Red Chile Pod Reclaimer Evaluations

New Mexico Chile Task Force: Report 27
George Abernathy, Ed Hughs and F. Ed Eaton
College of Agricultural, Consumer and Environmental Sciences, New Mexico State University

Authors: Respectively, New Mexico Chile Task Force consulting engineer and professor emeritus, Department of Civil Engineering, New Mexico State University (; research leader, USDA, ARS Southwestern Cotton Ginning Laboratory (; and engineer, New Mexico Chile Task Force and NMSU Extension Plant Sciences Department, New Mexico State University


Red chile harvest mechanization is becoming more widespread in New Mexico, Arizona and Texas in response to farm labor shortages and costs, and global competition. Existing harvesting machines, however, exhibit some shortcomings, one of which is that they occasionally pick whole plants or branches with attached chile pods. This material must be discarded or processed by hand to reclaim the pods. Either way, attached pods represent an additional mechanical harvesting expense. This creates the need for an efficient device to detach the marketable pods and discard the undesirable plant material.

The attached pods may be intermingled with loose pods on a conveyer belt coming from the harvesting machine's picking head, or they may have been separated from the loose pods by a cleaning device so that the attached pods are mixed only with sticks and other "trash" material. Ideally, the detaching or "reclaiming" process should occur on the harvesting machine, prior to delivery to the processor so that non-marketable material can be recycled in the field rather than hauled away from the processing plant for disposal.

The Boese harvester is the only commercial chile harvesting machine with a reclaiming device. The Boese reclaimer has three spiked drums that pick up sticks and plants from a conveyor belt coming from the picking head. Free pods pass underneath the drums, and the processed material, consisting of pods and sticks, is placed directly back onto the conveyor belt. The reclaimer is proprietary, and the company has not published any performance evaluation data. Field observations indicate that the mechanism breaks plants into sticks but does not remove all marketable pods from those sticks.

This paper reports trials evaluating some devices that may be suitable for detaching and separating branches from pods. Spring- and solid-toothed cylinder devices that pass material over grates or through stationary fingers were tested. The most practical detachment device was a solid-fingered pinwheel with one row of stationary spring fingers, although it would require a sorting mechanism downstream. The reclaimers were designed to be small enough to be incorporated into existing mechanical harvesting equipment.

Test results, Fall 2004 and Spring 2005

The first prototype reclaimer tested was a rotary drum with spring fingers spaced 3 inches apart (figs. 1 and 2). There were six rows of fingers on a 24-inch drum. The drum pulled the material over a concave surface equipped with three rows of spring fingers that bisect the spaces between the drum fingers. The concave fingers could be tilted in the direction of material flow to decrease the device's aggressiveness.

Illustration of cylindrical, spring-fingered reclaimer

Figure 1. Illustration of cylindrical, spring-fingered reclaimer

Photo of cylindrical reclaimer showing drum with spring fingers

Figure 2. Cylindrical reclaimer showing drum with spring fingers

The second device was a test bed for a fixed-finger reclaimer (fig. 3). A trough was assembled on a table top. It was equipped with three rows of fingers that could be placed 3 or 4 inches apart laterally. A Plexiglas® slide was fitted with two rows of fingers spaced between the trough pins. The pins were threaded 3/8-inch steel rods or bolts. Except for clearance, pins extended completely through the 3-inch space between slide and trough.

Photo of fixed-finger reclaimer test device

Figure 3. Fixed-finger reclaimer test device

The two devices tested would have different applications for reclaiming chile. The cylindrical device would be applicable to systems in which whole plants and branches with attached pods had been separated from other harvested material so that they constituted the entire stream of material being fed into the reclaimer. It would be necessary to later return the sticks and reclaimed pods to the harvested product stream for further separation and cleaning. The fixed-pin reclaimer should allow a pinned rotor above the product conveyor to pick up whole plants and branches with attached pods, while allowing free pods to pass underneath the rotor. An inverted concave with fixed pins should allow reclaimed pods to drop back onto the conveyor while rejecting sticks from the product stream.

The first tests were conducted using late-season 'Sonora' plants harvested at the NMSU Leyendecker Plant Sciences Research Center (LPSRC) at the end of the 2004 crop year. The plants were extremely dry with many of the pods discolored and/or shriveled. However, a reasonable number of plants were selected to test both devices. Another set of tests was conducted when a limited number of greenhouse plants became available. Tests are discussed in the following two sections of this report.

Spring-fingered cylindrical reclaimer

A test run consisted of four to six plants being fed through the spring-fingered cylindrical machine. The cylinder reclaimer was operated at 22 rpm for all tests on the dry 'Sonora' plants. The concave teeth were fully inserted and placed at 90 degrees for the most aggressive reclaiming. Runs also were made with the teeth one-half inserted at 90 degrees. For less aggressive reclaiming the teeth were tilted about 30 degrees in the direction of rotation for both insertions. Cylindrical reclaimer test run results are shown graphically in fig. 4. Each point is the average of two tests.

In fig. 4, the three rows of bars on the left represent the distribution of free pods, attached pods and broken pods. Percentages are based on total pods. The most aggressive treatment (full insertion, 90 degrees) resulted in about 75 percent free pods and 25 percent damaged pods, while removing all pods from plants and branches. The two medium-aggressive settings resulted in more free pods and reduced pod damage. However, a few pods (<10 percent) were left on plants and branches. The least aggressive setting resulted in an unacceptably high number of attached pods although it reduced pod damage to zero.

The second three-row set of bars in fig. 4 shows the effect on sticks. Data are expressed as percentage of total sticks. Aggressive reclaiming resulted in no attached sticks but a high percentage of long sticks. Aggressive treatment might be expected to break up long sticks, which are easier to sort out than short sticks. Medium-aggressive reclaiming showed an increase in attached sticks but no clear trend in the division between long and short sticks. The least aggressive setting had a large portion of attached sticks but no long sticks.

The set of bars on the right-hand side of fig. 4 shows small trash as a percentage of total sample weight. As expected, the amount of small trash increased with aggressiveness.

Bar graph of distribution of reclaimed material from the cylindrical separator using four concave-tooth positions

Figure 4. Distribution of reclaimed material from the cylindrical separator using four concave-tooth positions

In the spring of 2005, about 50 greenhouse plants became available. They were small-stature plants, approximately 18 inches tall with four to six medium-sized red pods and a few green pods on each plant. The plants were mostly vigorous with a full canopy of green leaves.

Due to the limited number of plants, each sample consisted of two plants. After a duplicate set of runs similar to those performed on the late season 'Sonora', it was obvious that the variability of result made conclusions difficult. Therefore, an additional set of runs was made on the spring-toothed cylinder reclaimer. The combined results are shown in fig. 5.

Bar graph of average of four trial runs of the spring-toothed cylinder reclaimer on greenhouse plants

Figure 5. Average of four trial runs of the spring-toothed cylinder reclaimer on greenhouse plants

The greenhouse plants had lost some moisture by the time they were used so the pods were easily damaged. The number of free pods was disappointing for all treatments. The best result was for the full insertion, 30-degree setting, which was the next-to-least aggressive setup. The percent of attached pods increased as aggressiveness decreased, as expected. Broken and damaged pods decreased with decreased aggressiveness, also as expected. Stick results were mixed with lots of short sticks for the most aggressive reclaiming and a high number of attached sticks for the less aggressive setups. Small trash consisted almost exclusively of green leaves, showing little variation among treatments.

An additional variable evaluated on the cylindrical reclaimer was the drum rotational speed. All tests reported to this point were run at 22 rpm. To test speed, a set of runs was made at 40 rpm. Results are shown in fig. 6. These tests were made at the medium-aggressive, full-insertion setting with teeth angled at 30 degrees from the direction of travel. Minor differences were seen between the two treatments, but plants tended to disengage from the drum more quickly at the higher speed. At 22 rpm, plants frequently became entangled in the drum teeth and made several rotations before exiting the device. Multiple rotations are not necessarily bad in that additional pods may be stripped from the plants. But in a production setting the cylinder may tend to become plugged with plant material at the slower speed.

Bar graph of effect of rotational speed on pod reclaiming with a spring-toothed cylindrical device

Figure 6. Effect of rotational speed on pod reclaiming with a spring-toothed cylindrical device

Fixed-finger reclaimer

The results of the fixed-finger reclaimer tests on dry 'Sonora' material are shown in fig. 7. The differences between 3- and 4-inch spacings were minimal. There was a slight increase in free pods at the 3-inch spacing, and a large increase of broken pods at the four-inch spacing, which does not seem logical. The four-inch spacing appeared to produce longer sticks than the three-inch spacing. Having longer sticks could help in the separation of material downstream. Although the data are somewhat variable, obtaining a 60- to 70-percent pod separation rate is encouraging.

Bar graph of effect of pin spacing on operation of a fixed-pin reclaimer

Figure 7. Effect of pin spacing on operation of a fixed-pin reclaimer

Results of duplicate runs made on greenhouse plants are shown in fig. 8. With green plants, the device was able to separate only 25 to 40 percent of pods as free pods. The percentage of attached pods was high for both spacings, but broken pods occurred less frequently for the 4-inch spacing than the 3-inch spacing. Stick breakup was greater for the 3-inch setup as shown by the percentage of short sticks.

Bar graph of results of fixed-pin reclaimer tests on greenhouse plants

Figure 8. Results of fixed-pin reclaimer tests on greenhouse plants

From these data, we conclude that devices based on the principle of combing plants across a bed of opposing teeth will break plants into smaller pieces and separate easily detached pods. However, they are not particularly adept at removing strongly attached pods from plant material.

Prior test

Prior to these tests, in early 2004, a different reclaimer concept was evaluated. The laboratory prototype consisted of a discarded wire drum washer with wires cut away to make openings that pods could fall through as the drum revolved (fig. 9).

Photo of rotating drum reclaimer with openings for reclaimed pods

Figure 9. Rotating drum reclaimer with openings for reclaimed pods

Plants were fed into the drum using a conveyor. We assumed that the tilt and rotation of the drum would cause pods to fall through the openings as plants and branches progressed through the drum. We also assumed that a stripper bar would be needed to detach pods that fell through the slots. In a test using greenhouse plants with attached pods, only 40 percent of pods fell through the slots. That was deemed inadequate, so no further tests were conducted.

Test results for the fixed-finger reclaimer were sufficiently encouraging that a test program was planned for 2005 and a small prototype machine was constructed.

Test Results, 2005

The first laboratory-tested reclaimer prototype in this test series (fig. 10) was based on the fixed-finger concept tested the previous season. It was designed to operate on pods and sticks separated from the product stream, with detached pods returned to the stream. Material was fed into the machine on the conveyor. Branches and plants were picked up by a pinwheel of short, fixed fingers, then moved over an open grate by a pinwheel with longer fixed fingers.

Photo of top view of the first reclaimer tested in 2005

Figure 10. Top view of the first reclaimer tested in 2005

After five test runs, it was apparent that the mechanism was not satisfactory. The pinwheels tended to retain material in the machine rather than providing a flow through. Also, all material tended to pass through the grate rather than having pods pass through and sticks flow over, as was planned.

The reclaimer mechanism was modified to address these problems. The small pinwheel was replaced with a blade pickup that was less aggressive in engaging the material. A shield was placed between the teeth of the large pinwheel to ensure that material would be disengaged after being pulled through the mechanism (fig. 11).

Photo of modified reclaimer showing the blade-type pickup wheel and shielding installed on the larger finger wheel.

Figure 11. Modified reclaimer showing the blade-type pickup wheel and shielding installed on the larger finger wheel.

The grate below the large finger wheel was changed to a smaller mesh to prevent all material from falling through. The change was from a slot width of 3 inches to about 2-3/8 inches (fig. 12). Although the difference appears small, it proved to be significant. In our tests, virtually everything fell through the larger mesh, while almost nothing came through the smaller screen.

Photo of grates installed below the large pinwheel to separate pods from sticks

Figure 12. Grates installed below the large pinwheel to separate pods from sticks

Eight tests were run with this setup, using fresh plant material harvested at the LPSRC. Pod counts show that about 35 percent of pods passed through the screen, with 40 percent of those being damaged; about 65 percent of pods were carried over the screen with 34 percent damaged; less than 1 percent of pods were retained in the mechanism, all as pods attached to sticks.

Effect on sticks was determined by weighing samples from the machine's outlets. About 80 percent of sticks passed over the grate, whereas 13 percent fell through and about 7 percent were retained in the mechanism. The data for both pods and sticks were very consistent for the eight test runs. Differences in pinwheel speeds between 88 rpm and 176 rpm had no observable effect.

These were the best results found to that date, although 35 percent reclamation is probably unsatisfactory. Also pod damage was high for pods falling through and for those passing over the grate. At 80 percent, stick separation was excellent. If no better mechanism is found, it would be possible to add another stage of separation by installing a second large pinwheel and grid to improve pod recovery.

In an attempt to improve pod separation, the grid below the large pinwheel was replaced with a curved sheet-metal plate with slots such that spring fingers could be inserted between the rod pins of the large pinwheel (figs. 13 and 14). This was a smaller version of the
spring-finger reclaimer that was built and tested the previous season.

Photo of close-up view of the spring fingers installed between the large-pinwheel fingers to increase separation of pods and sticks.

Figure 13. Close-up view of the spring fingers installed between the large-pinwheel fingers to increase separation of pods and sticks.

Diagram of cross-section drawing of the final reclaimer configuration

Figure 14. Cross-section drawing of the final reclaimer configuration

A total of 10 tests were run on this device, and data were collected on pod separation and damage. Since all material went through the machine, no data were collected on sticks. Obviously, a secondary mechanism would be required to sort pods from sticks.

The testing program consisted of two runs at each of four speeds. Only one plant was fed through the machine for each run. Two additional runs were made with one row of spring fingers removed since the apparatus appeared to retain material with two rows. Data were collected by pod counts and pod weights. Often only remnants of damaged pods were found, making pod counts difficult and probably inaccurate; therefore, only weight data will be reported.

The average of eight runs showed that 53 percent of pods were removed from the plants, but about half of those were damaged by the machine. Of pods that remained on the plants after each run, about one-third were damaged. Three percent of pods were retained in the device, which is a small number but indicates the machine is unsatisfactory for field application. When one set of fingers was removed, the number of pods separated dropped slightly, and pod damage was also reduced. With this configuration, no pods were retained in the mechanism.


From these tests we conclude that the most practical device to date is the pinwheel separator with one row of spring fingers followed by a separation device, probably a tumbler cleaner.

To find more resources for your business, home, or family, visit the College of Agricultural, Consumer and Environmental Sciences on the World Wide Web at

Contents of publications may be freely reproduced for educational purposes. All other rights reserved. For permission to use publications for other purposes, contact or the authors listed on the publication.

New Mexico State University is an equal opportunity/affirmative action employer and educator. NMSU and the U.S. Department of Agriculture cooperating.

August 2006