Version 0 of the M.U.R controller. © Quentin Chevrier
Since March, a citizen collective of designers, doctors, humanitarian workers and scientists has been working to produce an emergency medical ventilator that is open source, low-cost and easy to reproduce. Makery met them at several stages of the lockdown in Paris.
In late March, Makery received a message from the thr34d5.org medialab asking if we were going to do anything with the servomotors stocked in our office, which had been closed for the past two weeks. Servomotors, they said, were urgently needed for testing in a project to create a low-cost, open source ventilator. We got in touch with the project leader Antoine Berr, and a few days later hand-delivered a few servos to the team behind the Minimal Universal Respirator (M.U.R).
“Watching for critical degradation”
We wanted to know more, but the designers were buried in their work. On our own, we discovered that Antoine Berr had studied human-machine interfaces before adventuring around the world with his La Cool Co project and finally founded Club Sandwich Studio in Paris in 2019 with Paul Couderc, Arthur Siau and Simon Juif. Antoine redirects our media enquiries to Solène Touzeau. Both she and Antoine are active members of Objectif Sciences International, an NGO for participatory science and popular eduation that supports M.U.R.
She introduces us to Jean Karinthi from Hermitage, a large space for rural and citizen innovations situated on farmland north of Paris, which also supports the project. ACTED, another big humanitarian NGO, has also joined the M.U.R collective. She hopes that eventually, “M.U.R ventilators could be developed on ACTED’s intervention sites, such as their 3Zero House in Manilla,” a French-Filipino incubator of local collaborative and solidary projects. The name refers to zero exclusion, zero carbon, zero poverty.
Karinthi adds, “ACTED is well established in a lot of countries in crisis. They manage refugee camps and estimate that about 100 million people could potentially benefit from M.U.R in countries that have zero ventilators. In refugee camps in Jordan, there simply are no artificial repirators. In Mali, there is only one ventilator for the entire country. So Antoine wanted this project to be non-profit, open source, low-tech, low-cost and universally accessible. Because we know that this health crisis extends well beyond our national and regional borders, and that it could do a lot more damage in developing countries.”
“Paris Hospitals Association (AP-HP) has already welcomed M.U.R for testing,” Karinthi continues. “They have been following all the initiatives very closely, whether from companies and consortiums (Renault, Seb, Air Liquide…) or mobilized citizens. They are watching for critical degradation.”
Meanwhile, the M.U.R team is also testing and documenting in real-time as they work. “In the lab, ventilators are operating non-stop for several days at a time. Yesterday, one of them had been running for more than 48 hours straight without any anomaly, with a consistently stable display,” says Solène. “Everything is recorded on video, so that we can confidently say that our machines are working and they’re solid. We want to send out the message that yes, they’re made by makers, yes, it’s a citizen project, but behind all that, this project is reliable. Now we’re even supported by the National Centre for Space Studies (CNES).”
One month later
Antoine humbly admits that what they’re doing is “nothing more than what we’ve always done: make stuff, show it, share it”. He says that his experience with La Cool Co installing IoT sensors in cacao fields in Ivory Coast, with no stable Internet connection or electricity, somewhat inspired the concept behind M.U.R. “Our goal is to make a device that is extremely robust and reliable, all-terrain, that can operate in the bush.”
He gives a revealing example of the frugal engineering principles that guide their action: “We have an experience of jugaad, in Africa and elsewhere. For example, one of the problems with the ventilator is measuring the airflow, being able to extrapolate an injected volume. Because we want to stay low-tech, we chose not to use industrial sensors that are expensive and not always available. We came back to the idea of hacking an airflow sensor that you can find on almost any car. This device can be bought for 5 euros and produces incredibly precise and stable curves. And it has the advantage that you can find it in practically any region, because all cars have one.”
Since our first exchange in early April, a team of engineers from Grenoble has joined M.U.R to work on the proportional–integral–derivative (PID) controller, which automatically ensures that the device is operating as programmed. “The PID controller improves the performance of a servo system, which is a control loop mechanism,” Antoine explains. “This means that the device can better understand the patient’s reactions, if they’re coughing, if they’re trying to inhale, etc.”
A few weeks later
The M.U.R team has also been closely following Milano Ventilatore Meccanico (MVM), an Italian project initiated by physicist Cristiano Galbiati, who rallied a large number of nuclear physicists in Europe, even collaborating with scientists in Canada, under the supervision of Nobel-prizewinner Arthur B. McDonald, and the United States. “Researchers in dark matter designing a ventilator, that’s pretty extraordinary,” comments Antoine.
By this time, MVM has been cleared by the U.S. Food and Drug Administration’s criteria for Emergency Use Authorization to treat Covid-19 patients. Antoine specifies that MVM’s objectives are slightly different from M.U.R’s: “MVM is a response to a shortage of ventilators in hospital emergency rooms. Their device operates with a steady flow of oxygen supplied by the hospital. We’re also following MakAir, another start-up with industrial partners to deliver a turnkey project, in response to the shortage of equipment in France. M.U.R’s objective is to have a device that works even without a stable power source, where there is no Farnell or Lextronic that you can order your parts from, and even if a 3D printer is available, you should be able to build it using a turning-milling machine. We’re also working on an airtight case for easy transportation and deployment.”
When we ask Antoine how the project has evolved since the beginning, he says that after two months, their humanitarian remains intact, but now that they have a better understanding of the disease, they can better focus their medical objectives.
“Like many teams at the beginning of the crisis, including MakAir and MVM, we worked to meet the technical specifications for a medical-grade ventilator that were current on March 16. In the weeks that have passed since then, scientific understanding of Covid-19 has changed the diagnostics. So now we realize that it’s not just a respiratory disease, it also attacks the blood vessels, that intubation is not always the solution, that non-invasive ventilation (NIV) at the onset of the disease could be a good thing, etc.
It’s encouraging to see that MakAir and MVM got certified. But if we consider that this may only be the first wave or crisis, and that in some countries there is absolutely no equipment of this kind, very soon they will have to make a choice: either buy from foreign industries, or appropriate open source technologies and produce locally. We want to offer a device with the least amount of technical restraints for production, which requires much less technical training for medical staff, but which in all cases classifies as a ventilator with the capacity and functions for intensive care and anesthesia, while also having the potential to be downgraded for NIV, which corresponds to more than a third of the needs in respiratory assistance.”
More information on the Minimal Universal Respirator project