Dairy Calibration Standards
Calibration & Proficiency Products & Services
LABTEK Services has been supporting the UK Dairy infra-red and photometric analyser for over 30years.
In 2008, as a result of not being able to source accurate and reliable calibration samples for our engineers, we decided to introduce our own.
We now partner with Qlip (Netherlands), the largest accredited milk testing laboratory in Europe.
Control Samples – Take control of your FTIR / IR or NIR Analyser
Making sure you get the readings you need on your milk products is our priority. We use Qlip to provide us with the very best and most accurate control samples for all your Fat, Lactose & Protein tests to ensure that your milk and dairy composition is exactly what you think it is, and what is required by ISO governance.
Uniquely we can supply a wide range of low cost, ISO certificated control samples which enable the user to monitor the FTIR / IR or NIR analyser on daily basis between calibrations.
CONTROL Samples are available for:-
- Raw Milk
- Processed Milk 1.5%
- Processed Milk 3.5%
- Skimmed Milk
- Cream 10%, 21%, 31%, 35%, 40%, 48%, 52%
Qlip undertakes over 200,000 Rose Gottlieb(Fat%) & Kjeldahl (protein%) each year enabling them to confidently apply an Uncertainty of Measurement of +/- 0.01% for fat & protein.
- Fat: +/- 0,01%m/m
- Protein: +/- 0,01%m/m
- Lactose: +/- 0,02%m/m
These levels comply with the NEN-EN-ISO 1222 (Fat), ISO 8968-1 (Protein) and ISO 22662 (Lactose).
Milk & Cream Calibration Standards
The Qlip calibration sets have proven to be popular. A number of national milk groups and internationally recognised institutions are now using us for their monthly calibrations and instrument monitoring.
What sets us apart? LABTEK engineers are able to assist both product support and instrument support if required.
We can offer a range of calibration sets that are suitable for all infra-red dairy analysers, such as the Lactolyser, Lactoscope FTIR & Filter, FOSS F1,F2, FT120 Analysers, and Bentley Analysers:
- Raw Milk Fat % – 7 samples ranging from 0-6%
- Total Solids & Casein Certification for Raw Milk Samples
- Raw Milk Protein % – 5 samples ranging from 2-5%
- Raw Milk Lactose % – 5 samples ranging from 3-6%
- Skim Milk Calibration Set Fat, Protein & Total Solids
- Cream Extended Range 10-48%
- Cream Low Fat % – 5 samples ranging from 23-31%
- Cream High Fat % – 6 samples ranging from 33-44%
- Cream Total Solids % Certification
- Raw Milk FPD – 8 samples
- Whey Fat % – 5 Samples ranging from 0.05 – 0.35%
- Whey Protein % – 5 Samples ranging from 0.5 -1.1%
Understanding Calibration and Inter-correction of Infra-Red InstrumentationWe understand the need for precise measurements when testing the composition of your dairy products. Whenever a machine leaves the factory, it will have been calibrated against certain criteria on a wide range of fats, lactose and protein samples, but this will wane over time through wear and tear. How do you know that the machine you are using is still as accurate as it was on day one? A well calibrated infra-red instrument, should have a slope close to 1.0 and a bias close to zero on all milk channels.
Extended Range is critical to establishing an accurate calibration curve.
We are seeing an increasing number of calibration sample sets that simply don’t have the level of accuracy and stability that our qualified engineers would recommend for instruments. They can have very little range in their protein and lactose samples and this tends to confuse self-calibrating instruments.
Commercially Available Samples e.g. PROTEIN 3.30 – 3.60% LACTOSE 4.45 – 4.75%
LABTEK QLIP Samples e.g. PROTEIN 2.00 – 5.00% LACTOSE 2.97- 5.79%
Any instrument that has a slope of less than 0.75 and a bias greater than 0.75 either has a fault or is incorrectly calibrated. This is true for Fat, Protein & Lactose of any standard milk (i.e whole, semi or skimmed).
An Example:If the ‘commercially available’ samples mentioned above, were used when calibrating your machine and all the samples above 3.4% had a -0.03 bias and all the samples below 3.4% had a bias of +0.03 then it might appear to be a good calibration. The + & – would cancel each other out and you would end up with a zero mean bias. However a self calibrating instrument has to choose a best line of fit for these samples and we have seen on many occasions can actually end up with a bias of 3.50 and a slope of –0.05. What would this mean to your results: If you analysed water the instrument would read 3.5%, if you analysed milk with a true protein of 3.5%, the instrument would read 3.50%. Clearly something is not correct! Now Print out the slope and bias’s of your instrument, are they showing slopes of less than 0.75 or bias greater than 0.75? if yes then you need a solution to your calibration techniques.
The Solution?LABTEK Milk Calibration Standards:
- FAT RANGE 0.07 – 6.00 %
- PROTEIN RANGE 2.00 – 5.00 %
- LACTOSE RANGE 3.00 – 6.00 %