Where SustainX (and Vionx, Malta, Boston Metal, Phoenix Tailings) was a typical venture capital-backed, university technology spin-out, Building Envelope Materials – BEM – was just Doug (the owner) and myself at first. After SustainX, which I saw as a technical success but fell flat on bringing any product to market and lived & died by the venture capital model, I wanted to join a company that was much closer to bringing a product to market and was not in the venture capital system.
BEM is developing the techniques and equipment to achieve deep energy retrofits of buildings without the typical expensive & difficult hassles (tear off all the siding, add insulation boards, reset the windows, re-side the whole building). BEM is achieving this by injecting polyurethane foam directly into existing wall cavities, greatly improving R-value and air sealing.
R&D – Research & Development Engineering
In-Situ Proportioner (ISP)
A key to making good polyurethane foam is to get the proportions of its two parts just right. Incumbent solutions for maintaining the right proportions are basically at the extreme ends of the spectrum: ‘Kit Foam’ for a couple hundred bucks, zero proportion control, and generates a lot of waste (tanks & residual chemicals are thrown out when the kit is used up), or a spray foam trailer for around six figures, very good (but not perfect) proportion control, and a lot of maintenance for the equipment. We wanted to develop a product between those extremes: a proportioning system to use with reusable chemical tanks but without the expensive & complicated pumps and controls of spray foam trailers.
Hydraulic Pump / Motor
For the first prototype, I adapted a pair of small hydraulic (positive displacement) pumps to meter & proportion the chemicals flowing through them. Chemical A (the hardener) is less viscous than chemical B (the resin) thus the A side flows faster so we take some of that extra energy to help spin the shaft of the B side and keep the chemicals in lock-step. Under a head of air pressure in repurposed propane tanks, we used vegetable oil and castor oil to represent the chemicals A and B respectively as they have similar viscosity.
The trouble with this design was the hydraulic pumps shaft sealing is so tight (as they are designed to contain 1000s of psi), it takes a lot of energy to turn so the shaft does not spin fast at all – see the video above. But it was a useful proof-of-concept that gave us confidence to pursue the idea further, and the initial evidence needed to propose the development of this tech as as a project to get funding from the MassCEC (Massachusetts Clean Energy Center).
We also investigated other other kinds of hydraulic machines, such as this gerotor pump, which is a kind of internal gear pump where one gear is inside the other. Sometimes called a Trochoid pump, a brand name, short for trochoidal which describes the geometry of the inner rotor.
Magnetically coupled pumps
With the aim of reducing the shaft sealing friction, we went to a contactless technology: magnetically coupled pumps.
Kit Foam produces a lot of waste: that stack of boxes+tanks+residual chemicals ends up in the dumpster
Since these pumps are positive displacement, they can act like flowmeters where every revolution measures out a fixed volume of fluid. I used & programmed an Arduino and a magnetic pickup to count shaft revolutions to get that flowmeter functionality from the existing hardware.
From the stainless steel pumps, we learned that alignment of the pumps is critical as we found the flex coupling connecting the two was too lossy for our application. We shifted to mounting pumps to a gutted motor case.
One of the fundamental challenges with polyurethane equipment is that Chemical A, the isocynanate hardener (essentially Gorilla Glue), is moisture-sensitive and hardens after contact with moisture or water. We found out pretty quickly that magnetically coupled pumps – which are typically chosen specifically for their hermetic design – were not immune the permeation of moisture across their rubber o-rings so over time the A-side pump would seize up. Water (vapor) permeates through all polymers, plastics, and rubbers – only metal and glass are sufficient barriers to moisture permeation.
From the moisture issue and a number of other challenges, we found the ISP was not practical enough, and the effort/time/money to get it to the next level would be better spent on other projects. However, we kept the Arduino metering system and used it in the next design, the mobile foam system.
Mobile Foam System
Instead of using gear pumps to meter flow, I used an oval gear flowmeter for each chemical, and tied them to the Arduino system.
The Tower – mobile foam system, first evolution
The Tower shook apart when going up and down stairs, which was a nonstarter as traversing stairs/steps/curbs is a fundamental part of bringing the system to its point of use. But worked well in the lab!
For foam injection, we developed both the equipment and techniques (they go hand-in-hand at this stage of development), and the needle injection method (not pictured) is now patented:
The Trunk evolution
Packed the electronics and the flowmeter into this trunk, a mobile job box.
As you can see with the different evolutions, we were constantly trying to find smaller, more nimble solutions as rooms and apartments are so tight.
The Cart evolution
Elevators are necessary for the 4 tank cart as each tank is about 100 lbs when full.
Cathedral ceilings and Enclosed Roof Cavities
We won a DOE project for adapting the BEM tech to address cathedral ceilings and other roof cavities: Validation Study of Experimental Insulating and Air-Sealing Technology for Enclosed Roof Cavities
Voinx x BEM crossover
Prior to BEM, I had done some thermal modeling of Vionx’s vanadium flow battery system. For this project, they wanted to insulate & support their electrolyte tanks so BEM demonstrated its tech in their lab, then we actually foamed tanks in-situ on location in Shirley, MA.
