Planning dairy operation feeding systems for expansion


Source: Ontario Ministry of Agriculture, Food and Rural Affairs

Fact Sheet written by: B. Lang/Dairy Production Systems Program Lead/OMAFRA

When planning an expansion of a dairy farm feed-handling system, develop a strategy that both meets long-term goals and remains manageable during transition. This Factsheet provides an overview of tools and options for dairy producers considering expanding their feeding systems.

Units in this Factsheet are a combination of metric and Imperial. Where Imperial units are common usage, the Imperial unit is listed first, followed by the metric equivalent in brackets.

As with most technology, the tools used in the harvest, storage and delivery of feed on livestock farms tends to come in packages. As farms grow in size, the package of traditional tools such as mower, square baler, hay mow and tie stall manger, which were most appropriate for the 30-cow operation, becomes inadequate.

But the search for new tools is full of pitfalls and challenges. These challenges include accepting the need to change, choosing the right tools and making the new tools work during transition. Develop a strategy for expanding your feed-handling system that both meets your long-term goals and remains manageable during transition.

Trends in Management and Housing

Although no one can accurately predict the future, there are a number of current trends that will likely continue over the next decade or two:

  • Increased herd size. Herd size in Ontario has doubled every 17-18 years. At this rate, the average herd size will be 100 cows in 10 years.
  • Increased need for feed storage and handling. Since production and feed intake will also go up, farms will need double the feed storage and handling capacity 15 years from now.
  • Increased use of freestall housing. Livestock operations tend to switch from tiestall to freestall housing when their herds expand to between 50 and 120 cows. While a freestall barn may not be in your current plans, it may be wise to ensure that any future purchases are adaptable to this management style.
  • Increased per-cow production. Milk production per cow is increasing at about 3% per year. If 8,800 kg of milk per lactation is a typical average today, 12,000 kg will be the norm in 10 years. These cows will demand the best in feed quality, and a delivery system that assures fresh feed, balanced to meet requirements.
  • Changing labour demands. Labour costs – direct rate of pay, benefits, time off and provision of better working conditions – will continue to increase.

These and other factors will shape the future of feed handling on the farm. While emerging technology and economic differences will make the actual outcome unique for each farm, U.S. producers provide a model of which technology will dominate as herds increase in size. Their experience suggests feeding systems for larger herds will have the following characteristics:

  • drive-through feeding
  • total mixed rations (TMR), prepared in trailer- or truck-mounted mixers
  • forage consisting of 50%-70% corn silage and little or no hay
  • ensiled feedstuffs stored in bunker silos
  • high-moisture corn
  • use of by-products
  • use of protein supplements that arrive as commodities

Figure 1: bar graph showing the number of drying periods per week for hay and for haylage 4 out of 5 years in Ontario showing that on any cutting date before the third week of June, the likelihood of getting the 4 consecutive drying days is 1.8 to 2.4 out of 7, or 25 to 34%.

Figure 1. Number of drying periods per week for hay and for haylage 4 out of 5 years in Ontario.

Dry Hay vs. hay silage

While excellent quality hay can be associated with good rumen health and high production, this forage has no place on the large modern dairy farm. As illustrated in Figure 1, in Ontario, on any cutting date before the third week of June, the likelihood of getting the 4 consecutive drying days needed to produce dry hay is low: 1.8-2.4 out of 7, or 25%-34%. In other words, the odds are that two-thirds to three-quarters of the time, the crop will be rain damaged, often severely, if the rain occurs late in the drying period.

Rain-damaged hay is not a matter of bad luck. It is a normal and expected outcome of Ontario weather. By comparison, hay silage coming off in 2 days has a 50%-70% chance of coming in rain free, and at higher moisture, the rain is far less damaging.

Longer harvest days, bigger capacity equipment and fewer labour requirements give further advantage to silage over hay. The list of “solutions” to the problems of hay is lengthy indeed, but most of these solutions have problems of their own. Hay dryers cut field drying but increase labour and cost; dessicants lead to losses if it does rain; and big bales save labour but increase drying times.

As illustrated in Figure 2, hay is subject to the greatest dry matter (DM) losses. Raking and baling causes the loss of leaves. Respiration during drying results in dry matter losses of 2%-6% for hay that dries quickly. Respiration losses are higher for slow-drying hay. Alfalfa, the “queen of the forages” is particularly badly suited to hay-making, as it loses leaves easily when weather damaged.

Figure 2: graph comparing percentage moisture when hay is harvested and the percentage of dry matter loss showing that hay is subject to the greatest dry matter losses.

Figure 2. Estimated hay and haylage harvest and storage losses (adapted from Hoglund, 1964).

Based on this illustration, the forage handling system that maximizes dry matter preservation is 45% dry matter silage.

Hay is also very difficult to include in a total mixed ration (TMR) and when you do, much of its one specific benefit (effective fibre) is lost through particle size reduction. Mixers that can handle hay also make other ingredients finer, often negating the benefits of adding hay in the first place.

Tower Silos

Dry matter losses in well-managed tower silos are low, ranging from 5%-10%. Tower silos require no packing or covering, therefore, involve less labour at filling. Applying additives is simple: tower silos unload in any weather and require less space. When all factors are considered, tower silos appear well suited for herds of 80-100 cows. Once this herd size is exceeded, the disadvantages of tower silos quickly outweigh the advantages.

For storages larger than 165 tons (150 tonnes) dry matter, two 8-ft x 20-ft x 65-ft (2.4-m x 6.1-m x 19.8-m) bunker silos are lower in capital cost than a 20-ft x 70-ft (6.1-m x 21.3-m) upright silo. At this size, bunker silos become a feasible option in terms of feed quality. Note that a 20-ft (6.1-m) diameter silo has twice the surface area exposed to oxygen as the face of an 8 ft x 20 ft (2.4-m x 6.1-m) bunk. Excessive spoilage in the bunker results from improper management and not the size of the face.

The disadvantages of tower silos include potential safety issues with silo gas and the height of the structure, bottlenecks in the production process created by slow filling and the need to refill, and the incompatibility of slow unloading with TMR feeding.

Last but not least, tower silos taller than 70 ft (21.3 m) restrict the moisture content of the forage stored. Excessive leaching forces owners of 80-100-ft (24.4-30.5-m) silos to harvest corn silage at less than the desired 35% dry matter, thereby compromising feed quality. Low-moisture corn silage has lower starch and fibre digestibility. Fibre digestibility has been found to decrease by over 10% as moisture decreases from 70% to 58%. Kernels that are too dry will become hard and can pass through the cow undigested. The use of a processor or finer chopping can help compensate to some extent. When it comes to feed quality, it would appear that 30% dry matter, unprocessed corn silage, in a well-packed bunker is far better feed than 40% dry matter silage in a 100-ft (30.5-m) tower silo.


As herds expand and require additional forage, extra silage, bagged by a custom operator, placed on an existing hard base, may be a very good temporary measure. If the integrity of the bag is maintained, dry matter losses are low, and feed quality is excellent. For longer-term use as a permanent system, bags have a number of disadvantages. Unless they are placed on a well-drained weed-free base, rodent damage is a constant threat. Hail and birds can also be a problem, and fighting the mud adds to labour costs. Bags require a large storage area. When the cost of preparing a large well-drained pad for the bags is included, the system becomes costly for long-term use.

Bunker Silos


The best long-term storage choice for expanded Ontario dairy farms is bunker silos. While very large farms may choose to use piles of silage, bunkers with walls 6-10 ft (1.8-3 m) high will be more appropriate for 100-400 cows. The sidewalls provide greater safety in packing, reduce the size of the face, and reduce both packing and covering labour. While not intended as a complete guide to bunker silo design, the following are features of current interest.

Instead of a single wall between bunkers, the more common design for this “interior” wall is two walls, 4-5 ft (1.2-1.5 m) apart. Since crossbracing increases the strength of this design, less concrete and steel is used in each of the walls and total cost is only slightly higher than a properly engineered single wall. The space between the walls is tile drained to the back of the silo, and filled to 2-3 ft (0.6-0.9 m) of the top with gravel. When the silo is filled and crowned to the top of the wall, the space provides a safe working area for applying plastic and facilitates drainage of rainwater outside the silo. When the plastic is removed, the wall space has room for storing tires. Steps formed into the front panel of the double wall improve access to this area.

Bunker floors should slope 1% toward the opening for proper drainage. Asphalt has been gaining acceptance as a flooring material for bunker silos. Asphalt handles acids and freezing better than concrete. Place asphalt over a well-packed base. Use fabric under the base for stability. Construct a concrete curb around the edge of the apron to prevent damage from traffic entering and exiting the pad. On very hot days, tractor operators must take care not to spin tires, as the asphalt becomes softer in extreme heat.

Until recently, asphalt was considerably cheaper than concrete. In 2002, the University of Wisconsin estimated that asphalt floors in bunker silos were 30%-50% cheaper than concrete. In 2008, the cost of asphalt rose significantly, making the costs of asphalt and concrete floors similar. The price of asphalt declined somewhat in 2009.

Seepage from silage is a very potent pollutant and must be contained to prevent contamination of soil and water. Diverting seepage to a manure storage without also diverting large volumes of rainwater can be accomplished with a “high-flow bypass.” This consists of a sloped apron with a barrier along the lower edge and an outflow pipe located at the lowest point. The outflow pipe is positioned over a second pipe, open on top and located to capture low-flow seepage dripping from the outflow. The higher flow from a rainstorm shoots over the lower pipe and is not captured (Figure 3.)


The appropriate size for a bunker silo will depend on the amount of forage to be stored. Dimensions become a function of size and management objectives.

Figure 3: series of four diagrams showing the flow of bunker silo seepage and options for seepage collection including outside drain, floor drain and filtered vegetated buffer strip.

Figure 3. Bunker silo seepage collection systems.

Typical density of feed in a well-packed bunker will be 15 lb dry matter/ft3 (240 kg dry matter/m3). A minimum 6 in. (15 cm) removal rate per day is recommended to keep feed fresh. Assuming a forage intake of 24 lb (11 kg) dry matter per cow per day, 6 in. (15 cm) removal occurs with a total face area of 3.2 ft2 (0.29 m2) per cow (total of all open bunks on an all-silage diet). This increases to 4.5 ft2 (0.42 m2) if heifers are also fed silage. The recommended length for convenient packing is 150 ft (45.7 m) or less, and since two bunkers, emptied in alternate years, works best, two side-by-side bunkers of 90 ft (27.4 m) each provides 6 in. (15 cm) removal for year-round storage.

Some producers ask about opening both ends of a bunker, but since the floor must slope away from the face, working from alternate ends is not practical.

The minimum recommended width is twice the width of the packing tractor or about 18-20 ft (5.5-6.1 m). Wider bunkers are appropriate if more capacity is needed, but since a 1-to-4 sloped crown is desirable to drain off rainwater, wide bunkers become very high in the centre. Wide bunkers also require additional walking to apply plastic and tires.

Sidewalls should be 8-10 ft (2.4-3.1 m) high. Most loaders can reach about 14 ft (4.3 m). This will be the height of silage in the centre of a 30-ft (9-m) wide bunker with a 1-to-4 sloped crown, and a 10-ft (3.1-m) sidewall. For bigger bunks, higher walls cost more per tonne of feed stored than additional floor area.


Properly addressing this topic is beyond the scope of this paper. Seek detailed information before starting to fill a bunker for the first time. Many disappointments with bunker silos have resulted from inadequate packing. Achieving a density of 15 lb/ft3 (240 kg/m3) requires extensive packing; to measure this, monitor removal volume and weight in winter.

The time required for proper packing is determined by the weight of the packing tractor. Multiply the weight of the packing tractor (in kilograms) by the time spent packing the silage (hours). Then divide the result by the tonnes of silage dry matter stored. A total of 800 hour-kilograms per tonne (1,600 hour-pounds per ton) is required to achieve proper packing.

A forage wagon containing 3.5 tonnes of dry matter takes 12 min (0.2 hr) to pack with a 14,000-kg payloader and 28 min (0.47 hr) with a 6,000-kg tractor. Many custom operators harvest this amount in 8-10 min. To keep up to silage delivery at this speed, use a heavier packing tractor or an additional packing tractor.

The other weak link in bunker management is covering it with 6-mil plastic, split tires and sandbags on the edges. Split tires, or truck tire sidewalls, work better than whole tires because they don’t hold water, are easier to handle and stack readily when not in use. Unload the bunker by cutting a clean face with down pressure on the loader, or with a cutter, and always load and feed all loose material.

Storage of Other Feeds

The ability to handle truckload lots of wet and dry by-product feeds is an important economic advantage for larger herds. Where wet products are to be fed, a special bunker 14-16 ft (4.3-4.9 m) wide with 4-5 ft (1.2-1.5 m) side walls big enough to handle a trailer load of wet brewers or other by-products is an asset. Commodity sheds with four or more bays are well adapted for unloading trucks and loading trailer mixers, especially for non-flowing materials such as cottonseed. For more costly, finer ingredients, such as dry protein supplements, there is a strong trend back to enclosed storage bins, emptied by auger.

Studies have shown that when these ingredients were stored in sheds and added to the mixer with a loader, mixing errors and losses to wind, birds and spoilage were very high. Both lower losses and more accurate inclusion in the TMR more than pay for the increased handling time and labour of bulk bins for these feeds.

The use of bunker silos for high-moisture corn is also gaining widespread acceptance. For clean grain or cob corn ground or rolled in at 30%-35% moisture, removal rates of 4-6 in. (10-15 cm) per day are sufficient to prevent spoilage. With a dry matter density of 45 lb/ft3 (720 kg/m3) and a feeding rate of 25 lb (11.4 kg) per cow at 30% moisture, you can support 6 in. (15 cm) removal from a 16 x 6 ft (4.9 x 1.8 m) face with 125 cows. For smaller herds, use agbags or tower silos for high moisture corn.

Costs and losses

A 1998 study from Wisconsin comparing the cost of silage systems is summarized in Table 1. Costs are likely to be higher today.

Table 2 summarizes estimated losses from silage systems, which can range from 6% to more than 50% during filling, storage and feed out. Correct moisture at ensilage, restricted access to the atmosphere and proper base materials significantly reduce losses.

While these U.S. studies are more than 10 years old, they are still relevant for their relative comparison. One clear message from this: when handling costs and dry matter losses are included, silage is by no means cheap feed.

Table 1. Costs of stored dry matter
Silage System
$/ton of Dry Matter
Total Capital Cost (annual cost)
384 tons DM
768 tons DM
Steel-glass oxygen limiting (new)
427(82) 301(60)
Steel-glass oxygen limiting (used)
268(55) 187(41)
Cast-in-place oxygen limiting
285(58) 186(41)
Concrete stave
192(46) 138(36)
Above ground bunker
152(45) 103(37)
Packed silage pile
63(37) 41(32)
88(38) 53(32)
Wrapped bales
64(36) 38(32)

Adapted from Holmes 1998.

Table 2. Estimate of silage losses during filling, storage and feed out
Silo Type Moisture (%) Filling Seepage Gaseous Top
Feedout Total
DM Loss (%)
Conventional tower
801 1-2 7* 9* 3* 1-5 21-26
701 1-2 1* 8* 4* 1-5 15-20
65 1-3 0* 8* 3* 1-5 13-19
60 1-3> 0* 6* 3* 1-5 11-17
50 2-4 0* 5 3* 1-5 11-17
Gas-tight tower
701 0-1 1* 7* 0* 0-3 8-12
60 1-2 0* 5* 0* 0-3 6-11
50 2-3 0* 4* 0* 0-3 6-12
40 2-4 0* 4* 0* 0-3 6-13
Trench or bunker, no cover2
801 2-5 6* 10* 6* 3-102 27-37
701 2-5 1* 9* 9* 3-102 24-34
60 3-6 0* 10* 12 5-152 30-43
Trench or bunker, covered2
801 2-5 4* 9* 2* 3-102 20-30
701 2-5 1* 7* 3* 3-102 16-23
60 3-6 0* 6 4 5-152 18-31
Stack, no cover2
801 3-6 7* 10* 11* 3-102 34-44
701 3-6 1* 11* 19* 3-102 37-47
60 4-7 0* 12 24 5-152 45-58
Stack, covered2
801 3-6 5* 8* 2* 3-102 21-31
701 3-6 0* 7* 4* 3-102 17-27
60 4-7 0 6 6 5-152 21-34
Silage bags
801 1-2 2 6 2 1-5 12-17
60-701 1-2 0 5 2 1-5 9-14
Wrapped sileage
60-701 1-2 0 8 5 1-5 15-20
50-601 2-3 0 6 6 1-5 15-20

1Avoid ensiling hay crop above 70% moisture in structures and above 60% moisture in wrapped bales to prevent clostridial fermentation.

2Feed out loss is 3%-5% with good management on concrete floor. Use 4%-6% for asphalt, 6-8% for macadam and 8%-20% with earth floor, assuming good face management. With less than good management, add up to 7% additional loss.

*Based on Forages: The Science of Grassland Agriculture,4th ed. See Bickert et al. (1997).

Adapted from Holmes 2000.

Location of Feed Storage

With the use of trailer mixers, locating feed storage some distance from livestock facilities is no longer inconvenient. It is advantageous to keep all feed storage close together to avoid having to move both the trailer mixer and loader to another location. Locate bunker silos, commodity sheds, agbags and bins around a single paved apron to permit loading the mixer without moving it. Provide at least 50 ft (15 m) of hardened surface in front of commodity storage to accommodate delivery by tractor-trailer.

Alley and Manger Design

Use of mobile TMR technology will require a drive-through alley with 12-14 ft (3.7-4.3 m) of clear height and enough width to accommodate a trailer mixer. Assuming 3 ft (0.9 m) for the manger and 10 ft (3 m) for equipment, an 18-20 ft (5.5-6.1 m) drive-through alley is recommended – 16 ft (4.9 m) if feeding on one side only. While alleys 16 ft (4.9 m) wide (for feeding on two sides) or 14 ft (4.3 m) wide (for feeding on one side) are common, they leave no room for pedestrians.

Until recently, preferred barn layouts in four- and six-row barns have included drive-through alleys open at both ends and a “cross-over” for cow traffic. If using this layout, be sure to recess and slope the cross-over so manure liquids drain to the cow alleys. Use a non-slip concrete finish appropriate for cow traffic.

Biosecurity is important, and this crossover is a recognized risk factor, if the mixer is driven through manure and along the manger. Alternative designs include hinged plywood covers that form sidewalls for the cross-over, dead-end drive alleys, and three-row barns and other layouts with all cows on one side of the drive-through. In Europe, some barns using robotic milking have drive alleys on both sides, on the outside of the building.

Figure 4 shows a recommended design for mangers. Set the eating platform 4-6 in. (10-5 cm) above where the cow stands. The height of the wall and toprail are critical and must be decreased for small breeds or heifers younger than 18 months. Self-locking headgates are an excellent handling system but there is preliminary data indicating they may reduce feed intake by up to 5%. In barns using the manger design illustrated, it is customary to push up feed several times per day. A few Ontario herds are experimenting with specially designed alley scrapers, which automate this chore. One company has introduced a robotic feed pusher.

Figure 4: diagram of a post-and-rail feed bunk showing the height of the manger wall and size of supporting posts.

Figure 4. Dimensions for a post-and-rail feed bunk.


Development of a strategy and long-term goals for feed-handling systems will help ensure that these systems continue to meet the needs of the herd as it expands.

A Case Study

Perhaps the biggest challenge in planning a feeding system is dealing with the transition years. This outline shows how a tie stall operator with 50 cows and two tower silos might expand his feed handling as the herd grows.

  • Present facilities for heifers are lacking. Add a barn for older heifers and dry cows. Although current feed will be big round bales, ensure the manger and drive-through can accommodate TMR.
  • Nutrition management needs improving. Purchase a TMR mixer. Though it will be housed in a feed room and used stationary for now, it is trailer mounted and tractor powered to suit future needs.
  • The big bales get rained on and must be replaced by silage. Choose a future site for bunker silos and commodities. Strip one end of topsoil, levelled to 1% slope, and raise it with packed gravel 120 ft (36.5 m) long. This is the base for silage bags now and the future bunkers. Adjust cropping to more corn and bring in a custom bagger to fill a large bag, to supplement the tower silos (one each with haylage and corn silage).
  • Low priced by-product feed is available. Add brewers grain, delivered in a bag, on the gravel pad.
  • The herd grows some more and one bag has become two. Feed intake is now enough to justify bunkers for corn silage (freeing up the other tower for haylage, a real blessing in terms of feed quality, loading and unloading). A permanent 90-ft (27.4-m) double wall forms one side of the bunkers; pave the pad to form a floor. Purchase two moveable, pre-cast, free-standing double bunker walls and place them on top of the pad, to form two bunkers 8 x 20 x 90 ft (2.4 x 6.1 x 27.4 m).
  • The herd grows again. Widen the paved area and move the moveable walls. Expand the gravel pad to accommodate bagged high-moisture corn.
  • Unloading the deteriorating towers with worn-out unloaders is becoming a chore.Move the moveable walls again, replacing them with permanent 30 ft (9.1 m) wide bunker walls for corn silage. Add the new haylage bunkers using the moveable walls to the ever-widening pad.
  • There are no more trips to the old feed room. Move supplement bins to the pad and add a commodity shed with small ingredient storage and feed office.
  • A freestall barn, with drive-through feeding, now houses the herd. Its location and that of future livestock housing is far more flexible than before, since TMR fed out of a feed centre can be moved anywhere.


Holmes, B.J. 1998. Sizing and managing silage storage to maximize profitability. Four State Forage Feeding and Management Conference Proc.

Holmes, B.J., and R.E. Muck. 2000. Preventing Silage Storage Losses, University of Wisconsin Extension,

Hoglund, C.R. 1964. Comparative storage losses and feeding value of alfalfa and corn silage crops when harvested at different moisture levels and stored in gas-tight and conventional tower silos: an appraisal of research results. East Lansing: Michigan State Univ.; Agric. Econ. Publ. 947.

Clarke, S., and R. Stone. 2004. How To Handle Seepage From Farm Silos, OMAFRA Factsheet Order No. 04-031. Ontario Ministry of Agriculture, Food and Rural Affairs.

This Factsheet was written by Brian Lang, Dairy Cattle Production Systems Specialist, OMAFRA, Woodstock, based on an article by Jack Rodenburg, Dairy Systems Production Program Lead, retired, OMAFRA.