Calibration

RoboTech · 09-05-2003, 07:06 AM

Here is where we discuss calibrating equipment...

Both initial and ongoing calibration is essential to maintain accuracy and confidence. Post your thoughts on frequency, techniques and sources.

Gooserider · 10-08-2003, 08:59 PM

I've worked in many different places with equipment having different calibration needs.

All of them used a professional calibration house, which I don't see any way around, as otherwise the question of maintaining the standards against which one does the calibration becomes an issue.

Calibration procedures and frequencies are normally defined by the manufacturers of the equipment, though it can sometimes be a challenge to get that information (mostly on things that didn't have cal requirements) as I found on my last job where I had to help document calibration requirements for ISO certification. Fairly typical on the stuff I've worked with is annual calibration. Note that some equipment does not require calibration, especially digital gear.

It may be worth considering calibration requirements when selecting equipment - something with longer calibration intervals is less likely to get out of order between cals, and the savings on calibrations might justify paying more for a given peice of gear.

Gooserider

bigben2k · 10-09-2003, 12:15 AM

We actually have more than one type of calibration to deal with: first the various measuring equipment (DMMs, temp readers), then the test bench as a whole (should be done before each session starts).

For calibrating my temp probes, I'm thinking about using a glass-mercury thermometer that's ASTM standard. It should allow me to dunk it and a thermal probe in a cup of water (a la pHaestus) and see the difference. The accuracy of the thermometer is +/- 0.1 deg C.

Not sure if there would be a benefit to having it NIST certified to 25 deg C (coolant temp): it costs ~$60 .

Then there's the DMMs...

Gooserider · 10-10-2003, 12:11 AM

Quote:

bigben2k We actually have more than one type of calibration to deal with: first the various measuring equipment (DMMs, temp readers), then the test bench as a whole (should be done before each session starts).

Calibrating the bench is a bit outside my knowledge, I would assume that you would do something like making a short run with a known block to ensure that you get the same results each time.

As to the instrument calibration, I know that it is more expensive, but if I were doing the WBTA standards, I would insist that a 'standards compliant' test bench would have NIST traceable calibration certs on all equipment that requred it. IMHO not having such a trace would greatly reduce the credibility of the test results, and make cross testing more difficult.

Gooserider

murray13 · 10-10-2003, 08:13 AM

Two things:

A one point calibration on a thermometer is almost useless, it only lets you use the thermometer at one point. NONE of the rest of the range of that thermometer is of any use as it is NOT calibrated. The only use I've seen for one point cal's of thermometers is at 0C for use in monitoring a ice bath. When using a one point cal thermometer you must change the temperature of the medium the thermometer is in to the calibration temperature. Only then is the certification of that thermometer valid.

Calibrating the test bench is a much more complex issue. What are you trying to get out of calibrating the bench? Is it just the same readings for the same block tested many times. Is it that someone else's bench could come up with the same values. Is it to be able to certify the test object to within a certain specification. All of these would require some sort of calibration, some MUCH more than others.

It basically comes down to cost. How much is someone going to have to charge to keep a test bench NIST certified.

As a former metroligist I can tell you this is one area more complicated than you first think. While I don't quite have the time to do a work up on this I will keep an eye on this thread and try and keep you guys on track. Good luck.

BillA · 10-10-2003, 09:18 AM

we spend over $1500 pa on calibration, and to be in strict compliance it should be double that

quite correct on the temp range issue
each thermometer has its curve and each RTD its also
- knowing the actual temp, or difference, requires that the 2 devices (or 4) be corrected for their offset at that specific point
what the indicator says is merely the starting point in the determination of the temperature

09-05-2003, 07:06 AM	#1
RoboTech Cooling Savant Join Date: Sep 2002 Location: Cincinnati, OH Posts: 229	Calibration Here is where we discuss calibrating equipment... Both initial and ongoing calibration is essential to maintain accuracy and confidence. Post your thoughts on frequency, techniques and sources.

10-08-2003, 08:59 PM	#2
Gooserider Cooling Savant Join Date: Jun 2003 Location: North Billerica, MA, USA Posts: 451	I've worked in many different places with equipment having different calibration needs. All of them used a professional calibration house, which I don't see any way around, as otherwise the question of maintaining the standards against which one does the calibration becomes an issue. Calibration procedures and frequencies are normally defined by the manufacturers of the equipment, though it can sometimes be a challenge to get that information (mostly on things that didn't have cal requirements) as I found on my last job where I had to help document calibration requirements for ISO certification. Fairly typical on the stuff I've worked with is annual calibration. Note that some equipment does not require calibration, especially digital gear. It may be worth considering calibration requirements when selecting equipment - something with longer calibration intervals is less likely to get out of order between cals, and the savings on calibrations might justify paying more for a given peice of gear. Gooserider __________________ Designing system, will have Tyan S2468UGN Dual Athlon MOBO, SCSI HDDS, other goodies. Will run LINUX only. Want to have silent running, minimal fans, and water cooled. Probably not OC'c

10-09-2003, 12:15 AM	#3
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	We actually have more than one type of calibration to deal with: first the various measuring equipment (DMMs, temp readers), then the test bench as a whole (should be done before each session starts). For calibrating my temp probes, I'm thinking about using a glass-mercury thermometer that's ASTM standard. It should allow me to dunk it and a thermal probe in a cup of water (a la pHaestus) and see the difference. The accuracy of the thermometer is +/- 0.1 deg C. Not sure if there would be a benefit to having it NIST certified to 25 deg C (coolant temp): it costs ~$60 . Then there's the DMMs... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

10-10-2003, 08:13 AM	#5
murray13 Cooling Savant Join Date: Sep 2002 Location: Omaha, NE USA Posts: 216	Two things: A one point calibration on a thermometer is almost useless, it only lets you use the thermometer at one point. NONE of the rest of the range of that thermometer is of any use as it is NOT calibrated. The only use I've seen for one point cal's of thermometers is at 0C for use in monitoring a ice bath. When using a one point cal thermometer you must change the temperature of the medium the thermometer is in to the calibration temperature. Only then is the certification of that thermometer valid. Calibrating the test bench is a much more complex issue. What are you trying to get out of calibrating the bench? Is it just the same readings for the same block tested many times. Is it that someone else's bench could come up with the same values. Is it to be able to certify the test object to within a certain specification. All of these would require some sort of calibration, some MUCH more than others. It basically comes down to cost. How much is someone going to have to charge to keep a test bench NIST certified. As a former metroligist I can tell you this is one area more complicated than you first think. While I don't quite have the time to do a work up on this I will keep an eye on this thread and try and keep you guys on track. Good luck.

10-10-2003, 09:18 AM	#6
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	we spend over $1500 pa on calibration, and to be in strict compliance it should be double that quite correct on the temp range issue each thermometer has its curve and each RTD its also - knowing the actual temp, or difference, requires that the 2 devices (or 4) be corrected for their offset at that specific point what the indicator says is merely the starting point in the determination of the temperature

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)