Skip to Main navigation Skip to Left navigation Skip to Main content Skip to Footer

University of Minnesota Extension

Extension is almost done building a new website! Please take a sneak peek or read about our redesign process.

Extension > Agriculture > Livestock > Swine Extension > Quality control of on-farm swine feed manufacturing

Print Icon Email Icon Share Icon

Quality control of on-farm swine feed manufacturing

Lee J. Johnston and Jerry D. Hawton

All swine producers are faced with numerous questions concerning their feed supply. One of these questions is, "Should I buy commercially prepared feed, or should I purchase ingredients and manufacture feed on the farm?"

In an effort to reduce feed costs, many swine producers choose to manufacture all or a large part of the feed for their animals. When they do this, they assume responsibility for quality control. Unfortunately, quality control aspects of an on-farm feed manufacturing program quite often receive little attention. Researchers at Michigan State University found that 33% of feed samples deviated more than 1% from desired crude protein levels (1). A recent study conducted at Ohio State University revealed that only 20 percent of the sow diets sampled met or exceeded the sow's nutrient requirements for crude protein, calcium and phosphorus (2). Thirty-four percent of the diets were not mixed or sampled properly.

Errors in manufacturing feed have an economic impact on swine producers. Workers at Purdue University (3) have calculated the cost of producing market hogs when protein content of diets vary from 11 percent to 19 percent. When protein content rises above 15 percent, feed costs increase with no improvement in pig performance. On the other hand, low-protein diets result in depressed average daily gain (ADG), poor feed efficiency (F/G) and higher cost of production. These findings should alert producers that manufacturing high quality feed is much more than adding "a pinch of this and a dash of that."

Use high quality ingredients


The first step in manufacturing high quality feeds is to obtain high quality ingredients. It is virtually impossible to manufacture high quality feed with poor quality ingredients. Grain should be free of molds, insects, dirt, stones and other miscellaneous debris when it is stored. Avoid a high proportion of broken kernels because broken kernels are more likely to support mold growth than intact kernels. Grains should contain about 12-15 percent moisture for safe storage. Avoid excessive temperatures (above 300 degrees F) when artificially drying grains. Overheating compromises protein quality.

Producers must not let grain quality deteriorate while in storage. Rodent damage and mold contamination are two major threats to stored grain quality. Rodents will eat grain and contaminate grain with their excrement. Rodent droppings will reduce palatability of grains and ultimately reduce feed consumption by pigs.

Mold growth is enhanced when grain is stored with high moisture content. Molds produce toxins called mycotoxins which depress pig performance. The mold, Aspergillus flavus, produces aflatoxin (4). Aflatoxin is a carcinogenic compound that causes anemia and stunted growth in pigs when present at 200 parts per billion (ppb). FDA guidelines allow no more than 20 ppb aflatoxin in animal feeds. Aflatoxin contamination of grains is not a common problem in the North Central U.S.

Zearalenone and vomitoxin (deoxynivalenol-DON) are produced by Fusarium molds and commonly are found in grains, primarily corn. Fusarium mold can be identified by the white to pink discoloration of grain. Zearalenone has estrogenic properties that will cause swelling of the vulvas and teats of immature gilts when present at levels greater than 1 ppm. Vomitoxin causes feed refusal when present at .5 ppm. Vomiting will occur at much higher concentrations of vomitoxin.

Presence of mold does not always mean mycotoxins are present. Commercial laboratories can analyze representative samples for presence and concentration of mycotoxins. These tests cost between $25 and $50. Several companies have developed on-farm enzyme-linked immunosorbent assay (ELISA) tests for various mycotoxins.

Producers can use young gilts to conduct a preliminary test for the presence of Fusarium mycotoxins. Feed the suspect grain to young gilts (100-125 pounds) and observe feeding behavior and anatomical changes. Feed refusal or very low feed intake suggest the presence of vomitoxin. Near normal feed intake but swelling of vulvas and mammary systems after two or three days suggests zearalenone contamination. This screening only determines presence or absence of mycotoxins. Confirmation and quantification of mycotoxin levels must be determined by lab or ELISA analysis.

Protein sources

Protein supplements must be stored under the same conditions as those described for grains. Freedom from contaminants is necessary if the goal is to produce high quality feeds. Most soybean meal, the primary protein source in the North Central U.S., is produced by companies that adhere to National Soybean Processor Guidelines and meets established quality standards.

Producing a high quality protein source is more of a challenge when soybeans are processed on the farm. Producers must subject soybeans to the proper time and temperature to destroy antinutritional factors in soybeans without reducing protein quality. Roasted soybeans should be heated for three to five minutes at 240 degrees to 260 degrees F, while extruded soybeans should have an exit temperature of 280 degrees F. Underprocessing will not destroy all of the trypsin inhibitors. Urease enzyme is also destroyed by proper heating and can be used as an indicator of proper processing. To test for underheating, combine 1 part urea, 5 parts water and 10 parts soybean meal, extruded soybeans or ground, roasted soybeans in a small plastic bag. Close the bag and wait 30 seconds. If soybeans are undercooked, the urease in the soybeans will convert urea to ammonia, which can be detected by placing your nose close to the open plastic bag. Properly processed soybeans will not produce ammonia.

Excessive heating of soybeans or soybean meal will denature proteins in the soybean products. Denaturation of proteins reduces availability of essential amino acids, particularly lysine, to the pig. Over-heated soybean meal or soybeans have a color ranging from tan to dark brown, depending on the severity of overheating.


Vitamin and trace mineral (VTM) premixes should be purchased from commercial companies. Swine producers do not have the proper equipment to produce high quality premixes. Furthermore, trace minerals easily destroy potency of vitamins when not mixed properly. Commercial manufacturers possess proper equipment to blend vitamins and trace minerals together and the technology to protect potency of vitamins. Vitamin-trace mineral premixes should be used within 90 days of purchase. Potency of premixes decreases with long-term storage. Store premixes in a cool, dry and dark place.

Proper manufacturing procedures


Accurate formulation is essential to producing swine diets that satisfy nutrient requirements of the pig. Nutrient concentrations of feed ingredients can vary substantially from average values published in nutrient composition tables. For example, low-yielding, drought-stressed corn usually has a higher protein concentration than that published in nutrient composition tables. In contrast, high-yielding corn may have a lower protein concentration when compared with "average" values in published tables. The most accurate formulations result when laboratory analysis of ingredients is available. Swine producers should seek the help of trained professionals if they are not comfortable with the calculations involved in the formulation process.

Recipes from feed tags or special formulations should be followed exactly! Deviation from these recipes will alter nutrient content of the final feed and may compromise pig performance. Use supplements and premixes that have been formulated specifically for swine. Do not use a trace mineral premix designed for beef cows in a sow diet. Do not mix premixes and supplements from different suppliers. For example, do not use company A supplement with company B booster pack. Products from different companies are not made to go together in a diet.

Particle size

To achieve optimal pig performance, it is necessary to process cereal grains through a hammermill or roller mill to reduce particle size. Reduction of particle size increases surface area of the grain that can be attacked by digestive enzymes. Increased surface area for enzyme action improves efficiency of digestion and ultimately efficiency of body weight gain. Furthermore, particle size reduction allows uniform mixing of grain with protein, vitamin and mineral supplements.

As particle size gets smaller, feed efficiency improves and incidence of gastric ulcers increases. The incidence of gastric ulcers determines the lower limit of particle size. Researchers in Kansas reported that incidence of gastric ulcers increased from zero to 50 percent of examined finishing pigs when particle size was smaller than 700 to 800 microns (5). Very small particle size also increases problems with feed bridging in bins and feeders.

Many factors associated with the grain, such as kernel size, moisture content and fiber content, affect resulting particle size when grinding or rolling grain. Additionally, mechanical factors such as hammer speed, hammer shape, screen wear, screen size and roller speed influence particle size. A maintenance schedule should be developed and followed for hammermills and roller mills. Properly maintained and operated hammermills or roller mills can produce the desired particle size of 700 to 800 microns. Periodic particle size analysis of feed will allow producers to determine if hammermills or roller mills are working properly.

Adding ingredients

There are two types of feed mills — continuous flow (sometimes called mix-mills, volumetric or meter mills) or batch processing. Continuous flow mills add ingredients based on volume. This procedure assumes each ingredient has a constant bulk density. When bulk density of ingredients changes, the mill still adds a given volume of material, but that volume no longer contains the proper amount (pounds) of ingredient. Therefore, calibration of continuous flow mills should be checked, and adjusted if necessary, for all diets at least once a month and every time a new batch of ingredient is purchased. Consult the Pork Industry Handbook publication "Calibrating Meter-Type Feed Mills" for more details about this process (6).

With batch processing mixers, each ingredient is added individually. On many farms, this system consists of a portable grinder-mixer. Ingredients should be added by weight, not volume. Greater accuracy in feed manufacturing is possible when ingredients are added by weight. A portable grinder-mixer can be equipped with a scale for about $1,600 to $1,800. Using the data in Table 1, feed for finishing pigs that is 2 percent higher or 2 percent lower in crude protein than the desired 15 percent will add between $1.72 and $1.97 to the total cost of producing a market hog. Assuming that use of the scale will increase the accuracy of feed manufacturing, producers could pay for the scales with cost savings from producing 875 to 1000 pigs. The only substitute for a scale is another scale.

Make a list of ingredient names and amounts used in each swine diet, then check off each ingredient as it is added to the mixer. This will help guard against forgetting an ingredient or adding one ingredient twice. Ingredients added in very small amounts (less than one to two percent of diet) are difficult to mix evenly throughout a batch of feed. These ingredients should be premixed with the cereal grain and added in amounts of not less than 40 pounds per ton. A clean cement mixer works well for premixing.

The order that ingredients are added to the mixer will influence mixing time required to disperse all ingredients evenly throughout the feed. Add at least one-half of the grain first, followed by all premixed ingredients. Next add all the protein source and finish by adding the remaining grain. A good rule of thumb is to mix feed 15 minutes from the time the last ingredient is added. However, it is best to follow the manufacturer's recommendations on mixing time for the size and style of your mixer. Particle segregation or "unmixing" of feed will not occur in this time period if particle size of ingredients is uniform. Do not overfill mixers. Efficiency of mixing is reduced when mixers are too full.

Sequencing of batches

Every effort should be made to guard against drug carryover from medicated to non-medicated feeds. Drug carryover occurs because electrostatic charges cause medicated feed particles to cling to the inside of the mixer. Another source of drug carryover is residual medicated feed that remains in discharge augers and the bottom of the mixer after unloading. Proper sequencing of batches will minimize unwanted drug carryover from medicated to non-medicated feed. Mix medicated feeds first, followed by non-medicated sow feed, which will flush residual medicated feed out of the mixer. Non-medicated finishing feed should be mixed last. If proper sequencing is not possible, the mixer should be flushed with ground corn and cleaned thoroughly by hand to remove drug residues.

Quality control programs

Every quality control program should include periodic laboratory analysis of ingredients and feed. Success of a quality control program hinges on collection of representative samples for laboratory analysis. Great care should be taken to ensure samples are representative of material under inspection so that lab results will reflect the nutrient content of the ingredient or feed being sampled. The easiest way to obtain representative samples is to collect numerous samples from a running stream of material, mix the samples and subsample for lab analysis. When a running stream is not available, bulk lots should be sampled with a sampling probe. Six to eight probe samples should be collected around the outside edge of the structure and two or three in the center of the structure for a total of eight to ten samples that are then mixed and sub-sampled. For bagged material, collect a one-pound sample with a probe from 10 to 15 percent of the bags in the shipment. Mix samples and subsample for lab analysis. Retain a portion of the subsample in a freezer for possible later analysis.

Each new batch of grain should be sampled for lab analysis because grains tend to be quite variable in nutrient content. Diets can then be formulated based on nutrient content of grain. Soybean meal does not have to be sampled quite so often because processors are bound by law to meet guaranteed analysis on feed tags. It is still a good idea to sample soybean meal occasionally. If protein content does not meet tag guarantees, then the producer is eligible for a rebate. Soybean meal should also be analyzed for calcium content. Sometimes a calcium source is added to soybean meal to enhance flowability, resulting in calcium levels higher than normally found in soybean meal.

If premixes are purchased from a reputable company, it is not necessary to routinely send samples for laboratory analysis. Lab analyses for vitamins and trace minerals are rather expensive. Reputable companies pay close attention to the quality control aspects of manufacturing premixes. Nonetheless, it is a good idea to sample each shipment of premixes and keep samples stored in a freezer. If problems with the premix develop, a sample is available for lab analysis.

At the very minimum, final feed should be sampled and analyzed quarterly. Preferably, lab analysis of finished feed should be conducted every two months. Frequent analysis of final feed will help producers determine the quality of feed that is offered to pigs. Lab analysis is the first step in troubleshooting feed manufacturing problems.

Quite often, feed suppliers use synthetic amino acids, particularly lysine, to reduce the proportion of soybean meal needed in a complete diet. In this situation, crude protein concentration of the final diet may be lower than expected. Consequently, a more accurate evaluation of the nutritional value of a diet is possible by analyzing for lysine instead of crude protein. Lysine analysis is more expensive than crude protein analysis.

Interpreting results of a lab analysis can be confusing. The first principle to remember is that lab results are not carved in stone. Labs do make mistakes from time to time and there is normal variation in lab analyses. Acceptable variation in laboratory analyses is presented in Table 1. If lab analysis of a final feed indicates a particular nutrient concentration is outside the range of normal variation, then another sample should be submitted for analysis. Major changes in feeding programs should not be made based on results from one feed sample. If repeated analyses of the same feed indicate nutrient concentrations are outside the acceptable range, then it is likely that an error in formulation, mixing or sampling occurred. Producers may need to consult a nutritionist to help in troubleshooting their feed manufacturing process.

Table 1. Example of acceptable variation in laboratory analyses (adapted from Reese and Brumm)

Nutrient Variation Calculated
Crude protein 3% 16% 15.5 to 16.5%
Lysine 5% .70% .66 to .74%
Calcium 5% .65% .62 to .68%
Phosphorus 5% .50% .47 to .52%
Trace minerals
(zinc or iron)
5% 100 ppm 95–105 ppm

Several commercial labs are available for analysis of feed samples. Often, suppliers of feed ingredients will offer lab services to their customers. Contact your county agent, state extension specialist, or feed dealer for names and addresses of commercial labs. It is best to call a laboratory before submitting a feed sample to inquire about cost, sample size needed, type of analyses available and turn-around time.

Consumers demand residue-free pork. It is the pork producer's responsibility to supply safe, wholesome pork to the marketplace. A sound quality control program will help producers meet this goal. The National Pork Producer's Council has developed an educational program entitled "Pork Quality Assurance" that instructs swine producers in methods of producing wholesome pork. Producers should enroll in this program.

Feed quality after the feedmill

A primary goal of on-farm feed processing is to deliver high quality feed to pigs where they are housed. Therefore, attention to producing high quality feed does not end when feed leaves the mixer. It seems foolish to work hard to produce top quality feed, only to have it soiled or spoiled in the time between mixing and consumption by the pig. To maintain quality, store feed in a clean, dry place. Protect feed from insect and rodent infestation. Bulk storage bins frequently should be emptied completely so that stale feed does not remain in the bin. Be sure to clean the auger boot when bins are empty. Quite often, wet, moldy feed accumulates around the auger intake.

Feeder management also plays an important role in delivering high quality feed to pigs. Feed gets stale if allowed to sit in feeders for extended periods of time. Feed intake of pigs may be reduced by stale feed. Feeders should be filled frequently enough to provide fresh feed for pigs at all times. Consequently, enough feed for one week should be added to grow-finish feeders, only one to two days for nursery pigs and one day for lactating sows. Keep feeder holes free of wet, soiled feed and remove feed buildup in corners. These conditions lead to mold growth.


Producers who choose to manufacture feed on the farm also accept responsibility for maintaining feed quality. Quality control begins with purchase of feed ingredients, continues through the feed manufacturing process and does not end until pigs have consumed the feed. With feed costs representing 60 to 70 percent of the total cost of producing a market hog, it seems logical that producers pay particular attention to quality control aspects of on-farm feed manufacturing. A sound quality control program will help swine producers get the most "bang for their bucks" spent on feed.


  1. Parsons, M. J., T. J. Gall and M. G. Hogberg. 1982. Analysis of Michigan corn and feed samples. Michigan State University swine research report no. 437, p. 63.
  2. Shurson, G. C., T. Nye and D. B. Gerber. 1988. Summary of results from the 1986-87 Ohio feed analysis survey of gestation and lactation rations. Ohio swine research and industry report, p. 20.
  3. Bache, D. H., T. Cline and K. Jackson. 1985. Economics of good feed mixing. Purdue University, West Lafayette, Ind.
  4. Christensen, C. M., C. J. Mirocha and R. A. Meronuck. 1988. Molds and mycotoxins in feeds. University of Minnesota extension folder FO-3538.
  5. Wu, J. F. 1984. Effects of particle size of corn, sorghum grain, and wheat on pig performance and nutrient digestibility. Ph.D. diss., Kansas State University, Manhattan, Kans.
  6. Purkhiser, E. D. and P. D. Bloome. 1984. Calibrating meter-type feed mills. Pork industry handbook,factsheet 94.
  7. Reese, D. E. and M. C. Brumm. 1988. Mixing quality swine feed. University of Nebraska Lincoln extension bulletin G88-892.
  • © Regents of the University of Minnesota. All rights reserved.
  • The University of Minnesota is an equal opportunity educator and employer. Privacy