[SOLVED] Three point calibration in pH meter

[samic45mit1]

Hello,
I am working on a pH meter using a micro controller. Please tell me about the three point calibration process and its related math.Please also suggest some good books for instrument calibration and error correction

 

[tpetar]

Maybe this is useful for You :

pH electrode calibration

pH Calibration Procedure

How do I calibrate my pH electrode?

**broken link removed**

http://www.cbeng.com/resources/app_notes/FAQs_on_pH_sensors.pdf

 

[jpanhalt]

The resources cited above give differing views on the purpose of the third point in pH meter calibration.

The Monash Scientific discussion says the 3rd point is essentially for quality control. The first and fourth references say almost nothing about the "third point" or its purpose The fifth reference suggests the third point is really a 2-point calibration using an alkaline buffer (pH 10 usually) instead
of the acidic (pH 4) buffer. Wikipedia (pH meter - Wikipedia, the free encyclopedia) suggests that three points or more may improve accuracy (that way is too vague considering the chemistry in a pH electrode to instill much confidence). This (http://www.beckmancoulter.com/wsrportal/techdocs?docname=A59443AA.pdf) is a large Beckman reference. It describes multipoint calibrations, but the application is mostly to ion-selective electrode (ISE) measurements. As for pH, Beckman mentions a third-point as part of a temperature calibration of its electrode.

Unfortunuately many references, such as those from ISO or CLSI (USA, Standard C46-A2: **broken link removed** ) are only available for purchase and my copies of the CLSI documents are both way out of date and buried to deep to find right now.

The best source I found was from Radiometer ( **broken link removed** )

Source:Radiometer (emphasis added)
Temperature plays an important role as regards sample and buffer pH and an electrode's characteristics. The temperature dependency of the buffers is fully known and is shown on the rear of the buffer bottles from Radiometer Analytical. The pH variation due to temperature is minimal for inorganic acid buffers, whereas it is significant for alkaline buffers and some organic buffers (please see the buffer tables in the Appendix). As regards the electrodes, compensation can be made for the influence of temperature on the slope. On the other hand, no compensation can been made for the pH shifts caused by altered reference potentials or a change of pH in the inner solution in the glass bulb. Finally, almost nothing is known about the influence of temperature on a sample's pH. It is therefore essential that the temperature is registered together with the pH value. To sum up, samples, buffers and electrodes should all have the same temperature. Some compensation can be performed but it is not possible to calculate the pH of a sample measured at one temperature back to the sample pH at another (reference) temperature.

I would agree with the comments from Radiometer and Monash with the provision that if the instrument was being used in a highly regulated industry, such as a medical laboratory, I would follow the applicable regulation. The Radiometer document also give some of the math involved.

John

 

[FvM]

The resources cited above give differing views on the purpose of the third point in pH meter calibration.

Yes. In fact different methods are in use. It should be considered that pH electrodes are instrisically linear up to pH 12, so that the purpose of calculating separate slopes for the acidic and alkaline range can be doubted. More likely, accidental errors of the buffer solutions would take effect. For this reason, calculating a line of best fit when using three or more calibration points is usually the best method. It's e.g. suggested by the German standard DIN 19268.

 

[jpanhalt]

I respectfully disagree. Two points are adequate for calibration of any liner determination. Running a quality control (i.e., a third point) is also needed in some environments. If the meter is not linear, it should be repaired. In other words, I agree with Monash. Nevertheless, when the US government (CLIA 1988) said that linear instruments had to have 3-point calibrations, I did it and focused my efforts on other things.

More likely, accidental errors of the buffer solutions would take effect.

I would put that in the category of quality control. If you have to triple-test every calibration standard, you end up in an exponential loop of non-productive testing.

John

 

[FvM]

Two points are adequate for calibration of any linear determination.

Of course they are, but that wasn't the point of discussion. The question is, if you decide to do a third calibration point, may be intended primarly as a check, how can you take maximum advantage of it, assuming all three calibration points have basically the same uncertainty?

Using the third point only as a check discards part of the available accuracy. That's why the said DIN standard suggest a line of best fit, which is the appropriate method to get the best available accuracy from multiple calibrations. That doesn't mean that you necessarily need to use it if you are satisfied with a two-point calibration, but the method is straightforward and easy to implement. In my opinion, it's the preferred method if a three point calibration is intended. You get the the intended check information as coefficient of determination from the fit additionally.

 

[samic45mit1]

Thank you All of you
But I want to know the maths behind this process.Some of you saying that the two point is sufficient,Please note that some of company are providing five point cal.If any one can tell me about some books releated with this OR just the maths formula used for this
Thankyou

 

[FvM]

You should primarly search the instrument manuals for an exact explanation of what they are doing. I think, a user can expect it.

The line of best fit, also called linear regression method, that's suggested by some standards, can be found in high school and universtity mathematics or statistics handbooks. See also Linear regression - Wikipedia, the free encyclopedia