The idea is pretty simple. Made a Open Ventilator System with KiCAD based in Medtronic PB560.
The 3D makers communities could help a lot, they are helping a lot with other things.
I am also collaborating in some.
I have seen the project a little and it is not complicated, but it is somewhat laborious and a little obsolete, it would be necessary to change some pieces for other modern ones.
My idea is to make an open project in which anyone can participate.
If you are interested in joining a KiCad-related effort, I’ve started organizing a group of engineers here:
So far 3 of the 6 boards are claimed for initial design. You could either claim one of the others and contribute the KiCad translations or join one of the verification teams.
Ok, I can contribute in Spanish and Kicad as much as I can. I can comment this to the OpenPnP community in case there is anyone who wants to collaborate with the PCBA.
Right now, the CPU, blower and buzzer are claimed.
Yes, as long as they are commercially available. EoL parts are replaced with equivalent and documented in the schematic.
Join the project and send me a message.
Nightly builds.
Its entirely possible that someone will die, blame the ventilator and sue. If that’s what you are worried about right now then this is not the project for you. If your name is attached to the work, then you are potentially a target of lawsuit, legitimate or not, in any of the hundreds of countries in which this might be used. We are adding disclaimers to finished work products. I’ve reached out to a couple of lawyers I know to see if they can put me in touch with someone who wants to write pro bono legalese for this.
I have no intention of working on this project myself, but do wonder:
I’m still not sure if the intention of releasing this stuff is a genuine effort to help, or a publicity stunt from that company. Opionions are divided, but I do not read them much.
It has of course also been mentioned on EEVblog and Hackaday.
As re-creating this 10 year old thing is likely to be partly based on reverse engineering. some general calls on mentioned sites to post some detailed photographs of the internals of this thing seem like a good idea to speed up the effort.
Detailed photographs of mechanical parts such as pump, valves and filters, but also from the PCB(s) for component layout and general routing. There must be some of these things that are faulty and usable for this.
Asking on Hackaday / EEVblog / etc for such pictures will help the re(verse)-engineering effort, and also make more people aware of this effort. After all, the project is useless if nobody sees it on git-hub/lab.
Are you ready for pull requests for such pictures?
I just saw one on TV that used the handheld version of the ventilator that is used by paramedics. It used a simple motor driven cam to depress the bulb to replace the hand squeeze. I would imagine you could adjust the volume by moving it up and down. An adjustable pulley for rate? It seems local machine shops could crank these out pretty quickly.
Simple variations of “press on a thing to squeeze air into lungs at > 1atm pressure” are plenty available. Actual medical ventilator is much more complex, just read the requirements doc Medtronic published.
It looks like the PCBs were designed in Zuken CR5000. The file formats are test (s-expressions). I could try writing a plugin for pcbnew if someone here would be interested to test it and, if the quality is good enough, manufacture the PCBs.
Tom, we need to substitute components so this won’t be straight replacement although, if you’re interested, I have some code that reads CR-5000 (not KiCad yet)
Do you mean the CPU/display? These seem to be quite old/exotic. The software is ugly, but I think it can be ported to any relatively modern architecture in reasonable time. Also the electronics could be probably greatly simplified…
My wife was in and out of ICU several times over a three month period at the beginning of the year. She spent lots of time on a ventilator. Believe me, I know what they are. I asked lots of questions. What I’m talking about could be in hospitals in days if needed. This is literally next step up from getting volunteers to sit around squeezing the bulb. Sourcing a gear motor would probably take the longest for me.
But, thinking out loud. Those basics form a good fall back fail safe in case the electronics fail.
What you think if the PIC 28Mhz is replaced for a Wireless 48Mhz Cortex M0+ ?
So all ventilators can be monitored by a control room.
I’m using the NXP MKW41Z512VHT4 in other design, I have lib and footprint ready, isn’t expensive. It’s modern, low consume, in production and has Wifi and BLE , and support Thread protocol, Zigbee and others.
Is 20dbm but I have other design with PA with >500mt of range.
I wouldn’t think much about the mechanical parts now. Just focus on the pcb. There are a lot of people out there who are doing amazing designs for all kinds of machines, 3D printers in PLA, ABS, etc. FFF/FDM SLA/MSLA CNC Laser…
I think they have left it for health approval I was seeing it yesterday in the group of coronavirus makers. I’m printing visors for doctors and police in coronavirus makers.
I thought about brushless motors, I think that with one and a turbine generates enough air, They are quiet and durable. You have to find engines that can work more than 40 days without stopping. Turbine printed with dental resin in FDM or SLA could be pass the medics test I think. Or in metal with CNC or Laser.
Brushless motors need a ESC, I have some designs with fortior and others chips that are very used in fpv drones.
This could reduce and simplify the mechanical part of the respirator.
I think that respirator need a RTOS, this is mission critical, and if the
pressure sensor detects that there is no pressure immediately activates the other motor, with a support that facilitates the extraction of the motor in case it has to be replaced.
Another thing I’ve been thinking about is having the ability to add a UV filter. I have been reading that this improves air quality. It also has another function which is to disinfect. UV light with a color between 200nm and 300nm can kill viruses.