why its important to calibrate the measurement equipment
So measurements can be trusted. Often most of us only need relative measurements, i.e. to prove a square waves is generally square, but often for professional environments engineers have to trust their equipment, and use results in documentation so accuracy is required.
So measurements can be trusted. Often most of us only need relative measurements, i.e. to prove a square waves is generally square, but often for professional environments engineers have to trust their equipment, and use results in documentation so accuracy is required.
so the calibrate restore the accuracy of the measurement equipment
Fundamentally, yes. All equipment will drift a little, even the highest stability £££ gear, so the calibration means to correct for that drift with a known quantity, i.e. a precise voltage reference or whatever.
There are different interpretations of what calibrated actually means...
From a business point of view it makes sense to set up annual calibration checks for most items of test gear. However some items of test gear can have shorter or longer recommended calibration intervals. But usually the test gear just gets checked rather than adjusted.
At my place of work nobody enjoys losing test gear when it is sent for calibration. Nobody gets excited when it comes back either (from an accuracy point of view) because we all secretly hope that it just passes the checks and no spotty calibration tech has been inside it learning his/her trade. We just want it back if it is currently needed for a task.
For home/hobby use there isn't much point sending stuff for professional calibration checks like this.
To me 'calibration " is part of a quality control system from' cradle to grave', checking is more useful for the 'enthusiastic hobbyist' like me.
i vote .
having an uncalibrated piece of test gear is like grabbing the first stick you find and claim : this is my meter/ yard.
if everyone does that we're going to end up with some pretty funky constructions.
But a lot depends on what is meant by calibrated and uncalibrated.
Taken literally, an uncalibrated (effectively unscaled?) voltmeter is of limited use. But in the real world, none of my DMMs are calibrated yet they remain extremely useful because they are designed to provide an (adequately) reliable indication of A, V, O for many decades.
I only have one item of test gear here that is formally calibrated because it came with a 2 year cal certificate when I bought it as an ex rental sig gen. I didn't ask for or expect this 2 year cal certificate. But apart from that, all of my test gear is 'uncalibrated' in that it hasn't been to a calibration house in many years.
This doesn't stop me doing valid research work or designing or testing things successfully at home. A formal calibration will only add a tiny amount of 'value' to my test gear and the important thing is to be able to understand and manage measurement uncertainty and to still do 'useful' and valid work with test gear that isn't formally calibrated.
There are also ways to cross check test gear at home to improve confidence in the measurement.
If you trust them. I had issues with our calibration shop. I sent a scope with a dead channel in. It came back with a dead channel. Sent it back and the tech called me up and said he couldn't find a problem with it. Told him to connect channel 1 to to the test pin on the front of the scope. Then he saw it, told me they only test with 100KHZ and it tested fine with that. Wouldn't call that testing worth much.
why its important to calibrate the measurement equipment
What if you bought plans for a house, and your tape measure was off by an inch (uncalibrated)?
If you built the house using that tape measure, and then tried to install the windows which came from a manufacturer (that had calibrated measurements) what would happen?
The windows would not fit, i.e., the house would not "work".
Many companies sell product and share profits via contract (oil company is one example) that require verification/calibration on an agreed frequency. When money is involved all parties have a right to want accuracy to some agreed standard. The gas pump you use most likely has a dated calibration sticker on it showing it last being checked/measured. The equipment used to check such gas pumps also has to have it's own calibration/verification process. It can be a very big deal in industry, but for hobbyist use, it's mostly up to the individual how he/she goes about 'trusting' their measurement equipment.
I think the real issue here is when inexperienced users think that their test gear is 'accurate' because they just received it back from a calibration house. In reality, it usually means the test gear is proven to be within certain limits. In the case of RF test gear these limits are often very slack.
But an inexperienced user will often not realise this and may be seduced by the brand new calibration sticker and they will be convinced that the reading it gives will be very accurate. This is despite the reality above and also despite their lack of doing a basic uncertainty analysis of their overall test setup.
There will always be people who will cling to their 'perfect world' concept of 'calibration' and they can safely argue their point from deep inside their ivory tower. The real world is very different to this, however
Isn't calibration only valid if a whole series of other conditions are met, eg switched on for 30+mins, controlled temperature etc, it is also quite expensive.
In its basic sense was about matching parts from different manufacturers to fit together always by means of standards. See caliber of guns, the "happy killing rainbow devices ltd., china" manufactured bullet has to pass the gun just like the one from "United Ammunitions, mouth-lathed gold plated bullets division, USA-USA".
The most important kind of calibration I am aware is the volume of beer glasses, by the way - which cant be done by any company but the state by its Eichamt
Literally, calibration is the test if under defined conditions, a device under test matches a measurable property of a standard - which is necessarily again compared against another standard, and by how much it is off. A maximum size cannon ball needs to fit the gun. A minimum size cannon ball needs to work correctly, too.
Many companies sell product and share profits via contract (oil company is one example) that require verification/calibration on an agreed frequency. When money is involved all parties have a right to want accuracy to some agreed standard. The gas pump you use most likely has a dated calibration sticker on it showing it last being checked/measured. The equipment used to check such gas pumps also has to have it's own calibration/verification process. It can be a very big deal in industry, but for hobbyist use, it's mostly up to the individual how he/she goes about 'trusting' their measurement equipment.
I knew one garage where the one pump had some internal wear, in that it drooped quite badly at a low flow rate from fuel bypassing the positive displacement pump seals internally. Give it a good flow rate, like you normally used to fill the tank, and it was accurate. Slow down to a trickle, like I did when the tank breather blocked with dirt, and I filled the tank from empty to full in 10 minutes, but only put in, according to the fuel display, 10l. I did phone the regional office the next day, talked to technical services and told them of the issue ( here in South Africa the fuel stations are owned by an oil company, and the pumps are oil company property, not part of the station, which is a separate business) and about a wek later when I went in there that pump was a complete new one, a complete replacement from the floor up of the new styling as opposed to the rest which were 2 stylings old.
All Tokheim pumps in there, I have seen some which, in a brand new skin, were older than me, having manufacturing dates from the 1950's, and still having a mechanical gallon counter on the pump itself, which has been around the clock a lot of times. Old pump, new shiny outside wreapped around it.
why its important to calibrate the measurement equipment
What if you bought plans for a house, and your tape measure was off by an inch (uncalibrated)?
If you built the house using that tape measure, and then tried to install the windows which came from a manufacturer (that had calibrated measurements) what would happen?
The windows would not fit, i.e., the house would not "work".
One way of looking at it, is that absolute accuracy (and therefore calibration) is only necessary to the extent that other people need to work with your measurements. If you made the windows yourself, cut the glass, etc, rather than deal with another party (the window manufacturer), you would still be fine. (Of course, you could still run into trouble if your home made measuring stick expanded too much when you left it in the sun - i.e. it was not sufficiently repeatable)
The pioneers in many fields of science did not have the luxury of calibrated measuring instruments - in some cases, they would have had to design and build the instruments themselves, as well as choose a standard for their unit of measurement, since existing devices did not exist.
so the calibrate restore the accuracy of the measurement equipment
Not exactly.
There is calibration checking, and calibration adjustment, they are two different things.
Most of the time when you get an instrument "calibrated" they do
not adjust it, they check it against the specs and gives you the results.
When you do this enough times you can build up confidence in your instrument that it does not drift and will be within specification.
Calibration is essentially all about
measurement confidence.
If you trust them. I had issues with our calibration shop. I sent a scope with a dead channel in. It came back with a dead channel. Sent it back and the tech called me up and said he couldn't find a problem with it. Told him to connect channel 1 to to the test pin on the front of the scope. Then he saw it, told me they only test with 100KHZ and it tested fine with that. Wouldn't call that testing worth much.
Thats often the difference between cheap and expensive calibration labs. Some will follow the exact manufacturer test/verification procedures, others are only doing a minimum to fill up a data sheet.
For me, it has to do with troubleshooting. Troubleshooting is basically making a list of all the things that could go wrong. Unfortunately, a lot of possible culprits have a mutual influence on each other, making troubleshooting less straight forward.
So, the last thing you want to do is toss in another parameter you're not sure off. Ie, your measurement equipment
Other examples of requiring absolute accuracy.
Your ICs keep failing even though your voltmeter says applied volts are within spec.
You don't hear anything on selected frequency, after you just set it up with your shiny new RF signal generator.
You are buying gold at multi-thousands of dollars an ounce, using a scale that was clearly also used to weigh Roman trophy gold.
"Calibration" doesn't always mean absolute accuracy. Many moons ago our lab was shut down for a week or more when all three examples of a key instrument went out of their calibration interval at once. (How that happened is another fun story). Heavy management pressure resulted in a rush turnaround and all three arrived back the same day. Results seemed funny so connected all three instruments to the same source. Instrument A read 2X the value of instrument B. Instrument C read 1/2 the value of instrument B. Called the calibration lab and said WTF? Their response - tell us which one is right and we'll be happy to adjust them all to read the same!
