What is the difference between in house calibration and verification?

hi all,

all the time we sent the monitoring and measuring devices to external parties for calibration but this year in order to save cost, i was asked to carry out so-called "in house calibration/ verification". i was asked to send only one of each type of equipment for external calibration and keep that as a standard, for example ruler or micrometer. and using that as a standard, i measure things using the "standard" equipment and also non-externally-calibrated equipment, if i can get the same measured value, the equipment is considered "verified" or "calibrated". does this make sense? how often do i do the verification...do i need to have the working instruction? do i need to justify which equipment can be verified this way and which cannot?

thanks for reply.

Re: does this called in house calibration or verification?

kbng82 said:

hi all,

all the time we sent the monitoring and measuring devices to external parties for calibration but this year in order to save cost, i was asked to carry out so-called "in house calibration/ verification". i was asked to send only one of each type of equipment for external calibration and keep that as a standard, for example ruler or micrometer. and using that as a standard, i measure things using the "standard" equipment and also non-externally-calibrated equipment, if i can get the same measured value, the equipment is considered "verified" or "calibrated". does this make sense? how often do i do the verification...do i need to have the working instruction? do i need to justify which equipment can be verified this way and which cannot?

thanks for reply.

Click to expand...

If you have the proper training for performing in house calibration, then the usual way of doing it is to send out master equipment: gages, blocks, micrometers, rulers etc. for external calibration. Using the master equipment you then calibrate day to day equipment in house. Some equipment may need a high degree of confidence in the calibration and should therefore be calibrated externally. But, if this is the case then you definitely need to evaluate the use of the equipment in house due to the fact that the majority of a measurement uncertainty is related to handling.

Re: does this called in house calibration or verification?

I moved this to the 17025 thread because this type of question may receive more technically specific answers from calibration folks in this thead.

Re: does this called in house calibration or verification?

Yes,this is called in house calibration or verification,but what you did is not fully complied with the requirements of ISO9000.you are suppose to set up a series of procedures and standards.
If you want to comply with Iso9000, you have a lot of things to do.

Re: does this called in house calibration or verification?

thanks for your prompt reply.

mind sharing some procedure or WI in order for me to comply with ISO 9001? Simple equipment, RULER for instance, to check the measured value against the measured value using externally-calibrated, is there requirement regarding how many reading must be taken, tolerance, etc.... i have no idea about this but auditor did ask me to do so on simple equipment to save money... there is no such kind of training provided yet in this company on calibration...

another question...the standard equipment got to be the standard equipment for every year and can the equipments of its kind take turn to be the standard equipment?

help!!! i need a procedure on this...

Re: does this called in house calibration or verification?

mind sharing some procedure or WI...........

You can find procedures and WI in the post attachments list on page 9. You have to scroll about mid page to find what I think your looking for.

Hope this helps.......:

Also see these existing discussion threads on the difference between calibration vs. verification.

Really this all can be answered by what standard you are being held to, or holding yourself to. How complex is the equipment you're talking about? Take your ruler example for instance...you'd have to ask yourself what kind of precision you need, for linear measurement such as this there are many routes to go, you could verify your rulers on a calibrated profile projector, or against a standard gage block set, or against a calibrated caliper.

As for your calibration/verification intervals, this is left up to you, once again risk, cost, and common sense will give you guidance. Some of your customers or your registrar may require regulated intervals for certain instruments, OEMs may have suggested calibration frequencies, and finally historical performance data of the instruments, all this will give you guidance for setting your cal intervals.

One thing I'd like to point out however, comparing an externally calibrated instrument against an operational instrument of the same family may not be the best route to take with this. Such as a micrometer, if you're measuring a part with your master mic, and then comparing against your operational mic, this is all well and good, however, you'd be better off just using a NIST traceable gage block as the block will be uniform, calibrated...if you're using an appropriate gage block there is no need to have a "master mic" in this case and you'll be able to calibrate/verify most of your linear measurement instruments off this set of blocks, thus one reference standard, calibrated and traceable, is used for multiple instruments and the cost/upkeep will likely be lower and easier to maintain. Be sure whatever methods you use are acceptable by your customers and registrar...if they think you're just doing op checks vs a thorough calibration/verification they may not find this acceptable.

My suggestion, do some review, compile what type of standards will be needed, how much manpower will be needed, training, and what type of facility requirements will be involved, calculate the costs involved and present this to your boss...he will be glad to see what type of savings or lackthereof will be involved. Keep in mind, if you need precise calibration you will also need to have a controlled environment, temp, humidity, lighting, vibration, etc... The point I'm making, is before you jump off the bridge make sure your chute is properly packed, latched, the wind is right, and there are no crocodiles in the river below you.

Best of luck :

Oh, to add to this....how many points should be verified in the range, once again this depends...typically you want to verify an instrument over it's full range, acceptable practice in most cases is to do five points over the range of the instrument.

Benjamin, that was a superb post.

I would like to add support to Benjamin's post. If you took some time to add up your costs, is your company sure you will be saving money? There are a lot of costs associated with maintaining an internal calibration facility.

Suggestion:
Go through and look at Hershal's, Jerry's, and others posts in the calibration/ISO17025 forum. My point I am trying to make is there are a lot of variables in calibration, training, environment, etc. that will influence properly performing internal calibrations.

As far as your standards, you need to keep standards that are superior in accuracy to the instruments you are calibrating. Hence the need for gauge block standards, etc.

But.. if this is a mandatory reality in your situation, here are some suggestions:

1. Determine the costs for developing an appropriate calibration environment (temperature, humidity, vibration, etc.)

2. Determine the costs for obtaining/maintaining standards for the desired purpose. This includes downtime, calculating uncertainties, redundant instruments, replacement costs, etc.

3. Determine the costs for obtaining procedures, labor, training materials, cost of training, etc.

4. Make sure that this initiative will be funded for several years to come.

Providing procedures is a whole other matter. It will depend on the manufacture/model of the various equipment. If you are doing government contracts, then you may can access GIDEP for procedures. You may can contact the MFG. of the equipment and see if they have procedures available.

I hope I have not frustrated you. It's just I've seen so many times that organizations are saving a whole lot less money through internal calibrations than they may think. But, we're here to help, if that is what you want to do.

Benjamin28 said:

One thing I'd like to point out however, comparing an externally calibrated instrument against an operational instrument of the same family may not be the best route to take with this. Such as a micrometer, if you're measuring a part with your master mic, and then comparing against your operational mic, this is all well and good, however, you'd be better off just using a NIST traceable gage block as the block will be uniform, calibrated...if you're using an appropriate gage block there is no need to have a "master mic" in this case and you'll be able to calibrate/verify most of your linear measurement instruments off this set of blocks, thus one reference standard, calibrated and traceable, is used for multiple instruments and the cost/upkeep will likely be lower and easier to maintain. Be sure whatever methods you use are acceptable by your customers and registrar...if they think you're just doing op checks vs a thorough calibration/verification they may not find this acceptable.

Click to expand...

I believe the rule of thumb is that the standard be 4:1 more accurate (or, more precisely collective uncertainty) than the instrument being calibrated. For example, that is typically why gage blocks (of the correct grade) are used for standard for hand gages. It would also eliminate the comparative approach for calibration, although it might be used as an awkward verification. Don't forget, you need to do various lengths across the range of the device for linearity, etc. And, other checks may need to be done that your approach will miss, such as flatness (wear) across micrometer anvils usually require optical flats and its associated monochromatic light, etc. It may never show up on comparative checks.

One suggestion: get calibration training before calibrating. Pure and simple.