Developing Mixer Performance Testing Procedures

This page is part of the Guidance Document Repository (GDR).

Looking for related documents?
Search for related documents in the Guidance Document Repository

General Principles

  • Mixer performance testing assesses the level of one or more substances in a pre-established number of feed samples in a batch of feed.
  • The coefficient of variation (CV) for the batch is calculated to determine whether the batch of feed tested was uniform as an indicator of mixer performance.

The following principles should be considered when designing the mixer performance test used in a particular facility:

1. The mixer performance testing procedure must be defined

This written procedure should specify:

  • mixing equipment tested
  • mixing time
  • batch size

The equipment manufacturer’s recommendations should be used as a starting point to establish on-site guidelines for the operation of mixing equipment. The test batch should be manufactured using the facility’s current feed manufacturing practices (e.g., sequence of ingredient addition).

2. Schedule of tests

Mixer performance must be tested:

  • within a maximum period of 90 days after the installation of a new or replacement mixer
  • after a major repair or modification that could impact on the functioning of the mixer
  • periodically, but at a minimum once every one to three years depending on the risk profile of the facility

3. Testing procedures

Mixing operation settings

Sequence of ingredient addition to the mixer, test ingredient inclusion levels, mixing time and batch size should be those used under normal conditions of operation.

Selection of test substance

The test substance is something than can be measured to evaluate mixer performance.

Choosing a Test Substance

The following criteria should be considered when choosing the test substance:

  • The method to determine the level of the substance should be highly reproducible and have a low variation (e.g., for lab methods, the analytical variability should be less than the target CV for the mixer).
  • Only one ingredient should significantly contribute to the concentration of the test substance in the mix to avoid masking non-uniformity.
  • With respect to selection of the test substance, one or all of the following nutrients would appear to be suitable in most instances:
    • sodium
    • chloride
    • zinc
    • manganese
    • copper

Other test substances can be acceptable provided that they are indicative of product homogeneity and can be tested at an approved laboratory.

Sampling Procedure

  • A minimum of nine spot samples should be taken after mixing is completed.
  • These spot samples should represent the full batch.
  • Samples should be taken at, or as close to the mixer discharge as possible
  • Where this is impossible, another option may be to obtain probe samples from different locations in the mixer.
  • In continuous proportioning systems, samples should be taken with the system running, at even, predetermined time intervals, and as close to the mixer discharge as possible.
  • Take spot samples by holding a small box, cup or scoop in the stream of feed and collecting 9 equally sized samples, (between 100 and 500 grams for dry feeds and between 500 and 1000 grams for total mixed ration (TMR) mixes of feed ). Place each sample in the appropriate sequentially numbered bag or container. Do not make a composite sample.

Three examples of this basic methodology for different types of equipment are described in Appendix I.

Prescribed Critical Limits

The mixer is considered to be producing homogenous feeds when the coefficient of variation for the test batch is:

  • no greater than 5% for dilute drug premixesFootnote 1
  • no greater than 10% for micro or macro premixes and supplements
  • no greater than 15% for complete feeds and total mixed rations

Corrective Actions

Should the initial mixer test not meet the accepted standards, the results should be re-evaluated and original samples re-assayed if necessary. If the original samples are reanalyzed and indicate that the mixing is adequate, these results will be taken as correct. If the results of the original test are confirmed and the CV is above the critical limit, a second test must be performed following the same procedures. Should the second mixer test indicate that the mixing is adequate, these results will be taken as correct.

When the second mixer test verifies that mixing is not adequate, an immediate investigation must be made as to the cause. Continue corrective action and mixer efficiency testing until adequate mixing uniformity is achieved. For suggestions on potential causes of inadequate mixing/poor mixer validation results, refer to Appendix III.

Documentation and Records

The following documents and records are required:

  • Written mixer performance testing procedure
  • Verification records from mixer performance test including:
    • testing date
    • mixing sheet for the test batch
    • mixing time
    • analytical reports for laboratory results or raw data for non-laboratory methods and the calculated CV
  • Equipment maintenance records
  • Record of corrective actions taken when unacceptable CVs are determined

Appendix I - Sample Mixer Performance Testing Procedures

Note: One of the most accepted methodologies for mixer validation is the Test Procedure for Solids - Mixing Equipment for Animal Feeds, approved and adopted by the American Society of Agricultural Engineers, 1990. Standards 5303.2 and S380. In: Standards of the American Society of Agricultural Engineers, St. Joseph, Michigan. Several adaptations to this procedure have been made in order to suit specific circumstances and purposes.

Generic Mixer Validation

Purpose

  • To determine the adequacy of a mixer to distribute ingredients uniformly within a specified mixing time.
  • The assumption is made that all ingredients have properties similar to the tracer in terms of mixability, and will therefore be distributed throughout the feed matrix in the same fashion. In practice this may not always be the case due to varying particle sizes, densities, electrostatic properties, matrix interactions, etc.

Principle

  • A specified level of a tracer is added to a batch of feed. The feed is mixed for the specified time, then a minimum of nine samples are taken during mixer discharge. Each sample is analyzed for the level of tracer. A statistical analysis is performed on the resultant data to determine mixer efficiency.

Quality Control

  • All mixers will be tested at their normal capacity. Any plant mixing quantities less than the mixer capacity must verify adequate mixing efficiency by performing a mixer test at the desired capacity. (Minimum mix is ½ the mixer capacity).
  • The feed matrix chosen should be what is commonly mixed in that mixer. If there are drastic formula differences in the various products mixed (e.g., bulky complete feeds vs. concentrated supplements), mixer efficiency testing may be necessary for the different matrices, particularly if drugs are being added.
  • Mixer validations will be performed every one to three years depending on the risk profile of the facility as well as anytime the mixer has been adjusted, new parts installed (ribbons, screws or paddles), or procedural changes are made. It may be wise to test the mixer at variable mixing times to determine which is the most efficient, particularly after repairs have been made on the mixer which may affect its mixing characteristics. Going well beyond the optimal mixing time may eventually result in a demixing of the feed.
  • The lab doing the analysis should provide the coefficient of variation for the particular methodology used to assist the manufacturer in interpretation of the results. Ideally, the mixer CV should not be much greater (i.e. 5%) than the assay CV.
  • In addition to sampling at the mixer discharge, you may also want to sample the same batch of feed from various locations along the production system,or at least at load-out, in order to verify that the product stays adequately mixed throughout the manufacturing process.

A mixer performance test can be performed using one of the following three basic methodologies. The method chosen will depend on the type of mixer being used.

1. Batch-Type Mixers (except TMR Mixers)

Batch Size: Mixers should be tested at their normal operating capacity.

Type of Feed to be Tested: Any batch in the production schedule can be tested. Note: Ideally the type of feed chosen should be what is most frequently mixed in that mixer, and a feed that usually contains medications.

Number of Samples to be Taken: A minimum of 9 spot samples per batch is required.

Procedure:
  1. Accurately weigh the desired amount of the ingredient containing the test substance to be added to the mixer.
  2. Prepare a minimum of 9 sequentially numbered sample bags or containers.
  3. Calculate the sampling time interval as follows: Measure the time required for the mixer to fully discharge when filled to full capacity. Convert to seconds. For nine samples, divide by 10. Example: A five-minute or 300 second emptying cycle = one sample every 30 seconds starting after 30 seconds for a total of nine samples. An alternative is to sample at evenly spaced target weights during mixer discharge.
  4. Add the feed ingredients to the mixer in their normal sequence.
  5. Mix for the usual length of time, and record the actual mixing time.
  6. After mixing is complete, take at least nine (9) spot samples as close to the mixer discharge as possible.
    1. Take the first sample from the first 10% of the batch as it is discharged and the last sample from the last 10% of the batch as it is discharged.

    2. Take spot samples by holding a small box, cup or scoop in the stream of feed and collecting between 100 and 500 grams of feed. Place each sample in the appropriate sequentially numbered bag or container.

  7. Submit samples to accredited laboratory for analysis for the level of selected test substance or conduct the physical analysis of the test substance in each sample.
  8. Calculate the mean, standard deviation and coefficient of variation of the mixer based on the assay results. (Note: Many labs will do this upon request.)

2. TMR Batch Mixers

Batch Size: Mixers should be tested at their normal operating capacity.

Type of Feed to be Tested: Any batch in the production schedule can be tested. Note: Ideally the type of feed chosen should be what is most frequently mixed in that mixer, and a feed that usually contains medications.

Number of Samples to be Taken: A minimum of 9 spot samples per batch is required. Given the large variation in particle size for TMRs, extra care must be taken to obtain representative spot samples.

Procedure:

  1. Accurately weigh the desired amount of an ingredient containing the test substance to be added to the TMR mixer.
  2. Prepare a minimum of 9 sequentially numbered containers (volume approximately one to two litres).
  3. a) If sampling at mixer discharge: Calculate the sampling time interval by measuring the time required for the mixer to fully discharge when filled to normal operating capacity. Convert to seconds. For nine samples, divide by 10. Example: A five-minute or 300 second emptying cycle = one sample every 30 seconds starting after 30 seconds for a total of nine samples. An alternative is to sample at evenly spaced target weights during mixer discharge.
  4. b) If sampling from the feed bunk: Determine the length of the bunk filled with feed during mixer discharge, when the mixer is at normal operating capacity. If using containers, place the numbered containers at evenly spaced intervals along the empty feed bunk such that both the first and last portion of the mixer contents will be sampled.
  5. Add the feed ingredients to the mixer in their normal sequence.
  6. Mix for the usual length of time, and record the exact time
  7. After mixing is complete:
    1. If sampling at mixer discharge: Obtain at least nine (9) spot samples (sample size = 500 -1000 grams) as close to the mixer discharge as possible by holding a container in the stream of feed until it is full. Take the first sample from the first 10% of the batch as it is discharged and the last sample from the last 10% of the batch as it is discharged.

    2. If sampling from the feed bunk using sequentially numbered containers: Discharge the mixer contents into the bunk containing the sequentially numbered containers. Carefully remove the containers (which should contain at least 500-1000 grams of material) from the bunk and transfer samples to air tight containers.

    3. If obtaining hand grab samples from the feed bunk: Discharge the mixer contents into the bunk. Carefully obtain representative spot samples, through the depth of the feed (500-1000 grams), at evenly spaced intervals along the length of the bunk.

  8. As these samples have a high moisture content, place the entire contents of each container into an airtight bag (numbered as per the container). Submit samples to accredited laboratory for analysis for the level of selected test or conduct the physical analysis of the test substance in each sample.
  9. Calculate the mean, standard deviation and coefficient of variation of the mixer based on the assay results (on a 100% dry matter basis). (Note: Many labs will do this upon request.)

3. Continuous Proportioning Systems (based on Quebec Provincial Regulations)

Type of Feed to be Tested: Any feed in the production schedule can be tested. Note: Ideally the type of feed chosen should be what is most frequently made in that mixing equipment and a feed that usually contains medications.

Number of Samples to be Taken: A minimum of 9 spot samples per batch is required. There can be a large variation in particle size for continuous proportioning systems so extra care must be taken to obtain representative spot samples.

Procedure:
  1. Prepare 9 sequentially numbered containers.
  2. Calculate the sampling time interval as follows: Measure the time required (in seconds) to manufacture a batch of feed (or a tonne of feed if the batch size exceeds one tonne). Divide by 10.
  3. While the system is running, take 9 spot samples, at the time intervals determined in Step 2, as close to the discharge of the mixer as possible. Samples are obtained by holding the container in the stream of feed and collecting between 500 and 1000 grams of feed. Place each sample in the appropriate sequentially numbered sample bag.
  4. Submit samples to accredited laboratory for analysis for the level of selected test substance or conduct the physical analysis of the test substance in each sample.
  5. Calculate the mean, standard deviation and coefficient of variation of the mixing system based on the assay results. (Note: Most labs will do this upon request.)

Appendix II - Mixer Performance Testing Trouble Shooting

  1. Are ribbons/paddles worn and/or deformed?
  2. Is there excessive feed build up on ribbons/paddles?
  3. Have there been any mixer repairs, adjustments or new parts installed (e.g., ribbons, screws, paddles, different power motor) that might affect mixing function?
  4. Have there been any procedural changes (e.g., change in mixing time, charging mixer beyond the manufacturers recommended specifications) that might affect mixing function?
  5. Is the mixer running at the proper revolutions per minute and are ribbons rotating in the correct direction?
  6. Were more than 5-8% of liquid ingredients added to the batches tested? How were liquids added, e.g., sprayed, atomized?
  7. Was there a large variation in the particle size of the ingredients in the batches tested?
  8. Did the ingredient containing the test substance have properties (e.g., electrostatic charge) which render it difficult to mix uniformly?
  9. Was the mixer properly grounded?
  10. Was the ingredient containing the test substance dusty so that it could have been lost via the dust collection system during mixing?
  11. Was the degree of fill appropriate for the batches tested e.g., not above the mixer ribbon or below the mixer shaft?
  12. Was the batch size appropriate, e.g., within manufacturer's recommendations? Too small or too large a batch may contribute to variability.
  13. Was the sample size adequate? Larger samples may be required for TMR and liquid feed mixers or feeds which have higher degrees of particle size variation, e.g., multi-textured rations.
  14. Was the size of the auger used to deliver the ingredient containing the test substance appropriate (e.g., was a premix auger used for macro premixes)?
  15. Were the hammers/rollers and screens in good condition (particle size)?
Date modified: