain. I welcome any corrections to what I have added here.
Yes, that will give you an estimated uncertainty. There is a somewhat tricky part before using your method in that you need to normalize your tolerances. Tricky because a lot of manufacturers do not specifically state what sort of confidence level or statistical distribution their specifications adhere to.
Fluke is somewhat good about this as they list specifications at the 95% (and 99%) confidence level but they also reference these same specifications as "approximately 95%" which should raise the eyebrows of anyone who knows the difference between the two statements. But compared to other manufacturers at least they give you something to work with.
Anyway, depending on if you know how your standard's tolerance is reported (whether K = 2, K = 3, 95%, 99%, 99.7%, etc.), you would simply divide your reported tolerance by a certain factor. If you can't find this information, it is recommended that you assume it has a uniform distribution and you can divide the standard tolerance by sqrt(3) or 1.732. This normalizes the spec so you can combine it with your standard deviation of repeated measurements.
It is arguable whether or not to include your UUT tolerance in your calculation. I typically don't, but maybe you have a good reason to.
Quote from: N79 on 02-13-2016 -- 10:14:51
... but they also reference these same specifications as "approximately 95%" which should raise the eyebrows of anyone who knows the difference between the two statements. But compared to other manufacturers at least they give you something to work with.
...It is arguable whether or not to include your UUT tolerance in your calculation. I typically don't, but maybe you have a good reason to.
Approximately 95% is K=2 (94 or 95.something is actually K=2, can't be bothered to find the exact number, it is Saturday night after all). It should raise some eyebrows if they don't state approximately 95%......
If it is your measurement uncertainty, the only thing from your UUT you should be using is the resolution, iirc - depending on the type of UUT of course.
Quote from: CalibratorJ on 02-13-2016 -- 17:37:13Approximately 95% is K=2 (94 or 95.something is actually K=2, can't be bothered to find the exact number, it is Saturday night after all). It should raise some eyebrows if they don't state approximately 95%......
There is a (albeit slight) difference between K = 2 and a 95% CL, and they shouldn't be interchanged. If Fluke is going to list specifications at 95%, they shouldn't later refer to it as approximately 95%. They should list them as 2σ or K = 2, but not 95%. Also listing 99% specs just makes it even more confusing because if one assumes 95% is K = 2, then 99% must really be K = 3 which is quite a bit different than a 99% CL.
Yes, I can see that the confidence factor(s) of the tolerances do have to be accounted for. I will make a note of this.
I grasp that the standards used to certify an instrument are a major contributor to the uncertainty of a measurement and had assumed the tolerance of the UUT was also part of this.
I sat in a couple of Measurement Science Conference workshops on uncertainty and gaurdbanding and the certificates look nice but most of what I actually understand about uncertainty is self taught.
The last ISO auditor we had mentioned Quametec and I have heard Excel mentioned. Is there any "industry standard" software that most accredited labs are using to report uncertainty?
I bought the book 'Statistics for Dummies' to better understand measurement uncertainty and its principles.
Standard Deviation 2 = 95.45% [If you use 95% you are at a K factor of 1.96; perhaps close enough]
Standard Deviation 3 = 99.73% [If you use 99% you are at a K factor of 2.58; not close enough]
How i can calculate stability (12 MONTH) of DC Voltage if i have calibration between 1 and 2: near 1 year 9 month?
(1 - 2)/ 1.9 ? What document describe this calculation? Thank You.
Irvin--you need formal training for uncertainty budgets. You need more help than anyone on this site can give you.Self taught is OK for some budgets, but not for the questions you are asking. You can take classes from NIST or the NCSLI or MSC conferences. After that, go to quametec.com and take their software classes.
Here's an excellent resource:
http://www.isobudgets.com/