How Can I determine the calibration frequency for the measuring devices?

We have digital calipers, Micrometers, Temperature gauges, weighing scale, Block gauges.

Can I make calibration every one year, two years, less or more?

And are there any standard for this.

Many thanks in advance.

Hamed Radwan

It all depends on how much risk you want to take.

How Can I determine the calibration frequency for the measuring devices?

We have digital calipers, Micrometers, Temperature gauges, weighing scale, Block gauges.

Can I make calibration every one year, two years, less or more?

And are there any standard for this.

Many thanks in advance.

Hamed Radwan

If you only check it once a year, and it is wrong, then you have been using a bad gage for up to a year or two. Are you sure you want to take that risk, just to save 5 minutes of calibration?

What kind of environment? What is the risk (to the equipment, as well as product) How many devices do you have?

Do you have any data on some sort of daily verification done to make sure that the devices are stable from day to day? week to week?

:2cents:Start out at a frequency that is pretty short, collect your data, then double it. example. start out calibrations every week. then go to every two weeks. etc., until you find a frequency that doesn't allow things to go out of calibration, but also doesn't overwhelm your resources. I'm sure there is some "official" guide for setting up frequencies, but I am just a grunt in the field, not the general giving orders when it comes to cals.

In Alan S. Morris book “Measurement & Calibration for Quality Assurance” his method or what he calls is philosophy is to perform the following frequency of checks for instrument characteristics that might be appropriate for an instrument in use 24 hours a day:

Day 1: Once per hour for first 4 hours, then every 4 hours
Days 2-3: Every 12 hours
Days 4-7: Once per day
Weeks 2-3: Twice per week
Weeks 3-4: Once per week
Week 6: Once
Week 8: Once
Months 3-6: Once per month
Month 9: Once
Month 12: Once
Month 18: Once
Month 24: Once
Then once per year thereafter

Or a much simpler method to set the initial frequency is to simply ask the manufacture for their guidelines and then then set the frequency from the information they supply.

Also see: Determining Measurement Device Calibration Frequency (http://Elsmar.com/Forums/search.php?do=process&titleonly=1&query=calibration+frequency)

In Alan S. Morris book “Measurement & Calibration for Quality Assurance” his method or what he calls is philosophy is to perform the following frequency of checks for instrument characteristics that might be appropriate for an instrument in use 24 hours a day:

Day 1: Once per hour for first 4 hours, then every 4 hours
Days 2-3: Every 12 hours
Days 4-7: Once per day
Weeks 2-3: Twice per week
Weeks 3-4: Once per week
Week 6: Once
Week 8: Once
Months 3-6: Once per month
Month 9: Once
Month 12: Once
Month 18: Once
Month 24: Once
Then once per year thereafter

Or a much simpler method to set the initial frequency is to simply ask the manufacture for their guidelines and then then set the frequency from the information they supply.


I liked everything but the last line. Jumping for 1x per month to 1x per year is a big jump. Gages that can go wrong probably should be done more frequently.

In my experience measuring equipment manufacturers recommendations cannot always be relied on. Bizarrely, some even have no clue about true calibration!

Their ideas are just that, ideas, since they have no data about the use of the equipment. You could have a device which gets hammered for three sifts gauging threads in stainless steel, and it wouldn't last a week. Some things could go years before they need to be rechecked.

I'd recommend a study, by family of equipment and then by useage, especially if you have some high frequency use items in an application which is wear inducing.

If you read the ISO 9001:2000 standard carefully enough you begin to realize that part of the answer is there.

In element 7.6a is says " . . . calibrated or verified at specified intervals . . ." Who gets to set the interval? You do. Obviously, you don't want it to be so frequent that it is a burden, but what sets the other limit? The answer is in the paragraph after the lettered list: "In addition, the organization shall assess and record the validity of the previous measuring results when the equipment is found not to conform to requirements. The organization shall take appropriate action on the equipment and any product affected."

So you need to not set the interval so frequently that it is an unnecessary burden, but frequently enough that you don't run into problems that could make it necessary to go through the exercise of figuring out what if any bad product was shipped and how to resolve it.

As someone else said, a good place to start is with the equipment manufactures recommendations. Another good source is your local calibration lab. Then adjust from there pending the results of follow-up calibrations. Mix a little common sense into the equation as well.

It all depends on how much risk you want to take.

Exactly! You have summed it up nicely.

In my experience measuring equipment manufacturers recommendations cannot always be relied on. Bizarrely, some even have no clue about true calibration!

Their ideas are just that, ideas, since they have no data about the use of the equipment. You could have a device which gets hammered for three sifts gauging threads in stainless steel, and it wouldn't last a week. Some things could go years before they need to be rechecked.



Spot on!!:agree1: Man, don't get me started:mad:, Andy. I think it's getting a little better with mfg. specifications and all, but it is still pretty sad. Bottom line: Mfg. companies are in the business of making equipment and money... not necessarily interested in good metrology practices (Fluke is one exception, IMHO). Hence the reason you can rely on competent metrology sources for assistance. If nothing else is available, start with mfg. recommendations, but manage the process after that.


If you read the ISO 9001:2000 standard carefully enough you begin to realize that part of the answer is there.

In element 7.6a is says " . . . calibrated or verified at specified intervals . . ." Who gets to set the interval? You do. Obviously, you don't want it to be so frequent that it is a burden, but what sets the other limit? The answer is in the paragraph after the lettered list: "In addition, the organization shall assess and record the validity of the previous measuring results when the equipment is found not to conform to requirements. The organization shall take appropriate action on the equipment and any product affected."

So you need to not set the interval so frequently that it is an unnecessary burden, but frequently enough that you don't run into problems that could make it necessary to go through the exercise of figuring out what if any bad product was shipped and how to resolve it.

As someone else said, a good place to start is with the equipment manufactures recommendations. Another good source is your local calibration lab. Then adjust from there pending the results of follow-up calibrations. Mix a little common sense into the equation as well.

Good points, Jim. Taking up on your common sense suggestion...

Regardless of your interval, don't turn your brain off with your process. If your process is going along with decent stability and all of a sudden you get some noise, it might not be a bad idea to check the instruments, use a spare, etc. Also, if you are nervous about your interval, start short, and then stretch as you are able. Send your instruments to a good calibration source that provides as found/ as left data, so you can make informed decisions regarding calibration frequency.

The main thing is manage the process. Adjust the frequencies when the situation calls for it. When (or if) you get it right, you be having your instruments calibrated (and probably adjusted) at the right time frame, maximizing benefit and minimizing cost.

Also see: Determining Measurement Device Calibration Frequency (http://Elsmar.com/Forums/search.php?do=process&titleonly=1&query=calibration+frequency)Marc, why are both threads identical?

How Can I determine the calibration frequency for the measuring devices?

Can I make calibration every one year, two years, less or more?

And are there any standard for this.IMHO using time as an interval is a poor practice. Calibration frequency needs to be use based.

Identical tools will wear at diferent rates based on:
1. Work enviornemnt
2. Material being tested
3. Operator proficenicy
4. Consistancy of production at median size

Marc, why are both threads identical?Technically that's a Search link which yields a listing of existing Calibration Frequency discussion threads. They're listed by order of newest to oldest, so this thread is 'On Top' of the listing right now.

IMHO using time as an interval is a poor practice. Calibration frequency needs to be use based.

Identical tools will wear at diferent rates based on:
1. Work enviornemnt
2. Material being tested
3. Operator proficenicy
4. Consistancy of production at median size

But time is by far the easiest to monitor. It may be a little hard to track and check a micrometer every 100 uses. Anticipated use during a time frame just about has to be one of the factors. If heavier use is anticipated, the cycle can be closed up, which is often the case with something like thread gages.

Keep in mind that ISO 17025 says that calibration labs should not reccomend calibration intervals. As a manager of an ISo 17025 accredtited lab, I am asked several times a week how often something should be calibrated. Here is my standard answer:

"ISO 17025 says that I should not reccomend intervals. Of course that is not a good answer to your question. I can tell you what most other people do for intervals...."

I agree with the previous poster that said it is up to you to specify the intervals. You then just have to stick to that schedule.

For calipers, micrometers, height gages, dial indicators most poeple I work with seem to use a 6 or 12 month interval.

For temperature gages most people I work with seem to use 12 months (with a few using 6 months)

Out of the equipment you listed, your scales are most likely to need to shortest interval. We have many customers who use months or quarterly. Thisreally depends upon the environment. The more abusive the environment, the more frequent the calibration.

I do not know of any standard that lists specifically how often things should be done. As many have said before, it is all a matter of risk.

But time is by far the easiest to monitor. It may be a little hard to track and check a micrometer every 100 uses. Anticipated use during a time frame just about has to be one of the factors. If heavier use is anticipated, the cycle can be closed up, which is often the case with something like thread gages.Check out this related thread:  Never put an Expiration (Due) Date on a Calibration Certificate (http://elsmar.com/Forums/showthread.php?t=21953).

Check out this related thread:  Never put an Expiration (Due) Date on a Calibration Certificate (http://http://elsmar.com/Forums/showthread.php?t=21953).

Corrected link: Never put an Expiration (Due) Date on a Calibration Certificate (http://elsmar.com/Forums/showthread.php?t=21953)

Stijloor.

Check out this related thread:  Never put an Expiration (Due) Date on a Calibration Certificate (http://elsmar.com/Forums/showthread.php?t=21953).

Interesting concepts. Sort of puts new meaning to " . . . be calibrated or verified as specified intervals . . . "

I would think that for most companies the concept of fulfiling 7.6a without expiration dates would be more trouble than its worth.

Interesting concepts. Sort of puts new meaning to " . . . be calibrated or verified as specified intervals . . . "

I would think that for most companies the concept of fulfiling 7.6a without expiration dates would be more trouble than its worth.The trick here is that:
1. The scheduled recalibration date should not be called: expiration date.
2. The scheduled recalibration date is a decision made by the owner, not the calibration lab.
3. The scheduled recalibration date is not a blanked policy, but a studied approach to each tool.
4. The scheduled recalibration date should not be applied to the certifrication by the calibration lab. The owner could hand write the date on the front or back of the certification, or put it on a post-it, or just track in software.

Interesting concepts. Sort of puts new meaning to " . . . be calibrated or verified as specified intervals . . . "

I would think that for most companies the concept of fulfiling 7.6a without expiration dates would be more trouble than its worth.

Big Jim,

The standard gives us an option: ".....or prior to use...."

Stijloor.

Big Jim,

The standard gives us an option: ".....or prior to use...."

Stijloor.

I'm well aware of the "prior to use" stipulation, and I know it applies to instruments that are not very stable and need frequent calibration as well as on equipment that is used so seldom that it would be a burden to keep it in calibration all the time. Somehow I don't see it practical to calibrate prior to use for common, frequently used equipment, such as a micrometer. Calibrating a micrometer before each use would be a burden in most circumstances.

How Can I determine the calibration frequency for the measuring devices?
We have digital calipers, Micrometers, Temperature gauges, weighing scale, Block gauges.
Can I make calibration every one year, two years, less or more?
And are there any standard for this.
Many thanks in advance.
Hamed Radwan

First, lookup for your measuring devices manuals (if any), is there any suggestion for calibration time interval? If is, it will be just suggestion not an obligation! If you calibrate your devices in-house and also number of your devices respect to your calibration manpower is not a problem, you can use the methods as dear Jeff mentioned above but otherwise you can choose a calibration time interval (e.g. 1year) for each device according to: usage rate of the measuring device, quality of your measuring device, probable environmental parameters affecting on the device accuracy and etc., then (e.g. about dimensional devices you named) use a known sample block for daily inspection by end-user to insure about the devices accuracy before everyday use.:o

I thought to use kind of SPC concept, we daily/ weekly measure the so call golden/ standard unit, then plot the chart to understand whether the Measuring Device is within control or not, if out of control, then need re-calibrate, repair or scrap.

As for some instrument, we do calibration yearly and never see out of calibration but we spend so much money :D, which is not cost effective.

There is a Danish organisation (non-profit I believe) for workshop metrology that has a series of papers on most standard measuring equipment. Calipers for example have:
Calibration - Information
Calibration - Instruction
Check - Instruction
Data sheet
Calibration sheet

The biggest problem is that eveything is in Danish but iif you send me a PM I can give you their email address and fax number. I think they should probably know where you can get the information you want in English, as I can't believe Denmark is the only country to have this information.
As you can see there is distinguished between a calibration and a check. Check is done regularly - a calibration at longer intervals. Many Danish companies use these guidelines as part of their ISO 9000 system.

There is a Danish organisation (non-profit I believe) for workshop metrology that has a series of papers on most standard measuring equipment. Calipers for example have:
Calibration - Information
Calibration - Instruction
Check - Instruction
Data sheet
Calibration sheet

The biggest problem is that eveything is in Danish but iif you send me a PM I can give you their email address and fax number. I think they should probably know where you can get the information you want in English, as I can't believe Denmark is the only country to have this information.
As you can see there is distinguished between a calibration and a check. Check is done regularly - a calibration at longer intervals. Many Danish companies use these guidelines as part of their ISO 9000 system.

I believe it is called DANAK (http://webtool.danak.dk/Plone/english/).

The site is also in English.

The metrology site is here (http://webtool.danak.dk/Plone/metrologi/english/fundamental-metrology/).

Stijloor.

Stijloor,
Actually DANAK isn't the one I meant :) Here's another site for DANAK http://webtool.danak.dk/english, although I don't think it is the one to go to for guidelines, which is what I thought was originally being asked for. I think most (?) countries have their verson of DANAK.
The Danish name for the one I meant is Foreningen for Vaerkstedsteknisk Metrologi FVM (translated as best I can makes it The Association for Workshop Technical Metrology).
This association sells (among other things related to practical metrology), calibration procedures and calibration stickers. They also organize quality seminars and trips to domestic and foreign exhibitions that have a quality/metrology theme. Members get the chance to get together and discuss problems and solutions to everyday metrology problems. This gives it a more practical function than theoretical one. When "theoretical experts" get together it often ends in a discussion as to why a bumble bee shouldn't be able to fly :)
I could give their (FVM) address, phone and fax number plus email here, if you give me the OK. It is a Danish organization, so they could probably only give a reference to other similar organizations they may (or may not) know about.

I went into the FVM website but everything is in Danish :(
Here's a translation of their Vision and Goal:
FVM's Vision og Mål
FVM's vision er at være "Forum for styringen af egenskaber på mekaniske produkter".
FVM's mål er gennem en aktiv indsats for foreningens medlemmer at fremme udbredelsen og kendskabet til industriel måleteknik (anvendt Metrologi) i dansk industri, så danske virksomheder og institutter står stærkere i den stadig stigende internationale konkurrence og globalisering.
Translation:
FVM's vision is to be a "Forum for control of properties of mechanical products".
FVM's goal is, through an active effort for association members, to further the spreading and knowledge (usable metrology) so that Danish companies can be strengthened in the ever increasing international competition and globalization.

Is there anything similar in the USA?

The closest MAY be from NACLA.....but not too sure there.....

http://www.nacla.net

Hershal,
Thanks for your reply but NACLA looks more to me like a learning/training type facility, whereas the Danish FVM mainly focuses on utilizing the experience and know-how of members to help each other. There isn't any official "guru" type, or types, although there are one or two Guru wannabes :) when members congregate.
I might compare it to a thread forum in here if members could actually (physically) get together for field trips (exhibitions) and discussions to, in detail, discuss their interest in a particular field.
Measurement uncertainty (NACLA) is of course, something that should be considered (and checked) when measuring, but some "experts" have elevated much of it to a level that I often feel is way beyond practical use. If I had a suspicious mind, I'd suspect the measuring equipment suppliers of high precision (and costly) measuring equipment to lobbying for elevating measurement uncertainty to the level it has achieved :D
I talked recently to one guy that teaches (among other things) measurement uncertainty, and he said that 80% of digital caliper innacuracy was user influenced. He was very impressed when he tried a caliper measurement pressure device and said he'd mention that it did exist in future lectures, as a possibily to reduce (not eiminate) user influence on the result. Again, it's important when caliper measurement results should be as reliable and accurate as possible. If the digital caliper is used in more or less the same way as measurement with a steel rule or tape measure, then of course it makes virtually no difference at all whether a pressure evice is used or not.
One day I'll have to learn how to give short, precise answers without "moralizing" :D
If anyone disagrees with what I've written I'd love to hear why!

I can't disagree with the statements regarding calipers.....user influences are one source of the major Type B contributions.....

I should clear up that NACLA does sponsor some training, but they are actually the National collection of ABs and regulators.....not quite the same as APLAC or EA, but the general concept is basically the same.....

Hershall,
Please note I'm not an American so all the abbreviatons you use, might as well have been written in Chinese - no offence meant to the Chinese.

I can't disagree with the statements regarding calipers.....user influences are one source of the major Type B contributions.....

I should clear up that NACLA does sponsor some training, but they are actually the National collection of ABs and regulators.....not quite the same as APLAC or EA, but the general concept is basically the same.....

Wes (Bucey) gave me a link to look up abbreviations but there is rarely just one answer - in fact often dozens. I'm writing this as a help and not a "rebuke" to all those that think everyone knows the abbreviations they use regularly. Many would rather just "give up" instead of asking or admitting they don't know.

For Big Jim,

"Calibrating a micrometer before each use would be a burden in most circumstances".

Yes and no :) In the case of a micrometer, usually every micrometer above 1" (i.e. 1 - 2", 2 - 3" etc.) has a length rod in the case. Personally I always do a quick check on a micrometer to see if it's at least correct at "0". It takes seconds and if that's a "burden" then someone is headed for a stress breakdown :) I do the same with a caliper - a quick check to make sure everything is OK before I use it.

That covered the "No" part - as for the "Yes" part then a full calibration takes much longer, but making sure people take care of their measuring instruments will go a long way to avoiding the need for frequent full calibration. I'll bet if someone cared enough to check who used what, when, and for how long, then only a few "culprits" need to be monitored by the FBI (Frequent Bullshxt Investigators).

IMO the criterion for measuring device calibration frequency rests upon usage alone. The 'usage' got inadvertantly converted, illogically into 'time interval', which de-facto has no bearing or relation to calibration.

For example, a vernier or a mic. used by my patrolling inspector goes for calibration in a three months cycle, where as the one used by the Q/A engineer goes for six months cycle. The plant manager's instrument is on one year cycle. Again these are approximations based on experience. So you can't fix a timetable for any instrument, since it varies w.r.t. usage.

The most exessively used instuments in my plant are 'Thread plug gages', and their frequency for calib. varies from one to three months! In one perticular component my production has almost quadrupled, and I am going to switch the calib. frequency on 'per.. .thousands' basis. There simply is no other go!

Hope this helps

Umang :D

Umang,

This seems to be an excellent system based on experience and practicality. Am right in assuming that the most often used measuring devices are "dulplicated" i.e. there are extras to avoid loss of measurement capabiity during calibration?

I'm a little surprised at the calibration frequency for thread gauges, but of course some materials wear gauges down much more than others (stainless steel, cast iron etc.). Do you ever measure thread pitch diameter when setting up the machines? If the thread pitch diameter was as close as possible to the middle of the tolerance I'm sure wear on your thread gauges could be reduced significantly as many machine operators try and "force" gauges to get as "snug" a fit as possible.

The bottom line is, that to me, your company seems intent on producing high quality products that conform to specification. Congratulations :D

Umang,

This seems to be an excellent system based on experience and practicality. Am right in assuming that the most often used measuring devices are "dulplicated" i.e. there are extras to avoid loss of measurement capabiity during calibration?

You got it right Gordon. We maintain extra devices to avoid loss of measurement aka loss of production. The quantity of 'reserve' depends upon the duration of calibration cycle. Shorter the duration, larger the reserve & vice-versa. I believe this is standard in all the manufacturing units who care for quality.

I'm a little surprised at the calibration frequency for thread gauges, but of course some materials wear gauges down much more than others (stainless steel, cast iron etc.).

Your surprise at the calibration frequency is natural and understandable, since 'prima-facie' it is too high. One of the reason is, that high volume of components are being inspected by TPGs every day. As I've stated in my earlier post, I am going to shift the calib. freq. for 'high volume' components, from time basis to per...thousand basis.

Second, and more important reason is that these components are electroplated after the threading. In electroplating, the material deposit on the threads is always uneven id est high deposit at the crest with low or no deposit in the root and varying deposit on the flank. This contributes towards high wear and tear of the TPGs. "Electroless plating" is one way to control this evil, but is too expensive. That is why we have compromised with lesser of the two evils, and prefer high calibration frequency.

Do you ever measure thread pitch diameter when setting up the machines? If the thread pitch diameter was as close as possible to the middle of the tolerance I'm sure wear on your thread gauges could be reduced significantly as many machine operators try and "force" gauges to get as "snug" a fit as possible.

We do not routinely check pitch dia. It does play an important role, but the major contributor here is electroplating. The opertators' role in forcing the gauge is controlable. Supervisors and Engineers keep a close watch on operators, physically as well through CCTV, not with intent to punish, but to teach and educate them.

The bottom line is, that to me, your company seems intent on producing high quality products that conform to specification. Congratulations :D

Thanks for the compliment. It is our continuous endeavour to produce quality products. I have introduced 'Toyota way' (Toyota Production System) in the factory, and the results in the past four years are very encouraging, (though we are far away from the ideal). We are rated 'green' by most of our customers and our supplies go DOL (direct on line). Our company is known for quality products since long, but before introduction of 'TPS' we were selling quality products, and after 'TPS' we are making quality products. I am sure you appreciate the diff.

Umang :D

Umang,

I wrote a long reply and then hit the wrong button I think. To make a long story short everything disappeared. This reply will be shorter than my original one.

I’m impressed, and if your company doesn’t already do so, should charge other companies to come visit you and learn. OK, that could give unwanted competition :)

I worked as Quality Manager at a company that electroplated (mass production too) the majority of their products. Before I came they had constant problems with their threads as they just didn’t realise that with threads, electroplating removed 4 times the normal amount of material, as there are 4 flank angles on a thread.

Only one thing in what you write gives me a question:

“The operators' role in forcing the gauge is controllable”.

How on earth do you do that? Control the (human) “force” I mean.

I’ll bet you have your electroplating process well under control, so why don’t you monitor the thread pitch diameter by measuring before electroplating? I’ll also be willing to bet that you use SPC for just about everything except threads.

By making the thread to the pitch diameter necessary (required) after electroplating, means that you only need to check a very small number to verify conformance to spec. after the electroplating process is done. It would also mean that gauges don’t need to be forced at any time – you’ll always know where you are within the tolerance. Gauges can then be used to verify correct profile and pitch. Metals requiring electroplating are usually very rough on gauges – but I’m telling you something you already know :)

I’d appreciate if you’d send me a PM (private message) and let me know what type and size of threads you make most of. As you use electroplating I’m also pretty sure which types of industry you supply to. Threads are my speciality, so I might have some advice that could be useful :)

Gordon

In my experience measuring equipment manufacturers recommendations cannot always be relied on. Bizarrely, some even have no clue about true calibration!

Their ideas are just that, ideas, since they have no data about the use of the equipment. You could have a device which gets hammered for three sifts gauging threads in stainless steel, and it wouldn't last a week. Some things could go years before they need to be rechecked.

I'd recommend a study, by family of equipment and then by useage, especially if you have some high frequency use items in an application which is wear inducing.
Hi AndyN,

There’s nothing you write I can disagree with. I would add that, if possible, find out who takes good care of the equipment they use and who doesn’t. Items like micrometers, calipers and the like, that are used frequently, could (should?) be issued “permanently” and their serial number noted. They almost always have a serial number somewhere. I think the guy that got a “calibration” visit regularly would buckle up. Solid (and also expensive) inspection equipment can’t usually be permanently placed at any one person so this isn’t easy. There could be (depending on how mass production mass production is) separate gauges for different metals. Brass isn’t nearly as “wearing” on gauges as stainless steel. I think if many threaded components had to me inspected I’d measure occasionally (monitor) pitch diameter and only use the gauge to check for wear to the thread cutting tool. My experience tells me that a thread gauge that “feels” tight is usually too close to the top or bottom tolerance.

As you state, a study is required to determine what is appropriate in each case. What works for one company doesn’t always work for another.