Print This Page
Automatic Feeders, Extruded Feeds and Pelleted Feeds

Shrimp News: The following long discussion about shrimp feeds took place on The Shrimp List in late December 2016.  To understand some of the terms used in the discussion requires a little background information.  For example, both extruded feeds and pelleted feeds are marketed as “pellets”, making the distinction between the two a little confusing; however, on The Shrimp List, they are referred to as “Extruded Feeds” and “Pelleted Feeds.”  So that’s what I'm going to call them here.

What's the difference?

Extrusion is a process in which feed is ground, quickly cooked at a high temperature, and processed under high pressure.  The feed is then pushed through a die, which gives the feed a unique and uniform shape.  Adjusting parameters such as temperature and pressure allows the manufacturers to make pellets that suit different farming methods, for example, feeds that float or sink slowly and feeds suited to recirculation systems.  The more grains in a formula; the more likely extrusion will improve the feed’s digestibility.
Pelleted feeds, on the other hand, are not cooked and are prepared at much lower temperatures and pressures and then pushed through dies.

The similarity of pellet feeds and the extruder feeds can be summarized as follows:

• Both utilize steam, pressure and temperature.

• Both use dies to form the pellets.

• Both can improve the digestibility of the feed ingredients.

Here are differences between the two milling technologies:

• A pellet mill shapes the pellets by compression through long land dies, which makes the pellet denser than extruded pellets.  In most cases, a pellet binder is needed in the formula to hold the pellet together.

• Because of high temperatures during extrusion, substantial gelatinization of the starchy ingredients takes place, expanding the pellet and increasing its porosity, while reducing its density.  With extrusion, there is no need for additional pellet binders in most cases.

• Extrusion allows feed manufacturers to use higher levels of oil and fat, creating a tighter pellet with fewer fines (small, dusty particles).

The bottom line, neither processing method is better than the other.  The choice depends on the needs of the end user, the shrimp farmer in this case, who might use different feeds in extensive and intensive ponds, or in recirculating-tanks and open-pond systems.  The only way to ensure proper nutrition is to focus on the feed formula!
Sources: 1. Star Milling Company’s Website.  Which Is Better: Extruded or Pelleted Feed?  Website Visit on December 31, 2016.  2. Wikipedia.  Commercial Fish Feed.  Website Visit on December 31, 2016.  3. Insta-Pro’s Website.  Pellet Mill Vs. Extrusion.  Nabil Said.  Website Visit on December 31, 2016.


The Shrimp List Discussion

Jorge Cordova ( I would like to know if it’s better to use pelleted or extruded feeds in conjunction with automatic feeders.

Billy Setio ( Jorge, many farmers use automatic feeders in Indonesia, claiming they reduce feed conversion rations (FCRs) by 0.2 to 0.4, depending on their feeding strategy.  Keep in mind that many farmers in Indonesia stock at high densities, from 150 to 300 postlarvae per square meter (PLs/M2).  They say FCRs drop because they use less feed and more frequent feedings.  They use feeding trays to determine when to stop feeding.  If your FCR in less than 1:1, you don’t have much to gain with automatic feeders.  I’m not very comfortable with automatic feeders because I don’t have the controls in place to determine their effectiveness, but I’m testing them in on of my ponds.

One advantage of automatic feeders is that you minimize nutrient leaching because you feed a little bit at a time, and the shrimp eat it right away.

Laurent Queffelec ( Jorge, I was stocking around 25 PLs/M2 and did not activate automatic feeders until the shrimp reached three grams.  That’s when we could hear them feeding with AQ1’s Acoustic Feeding System.  We would start with two automatic feeders in a 3.5-hectare pond and add a third one when total daily feed consumption reached 210 kilograms (kg).  When it reached 280 kg, we could use all four feeders installed in the pond.  After our first partial harvest (+/- 4,000 kg), we dropped back to three feeders and then to two after the second partial harvest of 4,000 kg.  We also found that when shrimp reached 12 to 15 grams mean weight, the AQ1 software was attempting to feed the maximum amount and we had to adjust it to our cost structure and harvest plans.

Jorge Cordova ( Thank you Billy and Laurent for your comments.  Have you been able to observe any particular critical biomass that once attained would diminish say growth rate relative to the number of automatic feeders in a pond?

Here are some of my observations.  Perhaps you would like to comment on them?

• Extruded feeds seem to be consumed faster (suggested by AQ1 sound feeders).  Because extruded feeds are softer than pelleted feeds, shrimp might eat them faster.  Also, there’s no dust with extruded feeds to block the automatic feeders.

• FCRs definitely get better with automatic feeders.

Durwood Dugger (, Jorge, I’ve run hundreds of feed evaluation studies in the shrimp nutrition test labs I’ve built for both extruded shrimp feed producers like Purina and pelletizing feed producers like Zeigler Bros.  Comparing feeds with the same nutrition formulas and the same ingredients from the same source and batches with similar ingredient grinds, binding agents and in a controlled test tank environment, there are not enough difference in performance to measure the differences between extruded and pelleted feeds.

At risk of stating the obvious, shrimp, unlike fish, do not ingest whole feed pellets.  Animal nutritionists at feed mills that manufacture ten different animal feeds all of which consume their feed pellets in nearly whole form often miss the significance of this fact.  Additionally, as humans with ten meters of convoluted gut structure packed into their abdominal cavity, it’s hard for us to conceive how an animal with a straight gut, like shrimp, digests its food.

No matter what form the feed is in—extruded pellets or pellet mill pellets—the shrimp use their mandibles to grind the pellet up into particulates small enough for them to swallow.  Feed pellets fine enough for the shrimp to swallow whole are technically difficult and economically impractical to manufacture beyond the hatchery stage.  The degree of starch bonds and or other binders used in either pellet mill or extruder feeds do not prevent the shrimp from taking the food pellet apart and eating its most attractive ingredient components first.  If you have ever done feeding studies you might have noticed that the bottom of the tank will have soy, bran and other harder ingredient particles littering its bottom shortly after feed introduction.

Maintaining feed pellet integrity is not a simple matter.  Like most things, there is an optimum with negative extremes on both sides.  Having a feed pellet that literally explodes on impact with the water is obviously not desirable.  Contrarily, when you see a feed pellet intact two hours after introduction to a healthy shrimp biomass, its continued stability says far more about the feeds lack of attractiveness to the shrimp than it does about the value of its stability.  Additionally, observing the amount of “fines” in a feedbag also tells you whether the feed is formulated correctly, sufficiently bound and well manufactured.

The shrimp has a straight gut with a very high passage rate compared to fish and higher animals.  This is especially true for highly attractive feeds that are readily available, exciting shrimp into their hydraulic feeding mode, where the feed going in the front end hydraulically forces it out the back end.  I have measured large adult shrimp passing feed from mouth to fecal strand in less than six minutes!  If you know much about the biology and chemistry of feed digestion, you probably also know that not much happens enzymatically or bacterially wise in six minutes.  So, the question is raised as to how the shrimp gets adequate nutrition if it passes feed ingredients so quickly?

What many shrimp farmers and quite a few shrimp feed nutritionist don’t understand about shrimp like P. vannamei is that they are primarily detritivores.  They get the soluble added nutrients (including vitamins and minerals) off the feed particles they eat.  When they tear a feed pellet apart and eat it, they don’t digest many of the feed ingredients, but they do coat them with a gel of gut secretions—that contain a host of enzymes—and excrete them as a fecal strand.

Here comes the appetizing part: The shrimp re-eat the fecal strands.  Remember, these fecal materials have been exposed to the shrimp’s digestive enzymes and bacteria while passing quickly through the shrimp.  This enzymatic and bacterial action digests the feed ingredients in the fecal strand, making them much more absorbable on the second, third and later digestive passes through the shrimp’s gut.  This re-eating process continues until there are no longer any attractive nutrients that the shrimp can detect with its chemoreceptors on their forward walking legs and antennules.  Experiments confirmed this observation.  In a self-cleaning tank with a flow rate that quickly removed uneaten feed and fecal strands, the shrimp shrank and eventually starved to death, even though they had access to more than adequate fresh formulated feed.

Spend some time watching shrimp in an aquarium or diving in pond and watch how they continually probe the pond bottom for potential food particles.  Shrimp feeding activity does have photoperiod-based peaks; shrimp eat almost continuously.  Understanding this eating process should also tell you why nutritionally comparable diets—whether competently extruded or pelletized—have little if any difference on the diets performance in comparable environments.  Further, it should emphasis that the sizes of ingredient component grinds are very important in both the internal and external digestion process of the shrimp.  It also gives some insights into why shrimp do so well in biofloc systems: many bioflocs have feed particulates as their nuclei.

Understanding the extended (internal and external) digestion processes used by the shrimp when consuming formulated feeds also gives us insights into why multiple feedings provide measurable benefits up to a certain optimum point.  First, and probably foremost, strategic multiple feedings at peak feeding times insure that shrimp will ingest the feed while it still has some of its water soluble nutrients (vitamin and mineral premixes) associated with the feed particles, making water soluble leaching of nutrients less of an issue.

Understanding how feed manufacturers design their feeds to optimize the unique eating and digesting processes of P. vannamei allows us to determine how much we can invest in a quality feed before its costs exceed its benefits.  Nutritional quality standardization of the feed ingredient nutrition, the uniformity of ingredient sizing (grinding and sieving) and how the feed particles are bound and sealed in the pellet determine feed quality.  In a perfect shrimp-feed world, all the feed ingredients would be wrapped in a highly attractive (tasty to the shrimp), waterproof sealant that prevented soluble ingredient leaching, particularly the leaching of vitamin and mineral premix materials.  This would theoretically make all the various feed ingredients (wrapped in an attractant sealant) tasty and sought out by the shrimp.

Considering all of the above, we still have to adapt the feed and its formulation and design to the shrimp’s environment.  Earthen ponds and plastic lined ponds and or tanks should necessarily have their feeds adapted to them to optimize attraction, nutrition, consumption and digestion.

I would also like to point out that shrimp production levels are significantly impacted by the investments you put into feed design, multiple feedings and feeding equipment.  I remember an analysis of multiple feedings done in Hawaii a good while back (perhaps by Dr. Spencer Malecha) where a point of diminishing returns was reached at five feedings per 24 hours. I’m sure the particulars of the experiment impacted that result, but it might be worth looking up.

Hope these P. vannamei feed observations help.

Juan Aguirre ( The abstract of the following study might be of interest to some of you that are following this discussion: Spatial Distribution of Digestive Proteinases in the Midgut of the Pacific White Shrimp (Litopenaeus Vannamei) Indicates the Existence of Endo-Ectoperitrophic Circulation in Crustacea.

Frank Richardson ( Mr. Dugger—Excellent information!  You clarified everything for me.  (By the way, I did my summer internship for the Corp of Engineers shrimp project you led, around 1990).

Jorge Cordova ( Durwood, there was lots of good data in your post.  I will read it again and again, just to make sure I absorb it all.  A couple of years ago we used pellets and extruded feed in hundreds of hectares of earthen ponds, stocked at 11 PLs/m2.  We distributed the feed by casting it once or twice a day.  The harvest data showed no significant difference between extruded and pelleted feeds.

Once we started delivering feed with the AQ1 sound system and automatic feeders, we started to see significant improvements in production.  We have found that growth rates drop as more and more shrimp crowd around the automatic feeders and that it becomes necessary to add more feeders.

Now we are studying the advantages and disadvantages of extruded and pelleted feeds with these systems.  Since shrimp bite off pieces of the pellet to eat, we think extruded pellets may be better than milled pellets.  Do you have any comments on this?

Eric Muylder (, Durwood, I agree with everything you say.  To further add to it, I would like to clarify that extruded and pelleted feeds generally don’t have the same formula, but do have the same nutritional value.  As you state, if the grinding of the raw materials is not fine enough, some raw materials will be lost because they are too big for the shrimp to eat.  Also, extruded feeds, mixed at high temperatures and pressures, make it more difficult for the shrimp to select pieces, compared to a pelleted feed, where raw materials are only compacted together and pieces remain intact.

The main challenge in extruded feeds is making them 100% sinking.  If you use the same formula used in pelleted feed—high-quality wheat flour, gluten and binders—your feed will almost certainly float.  The advantage of extrusion is that you can use fewer binders and wheat or a cheaper starch source instead of wheat flour.  So generally, you can save about $60 per ton in raw material costs.  Of course, the cost of production with extrusion is higher than pelleted feed (investment, energy, spare parts), which is the main reason most shrimp feeds are pelleted.

Also extruded feeds generally absorb water faster (better gelatinization of starch), which is something the shrimp like, but the farmer gets the impression that the water stability is not so good.

I like your comment about the loss of attractability after leaching.  Most shrimp farmers still check feed trays after two hours, which seems to me too long.  With goldfish, whatever feed is not consumed in 15 minutes is considered overfeeding.  We should follow that rule.  I think the spectacular results obtained with automatic (continuous!) feeding prove this.  Small bits and pieces all day long provide the best retention time in the gut and the best digestibility.  As you state, a hungry shrimp consuming a lot of feed will have the first feed come out quickly, while the last part might take 4-5 hours before it has completely passed through the gut.

Jorge Cordova ( In our ponds, we have observed that when extruded feed is distributed from an automatic feeder, it covers a much smaller area than pelleted feed from an automatic feeder, which seems to have a negative impact on growth.

Laurent, partial harvests seem to bring a new challenge to acoustic feeding systems.  Resetting the acoustic feeding software to reduce the amount of feed distributed appears to be the way to go after a partial harvest.  Do you know the numbers of kilos per feeder that were being fed right before your first and second partial harvest?  What were the kilograms per feeder after your final harvest?

During one of my crops, I partial harvested and did pay attention to lowering the amount of feed, so in that particular case (2 ponds), I did not see the lower FCR effect.  In cases where no partial harvests occur, the FCR drop is evident.

Laurent Queffelec ( Jorge, before the first partial harvest, quantities per feeder per day were around 100 kg.  We estimated that no more than 30% of biomass was harvested, so the 400 kg daily feed was reduced to 280 kg in three remaining feeders (+/- 93 kg each).  After resetting the software, if the feeders are empty in less than 12 hours (we try to feed 20 hours a day), then we could increase daily feed to 300 or 320 kg.  Otherwise, we let them starve a bit and check weekly growth.  The same strategy is used after the second harvest.  The third feeder is abandoned, so the shrimps have to concentrate more around the remaining feeders, allowing for better sound response from the AQ1 system.

We also have tried to let the software manage it all by filling all the automatic feeders, so the shrimp could eat as much as they wanted.  Survivals dropped, FCRs increased to unsustainable values, and no better growth was observed.

Durwood Dugger (, I think it is important to understand that there is far less science in feed extrusion and pellet mill feed production processes than most people suspect and far more art in the mill operator’s skill level than most of us appreciate.  With comparable ingredient specifications, it is difficult to discuss extruded versus pelletized feed without production quality and technique variables confounding the comparisons and any differences that might be achieved.

Additionally, we tend to underestimate the value of nutritionally standardized feed ingredient sources.  Much of the feed performance variability we see in shrimp feeds in lesser-developed parts of the world comes from non-standardized ingredients that have variable nutrition impacts on shrimp growth and survival.  Again, with higher stocking densities, shrimp are more dependent on formulated nutrition inputs and more sensitive to nutrient variations in their diets.

I don’t completely agree with Eric assumptions that good pelletizing processes don’t generate high temperatures from the formulas and pressures used to produce adequate shrimp feed stability—if done correctly.  I agree that better binding is less assured with pelleted feeds, but this is circumstantial factor reflecting smaller less sophisticated pellet mill production operations more than an actual pellet mill capability factor.  Pellet mill operations tend to be less sophisticated in general and don’t necessarily realize the full quality capability of their equipment.  I would be more concerned about their ingredient buying capability and ingredient storage than in a high volume extruder mill.  That said there are some exciting new feed ingredient storage technologies that can minimize feed ingredient breakdown concerns.

We should also be aware that economies-of-scale are a big issue with extruded feed costs.  Extruders need to operate 24/7 to prevent extremely high maintenance costs.  When an extruder stops, the feed inside solidifies and has to be chiseled out, requiring the operator to have high enough sales to satisfy 24/7 operation (think millions of tons).  This means the extruder may run shrimp feed for a few hours once a month and then without stopping switch to dog, cat, monkey, hamster, iguana—whatever food it takes to keep the extruder operating.  Don’t be surprised to find and see some pet food shapes—stars and moons (or whatever cutesy pet food marketing shapes are popular)—in extruded shrimp feed because that was the pet food that ran in the extruder before your shrimp feed.  Equally, don’t be surprised if your dog complains because his food may have a “fishy” taste occasionally.  Also, understand that the machinery necessary to make formula/ingredient changes on the fly while the extruder is running and without stopping makes the extruded production process much more complicated and expensive.  Hopefully, if your extruder feed provider is large enough, its quantity ingredient purchases will offset the higher production costs of the extruder.

Pelleting operations are much more flexible and less expensive to build and operate because they don’t have as much start/stop maintenance and use less expensive machinery, but like the extruder feeds, they do require a very skilled mill operator and a nutrition formulator team to achieve an optimally nutritious and consistently formed feed ration.  I have yet to see any better FCRs using extruded versus pellet mill feeds when all other variables were the same in well designed and documented feed comparison tests.

I agree that pelletized feed production can be more sensitive to ingredients, often requiring reduced fat levels to produce the needed pressures (for a given diameter pellet) for good binding.  On the other hand, a post-production fat overspray can be an ideal time to add potent attractants and sealants.  Again, I don’t think that extrusion necessarily combines the ingredients any better under similar pressures and temperatures (though extruders can produce higher pressures).  Formula, starch and binder levels, comparable ingredient sizes and operator skills make greater differences in the feed outcome.  Larger ingredient sizes and the hydroscopic nature of many feed ingredients make the feed less stable the larger the particle size, even with comparable binding abilities.  In other words an extruded feed with large ingredients will be less stable than the same formula extruded feed with smaller ingredients.  Same with pellet mills.

Regarding feeding zones: Observing, understanding and interpreting shrimp behavior is one of the best management tools we have.  One of the interesting behaviors we have noted in underfed ponds that have ideal growth parameters is perimeter schooling of the shrimp.  Add some feed, and the schooling immediately disappears.  Beyond the obvious “I’m hungry” communication of the schooling should also be the idea that shrimp move around a lot when food is not present in sufficient quantity.  Consequently, if you are using a quality feed that the shrimp like, they will find it very quickly.  The number of feedings and number of feeding stations needs to be optimized with the given feed performance including its stability, particularly in terms of water-soluble nutrients.  I don’t worry so much about the solid ingredients of good quality and size.  The shrimp will eat them sooner or later and again and again.  However, the water soluble and very expensive vitamin and mineral premix materials are far more sensitive to how soon the shrimp find and eat them.

I don’t think there is a universal optimum feeding system beyond understanding these basic shrimp eating/behavior principles and adapting your growout environment to your feed resources and to your other specific production resources and economics.

Juan Aguirre ( Another study that might be of interest to the followers of this discussion: Protein Digestion in Penaeid Shrimp: Digestive Proteinases, Proteinase Inhibitors and Feed Digestibility.

Jorge Cordova ( Laurent, if you ever run extruded feeds with the AQ1 system, let us know if you see a different pattern of consumption and growth.

Laurent Queffelec ( Unfortunately, we never had access to extruded feeds in New Caledonia, and I think that’s still the case today.

Sources: 1. The Shrimp List (a mailing list for shrimp farmers).  Subjects: (1) Extruded vs. Pellets, (2) P. Vannamei Feed Design - Attraction, Nutrition, Digestion, and Feeding.  December 28 to 30, 2016.  2. Bob Rosenberry, Shrimp News International, January 1, 2017.

Print This Page