DS: Because biomass combustion is very complicated and not mathematically described I find that a random approach to experimental design can be productive. This is frequently referred to as the “Black Box” model. I asked Gemini AI about the use of Black Box vs. Open Box models.
Gemini AI: “Black box theory views a more random choice of experiment as a pragmatic compromise used when a system is too complex to fully understand.
Core Trade-Offs
Predictive Value: Black box models excel at predicting what will happen based on inputs. They fail to explain why it happens. Testing inputs and outputs is faster than mapping internal mechanisms. They can fail unexpectedly if hidden variables change.
When Black Box Productivity Wins
Complex Medicine: Testing if a drug works before understanding its exact molecular pathway.
Machine Learning: Using deep neural networks to predict things.
Early-Stage Research: Establishing that a phenomenon exists before funding microscopic analysis.
When Open-Box Productivity Wins
Causal Certainty: Proving a direct mechanism of action, preventing false correlations.
Generalization: Applying a discovered scientific law to new fields.
Troubleshooting: Fixing a faulty experiment by isolating the broken component.
Science uses a cyclical approach. A black box approach identifies patterns quickly and open-box inquiry later attempts to verify the underlying truth.”
https://aprovecho.org/wp-content/uploads/2026/07/black_box_7.17-1.jpg9341920Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-07-17 13:30:062026-07-17 13:30:08Using Black Box and Open Box Experimental Design
I’ll always remember my first moments at Aprovecho 26 years ago. I had just graduated with a degree in mechanical engineering, newly aware of the realities of household energy poverty, and had completed my first solo cross-country road trip to Oregon to work with a small group of folks who would become world leaders in clean cooking. As I arrived, Dean came rumbling down the driveway in his blue pickup and greeted me with his signature joyful warmth. He showed me to the rotegrity dome where I’d start my first experiments, the outdoor biomass-fueled kitchen where fellow interns would cook our food, and the forest clearing where I pitched my tent and shared the space with a few giant banana slugs.
At the time, I had no idea I would still be here decades later—now celebrating 50 years of an organization that has touched lives across the globe. Over the years, Aprovecho’s impact has grown in ways we could never have imagined:
Our rocket stove design principles have spread worldwide, with tens of millions of stoves built using these ideas.
Through our dozens of books, peer reviewed publications, and instructional videos, we’ve kept our knowledge open source and accessible.
We’ve installed emissions monitoring systems and trained researchers in over 67 locations, empowering communities to design solutions that work for them.
We’ve partnered with manufacturers like Shengzhou Stove Manufacturer to expand access to clean cooking technologies at scale.
We’ve also helped shape the broader clean cooking sector—contributing people and ideas to organizations such as BURN, Trees Water People, and StoveTeam International, and leading the ETHOS conference for 26 years.
Along the way, our work has reached unexpected audiences, from features in The New Yorker to awards from the US EPA and Prince (now King!) Charles.
Looking ahead to the coming decades, there’s still so much work to be done. We’d love to see our innovative designs for forced draft combustion be applied to cooking and heating stoves that protect health and climate. We’re also continuing to develop appropriate technologies and innovative measurement tools to support impactful design activities. And with the ongoing support of the Osprey Foundation, we’re committed to continuing to build capacity in projects globally to work with local users to develop and disseminate the best cookstove designs available.
Today, our team of eight dedicated staff members brings deep expertise and passion to this mission. Still, none of this work happens in isolation. We are incredibly grateful for our global network of partners, collaborators, and supporters who share our commitment to clean cooking and practical solutions.
Throughout our history, Aprovecho has largely sustained itself through earned income—grants, consulting, and equipment sales. As we look to the future, we’re working to ensure that this work can continue and grow for decades to come. If you feel inspired to support that vision, we welcome your partnership in whatever way feels right to you.
Join Us to Celebrate 50 Years
We would love to celebrate this milestone with you! Please join us for our 50th anniversary gathering:
Blue Mountain Campus July 25 3:00–7:00 PM
In honor of the occasion, we’re also offering a 50th anniversary t-shirt as a thank-you for a $50+ donation. If you’re in the US and interested in the t-shirt featuring the two images in this email, simply email me at nordica@aprovecho.org by June 22 and specify the size, cut (unisex or women) and shipping address.
Thank you for being part of this journey—and for all that you do to create a better world.
With gratitude, Nordica MacCarty, PhD Executive Director
Aprovecho Research Center
In our recent post “Improving Solar Distillation, Part 1,” we talked about the traditional design for a solar still and pointed out some design weaknesses. Here we show Dr. Larry Winiarski’s suggested improvements to address those problems.
The distillation of brackish water is a two-stage process. Water is encouraged to evaporate and is then condensed. Sealing water inside an airtight vessel results in almost immediate saturation of the inside air. Until the water vapor is condensed, distillation cannot occur. As well, in a classic still, the condensing surface is warm and production of potable water occurs at a significantly reduced rate.
By transporting clean water vapor to a cool condensing surface, Dr. Winiarski addresses this problem. In his design with chimney, ambient air is warmed, improving its ability to absorb water vapor. Condensation then occurs underground, stripping the moving air of moisture. A large, cool surface area results in more effective condensation. A tall chimney acts as the engine, moving warmed air through the system. Air moving up the chimney is dry, after depositing the potable water into a receptacle.
https://aprovecho.org/wp-content/uploads/2026/05/solar_distillation_sm.jpg4221200Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-05-29 13:15:572026-05-29 13:15:58Improving Solar Distillation: Two-Stage Distillation
On their recent trip to SSM in Shengzhou, China, Nordica and Jaden were reminded that many minds are always better than one. SSM were gracious hosts to ARC, OffgridSun from Italy, and Tango Energy from Tanzania for a week. Together, they designed a stove for Tanzania that addressed cooks’ needs, had improved performance, and could be jointly manufactured at SSM and Tango Energy. Meeting in person turned a several-month-long process of emails and phone calls into a streamlined week of design. Maybe when it comes to stove design, there is no such thing as too many cooks in the kitchen.
ETHOS 2026
The ARC team attended another successful ETHOS conference, this year in Portland, Oregon. Sam and Dean hosted Stoves 101, providing a valuable crash course in cookstove design. We also presented on the effects of forced air in stoves, LEMS testing around the world, RTKC capacity building, and much more. ETHOS is always a great time for us to reflect on the work we did and what we learned throughout the year. It’s also wonderful to see what everyone else is working on.
LEMS Around the World: Now in Burundi
Our mission to ensure everywhere has the capability to perform stove emission testing continues. Sam traveled to Burundi where he set up a LEMS and trained (in French) a team at the Laboratoire de Biomasse et de Cuisson Propre et Économe/Université du Burundi. Over 10 ISO tests on various stove types were run with the lab team as well as lab CCTs, a vital test to measure stove performance while performing a cooking task. ARC is now working to add CCT capabilities to their open-source data processing software.
Working in another language takes patience but it allows ARC to work in cross-cultural settings where lab testing, stove design, and market testing come together.
https://aprovecho.org/wp-content/uploads/2026/03/ARC-Logo.jpg10001000Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-05-14 11:07:222026-05-14 11:07:23What’s Cooking at Aprovecho
A typical DIY solar still. (Image from LSU Ag Center)
In this type of solar still, salt water is put in a sealed box with an angled glass top. As sunlight entering the box heats up the water, it evaporates into fresh water, condenses on the glass top and runs down into a collector.
Dr. Larry Winiarski pointed out that traditional solar stills, as above, have problems:
• Sealing water inside a box results in almost immediate saturation of the air.
• Until the water vapor is condensed, distillation cannot continue to occur.
• In a classic still, the glass condensing surface is warm. Effective condensers are supposed to be cold.
To improve production, Larry moved the hot, humid air (EVAPORATION) to a cold surface (CONDENSATION).
He used a chimney to pull air through the system.
In the next post, we’ll check out his most successful design.
https://aprovecho.org/wp-content/uploads/2026/04/solar_still_LSU.jpg13421200Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-04-03 13:32:132026-04-03 14:00:08Improving Solar Distillation, Part 1
“In a high-performance rocket stove, the average gas temperature within a pot skirt typically ranges between 800C and 1,000C. These gases lose temperature as they flow through the channel gap between the pot and the skirt, where they transfer heat to the water.”
One of the interesting things that our lab is trying to do is to “educate” Gemini and other AI models to know how Rocket stoves actually function. The Osprey Foundation is funding us to communicate with lots of folks each week from all around the world, trying to facilitate more improved stoves being in use. Improving and increasing the data publicly available to AI models seems like time well spent as a part of this endeavor. Maybe an easy way to change the world?
An SSM Pot Skirt, and International Training Coordinator Michael Saul with some test data
This week Michael Saul has been sticking thermometers half way down into 6mm channel gaps in pot skirts on four Rocket type stoves. The adjustable, inexpensive SSM pot skirt (as above) is 8cm high. Adding a pot skirt as an intervention may be the most cost effective way to save fuel (if proven useful by field-testing).
One of the reasons that thermal efficiency tends to top out around 50% is that actual temperatures inside pot skirts seem to be lower than Gemini suggests.
Channel Gap Temperatures in Four Rocket Stoves with Skirts
High Power
Medium Power
Low Power
Stove One
320C
240C
190C
Stove Two
330C
230C
150C
Stove Three
290C
220C
160C
Stove Four
335C
260C
180C
As David Evitt says: “Every Day Less Wrong!”
https://aprovecho.org/wp-content/uploads/2026/03/michael_saul.jpg1029772Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-03-27 14:17:202026-03-27 15:02:11Real World Temperatures in a Pot Skirt
We are currently studying products that help households in rural developing communities to meet their basic needs. Our goal is to develop a catalog of the most affordable, durable, usable, functional products available that have been rigorously user-tested and optimized. To help us choose where to start, could you please take 5 minutes to let us know what sort of products would be most needed in the communities that are experiencing energy poverty where you work? Please forward to your colleagues who work closely with these communities.
https://aprovecho.org/wp-content/uploads/2026/03/ARC-Logo-1.jpg10001000Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-03-13 12:43:412026-03-13 12:54:31Requesting input: What technologies are most needed in households?
For example, a new Bosch prototype cook stove with Super Pot. Thanks, Osprey Foundation!
HTE Design principles: Increase temperature and velocity of gases, exposed area in pot(s), radiation, proximity of gases to pot(s) without decreasing velocity. Use dry wood. Doubling temperature, velocity and area doubles heat transfer efficiency! Doubling radiation is much more effective!
Wood Moisture Content: This is often a critical variable. Ideally, wood should have a moisture content of less than 20%. Water in the wood must be evaporated before the wood can burn, consuming energy that could go into the food. Burning wood with 30% moisture content can reduce effective heat output by nearly 40% compared to dry wood.
Excess Air Ratio: Too little air into the combustion chamber causes smoke and incomplete combustion. Too much air also cools the gases before they hit the pot, decreasing how much energy enters the food. In natural draft cook stoves, velocity is usually something like one meter per second which is SLOW.
Design: Influences how much of the heat is successfully captured or lost. Dr. Larry Winiarski suggested maintaining constant cross sectional area throughout when designing a stove. A gap of 6mm to 8mm seems to work well in a pot skirt. The narrow channel forces hot gases to “scrub” against the pot surface, thinning the insulating boundary layer of still air.
Materials: High-mass stoves, often used for an hour or so, absorb a significant amount of heat. Using lightweight, insulating materials ensures heat is reflected back toward the pot rather than being “stolen” by the stove.
Bigger Pots: For highest thermal efficiency 1.) Expose as hot as possible gases at 2.) Fastest natural draft velocity 3.) As close as possible to the bottom and sides of 4.) The biggest possible pot. ARC now uses the constant cross sectional area of the stove reduced by 25% to calculate the gap for a pot skirt. We then fine tune prototypes under the emissions hood trying to find a desired compromise between thermal efficiency and emissions of CO and PM2.5.
Increasing HTE
Variable
Target
Effect on HTE
Wood Moisture
< 20%
Increases by reducing energy wasted
Channel GapIn Pot Skirt
6mm – 8mm
Increases (~25%)
Insulation
High
Increases by reducing losses
Pot Lid
Keep humidity above water at 100%
Increases by decreasing heat loss via evaporation
Excess Air
Keep Temperatures Hot
Increases when not more air than needed is supplied
https://aprovecho.org/wp-content/uploads/2026/03/image-e1772837736537.jpeg16001200Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-03-06 14:59:092026-03-27 14:11:49 Improving Heat Transfer Efficiency (HTE)
The other day, I watched as Dr. Winiarski’s stainless steel stove top (sold by BURN and SSM) helped to force 43% of the heat from a hot fire into a 30cm in diameter flat bottom pot without a pot skirt.
The improved stove top adds a lot to a stove! It is probably the most cost effective way to start improving a stove. What do we think it does?
Maintaining ~0.75 of constant cross sectional area in the stove top may help to thin the boundary layer of still air next to the bottom of the pot so hot molecules in the gases can replace cold molecules close to the bottom of the pot more effectively.
The restricted flow may help to maintain a beneficial air/fuel ratio (elevating temperatures) by decreasing the excess flow of cold air into the combustion chamber.
Evolving heat transfer “rules of thumb”:
Raising the temperature of the gases will increase efficiency.
Moving hot gases closer to the boundary layer will increase thermal efficiency until gas velocity is slowed.
Increasing exposed surface area will increase thermal efficiency until gas temperature reaches the temperature of the water in the pot, for example.
Increasing radiation will improve efficiency.
Increasing the velocity of the gases will also increase thermal efficiency, making sure that gas temperatures are not reduced by excess velocity.
https://aprovecho.org/wp-content/uploads/2026/02/30-cm-pot-w-winiarski-top.jpg6821200Kim Stillhttps://aprovecho.org/wp-content/uploads/2015/11/Aprovecho-Logo.pngKim Still2026-02-27 15:36:452026-02-27 15:49:20The Stainless Steel Winiarski Stove Top
Auto damper illustration from “Clean Burning Biomass Cookstoves”
Although airtight biomass heating stoves in the USA on average emit a lot more smoke than Southern Hemisphere cook stoves per unit of time, legally mandated chimneys move it outside where it is diluted enough to generally meet WHO standards. That’s great! Anyone who uses a wood burning heating stove knows that a chimney is necessary as a first step in the attempt to protect indoor air quality. Then, combustion efficiency must be high when density of wood stove use approaches urban levels, external air becomes stagnated, etc.
After decades of trying, research shows how hard it is to protect health with biomass cook stoves. It is not hard to design a cookstove with a chimney that achieves ~50% thermal efficiency when sufficient pot surface area is exposed to hot gas flow, but more than a chimney is needed for best protection.
I wonder if a vented airtight stove with sunken pot or pots (with or without griddle) would be “vale la pena,” worth the hassle? Dr. Kirk Smith spent ten years trying to protect health in Guatemala by removing smoke from houses with chimneys added to plancha cookstoves. The cooking pot would sit over a hole in the plancha for best heat transfer, but when the pot was removed from the stove smoke would pour into the kitchen through the open hole still harming health. He might say that an automatic damper must close when the pot is removed to keep smoke out of the room.
