BSF Metrics & Yields| Scale Up Production of Black Soldier Flies : The Life and Times of BSF (Black Soldier Flies)
DipTerra LLC      
HomeAbout UsProducts & ServicesPublicationsForumsQ&A'sBlog IndexBlog

image of DipTerra logo
Insect-based Processing of Biodegradable Waste
Contact Us

BSF Metrics & Yields| Scale Up Production of Black Soldier Flies

by Terry Green on 01/31/14

A major barrier hindering large scale continuous year round production of Black Soldier fly larvae (BSFL) is the absence of practical knowledge on how to design, build and economically operate a BSFL production facility. This blog looks at (i) food scrap loading rates relative to BSFL processing space needed; (ii) cyclic fluctuations and average yields in BSFL production accompanying continuous operation of a BSFL facility; and (iii) the average feed conversion ratio associated with raising BSFL on food scrap waste.

We collected our data from field tests at our pilot BSFL production facility (see Scaling Up Black Soldier Fly Food Scrap Processing | Phase III).  Total bin space dedicated to these measurements in our pilot plant is 6 square meters. Here are our findings:

  • Propagation (mating, egg laying, and the hatching of new larvae), larval processing of food scrap and harvesting of larvae works best (with but a exceptions) when all operations are done in the same enclosed environment;

  • Larval exit ramps are unnecessary (Fig. 1);
  • The installment of passive continuously operating drainage systems in bins reduces time and labor in dealing with leachate as larvae feed and grow off of the waste added to bins. The volume of leachate (in L) varies from about 3% to 5% of the initial wet weight of food scrap (in Kg) placed in bins;
  • Bins without thermostatic controls run on average about 5 to 10 C above the ambient temperature outside the bins;
  • Cyclic fluctuations from day to day in BSFL harvest yields (peaks and troughs) are normal and to be expected (see Fig. 2); and
  • There is an optimum food scrap loading rate in bins (~ 5 Kg per day per square meter) which produces the best yield of harvested BSFL while operating bins continuously year round. Loading bins with food scrap in excess of what larvae are able to consume results in accumulation of sludge and a reduction in BSFL yield. Loading bins at too slow of a rate results in slower growth, avoids build up of sludge, but also results in a sharp reduction in the yield of harvested BSFL.
image of prepupae climbing vertical wall
Fig. 1. Image of BSFL prepupae climbing with ease vertically up and over the top walls of bins. Larvae have no difficulty climbing vertically up and over vertical walls as they egress from the bins.  Copyright © 2014, Terry Green, All rights reserved.

We consistently and repeatedly observed dynamic oscillating changes in the BSFL population density inside the bins driving cyclic fluctuations in larval egress rates  (Fig. 2).  Very young larvae (newly hatched from egg clutches) are so small that the density of young larvae per unit volume reaches very high levels (>15 per ml). As the young larvae grow, the volume they occupy per unit volume of bin space increases sharply. With heterogeneous populations of larvae of varying size co-existing in the confined spaces of the food scrap processing bins, population pressures (larval neighbor to neighbor interactions) appear to trigger cyclic intervals of mass egress linked with the growth cycle of the new larvae competing for feed and space with older larvae.

image of graph showing BSFL yield at varying intervals
Fig. 2. BSFL harvest yields (Kg per day per square meter) measured on food scrap processing bins beginning at the time the bins received food scrap and an inoculum of BSFL (day 1) and progressing over an approximate four month interval at an optimal average food scrap loading rate of 5 Kg per day per square meter. Temperatures in the bin held relatively constant between 30 and 40 C throughout the interval monitored. Copyright 2014, Terry Green, All rights reserved.

Table 1 summarizes the metrics of BSFL production while holding the food scrap loading rate of bins constant. We calculated the cumulative average yield of BSFL (Kg wet weight per day per square meter bin space) while aiming for a target optimal loading rate of ~ 4.8 Kg food scrap (wet weight) per day per square meter bin space based on prior work designed to find an optimal food scrap load rate. The optimal food scrap loading rate was determined by measuring BSFL yields taking into account loading rates ranging from less than 5 to as much as 30 Kg per day per square meter. At loading rates in excess of ~ 5 Kg per day per square meter we observed that the larvae fell behind in their ability to keep up with the food scrap loaded into the bins. They handle all of the food scrap presented to the bins at loading rates of ~5 Kg per day per square meter or less.
Table of data summarizing BSFL metrics and feed conversion rate

Table 1 provides information that can help those interested in scaling up BSFL production to predict the total footprint size of processing bins and troughs needed to achieve a specific annual yield, and how much food scrap would be needed in meeting a set production rate. Using the FCR in Table 1, for example, it takes on average about 9.35 MT of food scrap (wet weight) to produce 1 MT (wet weight) of harvested BSFL (~ 18.7 MT to produce 1 MT of dried BSFL). To calculate the footprint in bin space needed to produce a target amount of BSFL, simply divide the cumulative average BSFL yield into the target production rate desired. For example, a plant designed to produce 100 MT of BSFL (dry weight) will require 100/0.09, or ~1100 square meters of bin space.

Comments on this blog, or any of our other blogs, are always welcome. Follow us through our RSS feed. For additional information or follow-up questions, visit our Q&A's or Forums page, or Contact Us (http://www.dipterra.com/).


Comments (15)

1. Holly Cnana said on 5/9/14 - 10:47PM
What would a 2 square meter bin produce of BSFL in dry weight? I don't understand MT, could you give your answer in kg or lb.
2. Terry Green said on 5/12/14 - 06:24AM
The yield (dry weight) per day of self-harvesting BSF larvae, mainly prepupae, per square meter fed food scrap waste is approximately 0.2 Kg (0.44 lbs) per day (73 Kg per year). For a 2 square meter bin the yield is double this value. One MT (metric ton) is 1000 Kg (2,204 lbs), slightly more than a ton.
3. Jeff C said on 11/11/14 - 08:01AM
Would you expect the FSR to be higher if manure were used as the substrate?
4. Terry Green said on 11/11/14 - 03:41PM
Yes. Manure has less nutrient value remaining then food scrap. Microbes growing off of the manure probably would increase its nutrient value somewhat, but generally one would expect the FCR to be higher for manure relative to that of food scrap.
5. Daniel Michaelson said on 11/22/14 - 05:01PM
Questions: 1. Are you breeding BSF in the same space as your reactors? If so, then are you supplementing the light over the winter? 2. Did you have any problems with excessive heat in the shed over the winter? 3. Can I come see your operation and discuss research needs with you?
6. Daniel Michaelson said on 11/22/14 - 05:11PM
How did you calculate your last metric of % FS converted into BSFL i.e. bioconversion rate? .51/4.76 would yield a 10.7% conversion rate.
7. Terry Green said on 11/23/14 - 07:12AM
The % FS converted into BSFL is calculated on a dry weight basis. Feedstocks vary in water content. Water, itself, has no nutrient or caloric content even though it obviously serves an essential role in sustaining life. Based upon an average water content of BSFL ~ 50%, and that of FS of ~ 80%, bioconversion (on a dry weight basis) is 26.75% (0.51/4.76 x 100/20 x 50/100), or 27% rounded to the nearest significant figure as expressed in Table 1. We breed BSF in the same space as our processing bins year round relying solely on natural light. We have not experienced problems with excessive heat during the winter. Our pilot plant is not open for touring. For general questions concerning the management of BSFL or questions about consulting work we do, I can be reached by email at greent@dipterra.com.
8. Chris slawson said on 1/17/15 - 01:40PM
These figures aren't consistent with an ebook you guys made: "under ideal conditions a BSF recycling bin with about 100,000 larvae per m2 processes upwards of 3 lb per sq. ft (15kg/m2) of food scrap per day". Assuming a 10 dry weight conversion, this is .54MT per year per m2. this would mean 1m2 would produce 3kg of bsf per day going at 25% conversion..?
9. Terry Green said on 1/17/15 - 02:37PM
Chris, the dry weight conversion factor (wet weight food scrap:dry weight larvae) is ~ 20:1, not 10:1. We also have found through experience over the long haul that on scaling up it is necessary to back off on the loading rate of food scrap. We have found as stated in our blog that the optimal feed rate is closer to ~ 5 Kg food scrap per day per m2 in scaling up production. In small backyard bins, there is more tolerance for overfeeding. Adding excess food scrap in scaled up operations results in an accumulation of partially degraded food scrap which becomes difficult to work with and which interferes with the efficiency of BSF production. The numbers reported are consistent if you work through the calculations as noted in our table. Depending upon the feedstock used, there may be some variation in the BSF yield. Several complex practical and environmental factors in combination bear on production yields, so attention to detail is important in producing BSF from food scrap waste.
10. Chris Slawson said on 1/17/15 - 03:24PM
So the "compost" leftover from the larvae eating so much interferes with production at that level of feeding eventually? Also, in the book it says "the production of larvae... Ranges from 10 to 30% with food scraps as the principle foodstock." even if the larvae were 44 percent dry weight, that can be quite a bit more than 5%?
11. Terry Green said on 1/17/15 - 04:12PM
The range mentioned refers to the production efficiency of BSF in converting food scrap into larval biomass. This is calculated by dividing the dry weight yield by the dry weight of food scrap fed larvae. The water content of food scrap is typically 70 to 80% and must be corrected for in calculating the true conversion rate. Excess food scrap decomposes into a complex mix through the action of larvae feeding on it, but also microbes, and can lead to colony collapse if not addressed properly. We will be posting more on this subject shortly.
12. Irwan Santoso said on 9/13/16 - 01:37AM
Have you experimented FS with different protein content? Would a higher protein content result in lower FCR? If you use FS with 100% protein content, would the ratio be 1:1?
13. Terry Green said on 9/13/16 - 12:21PM
Irwan, I do not know the effect of varying FS protein content on the FCR. I however doubt that this would have much effect on the FCR. Keep in mind that larvae growing off food waste compete for nutrients against microbes and detritivores. This competition, itself, impacts the FCR negatively. An FCR in the range of 4:1 (25% conversion efficiency) for BSF grown off decaying FS under the circumstances is not bad. Fish, growing under conditions absent this type of competition, generally among the most efficient feeders known, can in some instances achieve FCR’s in the range of about 2:1. Note it is not possible to achieve an FCR of 1:1. This would imply 100% conversion of feed into biomass which violates the “Law of Conservation of Energy” since some of the nutrient energy recovered from feed must be spent in powering muscle activity and motility, some is dissipated as heat, some in varying forms of chemical energy required in building cellular structure, etc.
14. Bernardo Kostich said on 11/2/18 - 07:52PM
Hello, I'm trying to read all of your articles and willing to buy your books because I want to start a small scale BSF production. There is something in your model of larvae production that I can't figure out. The fact that the adult flies are not isolated form larvae means that there are larvae at mutiple development stages inside the bins right? I know that prepupae crawl outside the bins by themselves but how can you separate the residue from the young larvaes when it accumulates? I have seen many videos and most of them have the adult flies and larvae in separated enclosures. You are one of the few people who suggest that adult flies and larvaes should be in the same place. Why do you say your method is more efficient? Thank you and regards from Chile.
15. Terry Green said on 11/3/18 - 10:43AM
Bernardo, see http://www.dipterra.com/blog.html?entry=steady-state-farming-of-bsf. Growing larvae continuously seek out and feed off nutrients in waste. Adding food scrap waste and newborn larvae to secondary bioreactors at a rate commensurate with the turnover of nutrients in the waste, and larvae (linked with the exit of prepupae from the bioreactors), provides for continuous, uninterrupted harvesting of prepupa year round. Steady-state operations require a priori sustaining a heterogeneous population of larvae in bioreactors. About 5 to 10% of the waste residue (including larvae growing in the latter waste) gets set aside in separate bioreactors. These latter “spent waste” bioreactors receive no incoming waste or newborn larvae. Larvae in these bioreactors within a week or so clear free of the left over waste in seeking nutrient-rich secondary bioreactors receiving incoming waste as the nutrients get depleted in the latter bioreactors leaving behind principally “spent waste”.


Leave a comment


Dipterra's Blog - The Life and Times of BSF (Black Soldier Flies)