Six axes of calibration service

Want to make APP Same thing as WeChat , Can run small programs smoothly ？ | Experience will send you to Xinjiang 、 Huawei 、 Cherry keyboard ！>>>

This issue is from Keysight, There is no change in the original text ！

----------------------------------------------

Abstract

Choosing a calibration service provider seems to be a very simple matter , It seems that calibration is just calibration , As long as the supplier has a normal business license （ISO9000 registration 、ISO17025 Qualification, etc ） So no matter what kind of service or supplier you choose , The services provided should be similar . But that's not the case , Due to the lack of regulation of calibration services , As a result, the service provided and the calibration results are uneven .

introduction

To help illustrate the difference , Instruct purchasers who are not familiar with the technology to understand the importance of all aspects of calibration services , We split the calibration service into six different axes , Each axis affects the practicality of calibration or the direct or indirect cost of choosing calibration services . Each axis is discussed in depth in this paper , They are ：

1. The scope of the test （ How many parameters to test ？ How many test points need to be tested for each parameter ？）

2. Information provided （ What data is provided ？ Is it the data before adjustment or after adjustment ？）

3. Measurement uncertainty （ How accurate is the test ？）

4. cycle （ How often do you need to calibrate ？ Can you change this time interval ？）

5. Speed （ How long can you tolerate not using the instrument ？）

6. Supporting services （ What are the supporting solutions and services ？ What items are included in these services ？）

axis 1. Test range

Obviously , The more comprehensive the instrument is tested , The more confident the instrument performs the measurement . Generally speaking , The manufacturer recommends which parameters should be tested and the number of test points that should be checked for each parameter . However , It is not uncommon for low-cost suppliers to ignore many of these tests . Since there is generally no technical agency in our industry to specify the scope of testing required to perform legal calibration , Therefore, the above practice does not violate any regulations or guidelines .

There are three main reasons for ignoring some of these tests ：

1. The supplier does not have the equipment to perform the test

2. The test is too complicated , Execution time is too long , Or require specialized skills

3. The negotiated price does not allow all tests to be performed

Usually this means that calibration services cannot adequately test important parts of the instrument , As a result, , Users may be unaware of , Measurements performed with these devices may be wrong .

When testing products with these devices , There can be two harmful situations ：

1. Qualified products may be judged as unqualified during the test , Lead to costly rework , Improve the rework department NFF（ No fault found ） Probability （ This is a key indicator in the production cycle ）.

2. Unqualified products may be judged as qualified during the test , This leads to an increase in the number of warranty returns , The cost of repair has increased , At the same time, it damages the company's image and reputation .

example

A company sent a scalar analyzer probe to a famous company for calibration . The company's calibration process is at the low end of the range 、 Middle end and high end three point frequency test equipment . The test report shows that , The performance of the equipment is in accordance with the specifications , And issued a calibration certificate . But the customer had strange measurements when using the device , So it sent the device to Shide Technology , Calibrate again .

The program of SciTech is to use sweep frequency to test the whole range of the equipment . This test clearly shows , The probe has problems at frequencies not tested by the first supplier . If you use sweep calibration from the beginning , Customers will save a lot of time and money .

A metaphor can better explain all this ： On a sailing ship , The watchman looked out through the mast , He first looked through a telescope at Zhengdong , And then I took a look at due north , Look at the west again , Finally, the report says “ I didn't see the ship ”.

exception

It must be pointed out , If you control it properly , Reducing testing is not necessarily a bad thing . For devices used in specific applications , If only part of the well-defined functions of the instrument are used , Reducing testing is certainly acceptable , And it could offer a big advantage , Because it can focus calibration on these specific functions or ranges . Especially if compared to full calibration , More comprehensive testing of these areas .

however , This method may not be suitable for multi-purpose or general purpose equipment .

axis 2. Information provided

Information provided by the calibration supplier relating to the tests performed during the calibration process , It's the basis for users to make many decisions , So get the information you need 、 At the same time, it is very important to ensure the accuracy of the information .

Users want calibration suppliers to provide comprehensive data , Show the performance of the device at each test point . these

The information indicates the scope of the test , Can be used to perform an objective comparison between suppliers .

But the value of information is more important . When the calibration supplier found that the performance of the instrument fell outside the specified allowable error range （OOT—— Out of tolerance / Out of tolerance ） when , This means that the user may have used the device to perform some unqualified measurements . To analyze whether this will impair the effectiveness of these completed tests , You have to know which parameters the instrument falls outside the allowable error range , And how far out of range . Then you can decide if corrective action is needed , Such as retesting or recalling your product . Without this information , You may not know that the test has made errors , Or maybe the calibration report simply states that the device is out of tolerance , But there was no support

data , Then you won't be able to make informed decisions .

The really useful data is when the calibration supplier receives the equipment , A complete set of performance test information collected by performing measurements on the equipment . During this test , Any adjustment to the device will destroy the integrity of the test in the subsequent program . The second set of data collected after adjustment will confirm that the instrument is within the allowable error range after adjustment , To provide confidence that the device will be used to perform tests in the future .

Another key message is “ Measurement uncertainty ”. Its simplest form is a statement , State that the equipment used to test the user's instrument is at least as accurate as the calibrated instrument 4 times .（4 Generally considered to be the lowest test accuracy ratio for calibration services , namely TAR）.

The calibration supplier will provide more detailed uncertainty information through the summary table , The parameters and associated uncertainties are listed . Parameter values can be relative values （ Such as ± 0.5%）, It can also be the absolute value of the unit of measurement （ Such as ± 0.2 mV）.

Next to the measurement data is the most detailed uncertainty information , Ideally , The measured values shall be provided in the same item 、 Technical index range and uncertainty information .

axis 3. Measurement uncertainty

The measurement uncertainty can indicate the quality of calibration service testing , That is to calibrate the accuracy of the measurements performed by the supplier . This axis is invisible , Unless the measurement uncertainty is known , But it can have a significant impact on whether calibration is really useful .

The uncertainty clearly indicates whether the supplier has the technical qualification to perform the measurement . The uncertainty can be evaluated by the supplier itself , It can also be assessed by a certification body . Recognized uncertainty is more important , Because it is independently evaluated by a recognized certification body , Such as the United Kingdom UKAS、 Japan JCSS、 Germany DKD、 China CNAS And the U.S. A2LA wait .

You can examine the best measurement uncertainty data from an accredited Calibration Laboratory , Because the information is public , It is usually available through the website of a certification body . Through this information , You can know the technical level of the laboratory , And compare the ability of laboratories to perform certain measurements . This allows you to conclude that some laboratories may not have the real qualifications to perform certain calibrations .

In terms of certification , Some laboratories may only be certified on certain tests , But accepted equipment calibration beyond the scope of certification , This is not uncommon . They may only get one measurement certification , But they're going to show the certificate , Claiming that they are a certified Laboratory , And in fact, they don't have the ability to calibrate most instruments .

Illustrate with examples

By assessing the UK calibration supplier's certification arrangements or scope on two comparable basis measurements , You can see very interesting results .

The measurement we chose was 10 MHz The frequency and 1 V DC. This information comes from the UK certification service (UKAS) Website .

stay 10 MHz Under the circumstances , The uncertainty range of certification is very large . The best range of uncertainty is 10-12, The worst way to reach 3 x 10-6. So when measuring 10 MHz when , The most accurate results may also have uncertainties ± 0.00001 Hz, In the worst case, the uncertainty may even reach ± 30 Hz. Due to the technical specifications of most crystal oscillators ± 30 Hz Be narrow , So obviously , The company's calibration of crystal oscillators or instruments with certain frequency measurement accuracy is meaningless .

stay 1 V DC Under the circumstances , The range of uncertainty is relatively narrow , But it's still very obvious , The best is 4 x 10-7, The worst is 2 x 10-5. The resolution of many digital multimeters is 6½ position . in other words , It should be able to measure 1 V DC Is shown as 1.000000 V（ That is, accurate to 1 Microvolt ）.4 x 10-7 The uncertainty will be realized 0.4 Microvolt accuracy , This is accurate enough for the measurement task .2 x 10-5 Uncertainty can only be achieved 20 Microvolt measurement accuracy . In this case , The last one on the multimeter is completely untrustworthy .

It should be noted that , released “ Best measurement qualification ” It is not equivalent to the actual uncertainty specified for the calibration of a product . By comparing some typical certificates / Report examples , It allows us to understand the essence of the problem from another perspective .

axis 4. Calibration cycle

The calibration cycle is how often the instrument is calibrated . In general, the manufacturer recommends the calibration cycle for each instrument , It's usually a year or two . This is based on the average drift rate of various components . In the development process , What designers are most concerned about is the design with the least drift , But drift is a factor that cannot be completely eliminated , That's why calibration services are essential .

In order to analyze whether it is possible to calibrate without following the manufacturer's recommended period , You need to know two things ：

1. Calibration history of an instrument

2. The use of instruments

The first point is to refer directly to the information provided by the calibration service discussed earlier （ axis 2）. A widely accepted algorithm is , If you calibrate it three times in a row , The technical specifications of the instrument are all qualified , Then the calibration period can be relaxed appropriately . Empathy , If you calibrate twice in a row , We need to adjust the instrument , Then the calibration cycle should be shortened . And so on .

The second point is about risk . If the measurement performed by the instrument is critical , Then the calibration cycle must be shortened , To reduce the risk of poor measurement results . If the measurement results are not good or bad , Then the calibration period can be extended appropriately .

When both risk and calibration history are important , An efficient cycle management solution can be adopted , This minimizes the risk of deterioration of key measurement results , Control costs at the same time .

axis 5. Speed

On time at school , Generally, the instrument needs to be transported away from the user's place of use , Send it to another place for calibration, which usually takes a week or more to stop using the instrument . in other words , The user either stops what the instrument is doing , Or in a critical transactional environment , Buy or rent a spare instrument .

Knowing that the instrument will be moved for two weeks , It means you can plan ahead , Arrange for the use of spare or rental instruments . But if the two weeks promised by the supplier become three weeks or more , The cost will be greatly increased , You will lose control . therefore , Users would like calibration suppliers to promise a very short delivery turnaround time , It must also be able to deliver on promises reliably .

If the company can provide on-site calibration service , You can shorten the turnaround time to a few hours , In fact, it's just the time it takes to calibrate . This service may be expensive , But compared with the cost savings of purchasing and leasing spare instruments , The added cost is not much .

I want to remind you that . Field calibration seems to be an ideal solution , However, it is still necessary to ensure that the calibration supplier's technical level can meet your needs . Companies that provide on-site calibration services usually offer only a fraction of the tests , It may also bring a lot of measurement uncertainty . Only when they arrive at your place , You will find that , They carry limited testing equipment , The ability to test is even more inadequate .

Whether it's in the service center or at the customer site , We can provide high quality calibration service of the same standard .

axis 6. Supporting services

When it comes to calibration costs , The price of basic services is usually the main consideration . But if you choose supporting services , It can reduce the hidden cost , For example, to submit an order 、 Pay the invoice and other related costs , Or reduce the service price by signing a service agreement .

Is Germany technology to provide a wide range of supporting services , We hope to help you reduce the direct and indirect costs of purchasing calibration services . Many services can spread the service charge evenly over a year , Avoid peaks and lows in your spending .

There are other factors to consider in this regard . Shipping costs and routine repair services should be considered . If these items are included in the supporting services , Then you'll save a lot of money . conversely , The extra cost will surprise you .

example

A company calibrated AVO Meter （ Universal Analog Multimeter ） The quotation is every time 15 dollar . After a while , The customer found out , His AVO The instrument calibration fee has always been 40 dollar .

The customer didn't know until he asked , stay AVO The battery needs to be replaced during instrument calibration . According to the rules of this supplier , Replacing the battery is equivalent to a repair service , Additional charge per visit 25 dollar .

In this case , Although the amount of expenses per occurrence is very small , but 12 Months add up to , The total amount will be very considerable .

Conclusion

In the center of the six axes , Each axis affects the cost and quality of calibration services . I hope this article can help you make informed decisions , Not exactly on the seventh axis ―― The price will decide .

It's the same as shopping , You get what you pay for . Understand the content of calibration service and its significance to your company , Can make your decision more scientific , Little detours .

disclaimer ：

The official account is purely personal account , Only for the purpose of technology sharing and exchange .

The information in this issue comes from Keysight, If you intrude, please delete the message immediately , thank you ！

This article is from WeChat official account. - Test those things （Jackiy_455750304）.
If there is any infringement , Please contact the [email protected] Delete .
Participation of this paper “OSC Source creation plan ”, You are welcome to join us , share .