Introduction

Most dairy farms in North America rely on daily bulk pickup weights and monthly DHIA test milk weights to monitor milk production. However, many newer, larger dairies are installing parlors with milk meters and electronic identification. One of the reasons these dairies are purchasing these parlors is due to the recent advances in the quality and volume of available information. A number of other papers at this conference discuss the uses of these systems and attest to their benefits, however, the capital investment required for procuring and maintaining these systems can be substantial and may be out of financial reach on some dairies.

Obviously, there is a need to bridge the large gap between measuring one milk weight a month versus measuring individual cow milk production at every milking. An intermediate alternative is available from in-line milk meters, mounted on the receiver jar discharge pipe. By adding software and minimal milker participation, the total milk yield for each pen can be measured. Additional information can be easily and automatically calculated.

Description

A basic system consists of a sanitary "in-line" turbine meter connected to a "flow computer". The turbine meter is located between the milk pump(s) and the plate cooler on the discharge pipe and measures flow when the milk pump runs. Typically, a temperature probe is also attached to the milk line after the plate cooler. The flow computer is connected to a personal computer via a software program that is essentially always monitoring the process. When the current pen is finished milking, the milker presses a reset button to clear the total. The pen's milk total is recorded along with the start, stop, gaps, and total times, and also the highest temperature.

With the additional information of pen counts from the management software, a number of calculations can be performed: Milk/Hour, Milk/Cow, and Cows/Hour.

Grand totals are calculated for each shift and for each day.

Because the meter is an in-line sanitary meter, it also can measure wash-up. For each phase (rinse, wash, and acid), the start, stop, total and gap times are recorded, along with the minimum and maximum temperatures achieved.

Graphic displays are available for any combination of the herd and/or individual pens, by shift, and even of each wash-up phase. The current temperature and bulk tank volume can be monitored and alarmed if desired, including contacting a pager or cell phone. A one-page printout is produced after every milking, which can be automatically emailed to a third party.

Benefits

The initial goal of these systems was the automatic measurement of production by pen every day, in order to more rapidly detect changes in milk yield in each pen. It was hoped that recent ration changes could be quickly evaluated. Total bulk tank yield is often unsatisfactory for this purpose, especially if the change only occurs in certain pens. Although this milk-per-pen goal has been met, numerous other uses are of interest.

On most dairies, the reports that are printed after each shift are placed on a separate clipboard. This allows the employees to compare the milk yield, and performance of their shift with the others. As a general rule, the milk filter socks are also hung with the report, so that the milk quality message is even more visible than the milk/hour message. Self motivation of employees appears very effective. In addition, graphs of milk production by shift can document progress (or lack thereof).

On the majority of installations, the simple wash-up monitoring described here has detected a problem within the first two weeks of installation. There have been cases of broken hot water heaters, stuck cold-water valves, insufficient hot water capacity, a milker short-cutting the automated process, plate cooler failure, keeping the chiller on during wash-up, and numerous others. Detecting wash-up failures prior to notification from the creamery is an important goal.

For dairies directly pumping into tankers, audible, visible, or digital (cell phone/pager) alarms have been helpful in preventing overflow situations.

However, feeding is still the largest expense, while milk production still pays the bills. When using for detection of feeding errors, an inappropriate feeding change is quickly detected, possibly with a significant financial return. Perhaps even more important is that a positive intervention can be evaluated in a much more timely fashion. The system has the capability to display the milk production for the past two months. The default graph is the overall herd average, but any combination of pens can be selected. Any notes entered will be displayed when the corresponding pen is selected. Examples of notes include BST supplementation, feeding changes, liner, personnel, and severe weather changes.

Problems

A number of problems have arisen with the daily operation of these systems. The frequency varies by farm.

• Originally, these turbine meters were connected to a simple "totalizer", rather than a flow computer. These totalizers showed the current total, but could not be connected to a PC or a temperature probe. Milkers were required to record all the information by hand, then later enter it into a spreadsheet. Precise calibration was not possible (5 percent errors were common), so rather small turbines were used to improve precision. Unfortunately, these tiny turbines spin much faster, especially during wash-up, and needed to be repaired frequently. When a flow computer is used, calibration provides less than one percent error. All data are automatically recorded, along with continuous temperature readings.
• Occasionally, foreign material gets lodged inside the turbine and the flow measurements are inaccurate. Hopefully, installing the turbine past the filters will minimize this.
• Certain farms have significant electrical interference. Proper grounding is important, as is running the 120 VDC power in a separate conduit from the low voltage signal from the turbine to the flow computer.
• Frequently, check valves leak. This allows backflow through the meter, which is indistinguishable from forward flow. The software will print a message regarding continuous flow that should arouse suspicion. Replacing valves when they first start to fail is better than waiting until they stop working completely.
• The software is turned off because the program is exited, or the PC is powered off for more than one milking pen. The system is unable to monitor the parlor when the PC is turned off..
• The temperature probes are connected to the outside of the stainless pipes. If overzealous washing is done to the outside, it is possible to both damage the temperature probe, and/or change the readings. Protecting the probes with insulation has eliminated this problem.
• From time-to-time, milkers forget to press the reset button identifying the end of the pen. The software has an expected yield so it can usually detect this mistake and estimate each yield for the combined pens.

Conclusions

Although daily milk meters with individual cow identification provide significantly more complete information, an inline turbine meter may be an alternative for certain dairies. The information available is helpful in monitoring the cows in each pen, the milkers in each shift, and the functioning of each phase of the wash-up system.

References

Bethard, G., Stokes, S. On-Farm Tools For Monitoring Feeding & Production. Proceedings, Fourth Western Dairy Management Conference, Las Vegas, NV 1999.

Buelow, K. Daily Feed Cost and Production Accounting for Dairies, Compendium on Continuing Education, 19 (2) S56-64, 1997.