Sprint Day 2: What developing world substitutions are there for flow meters?

[Enginerding]

Sprint Day 2 has begun;  here's the critical path question:



What component substitutions are available to the developing world for flow meters?

This could come from off-the-shelf commercial and household items.  E.g. automotive Mass Air Flow sensor (bonus: ~$50).

This is a dialogue oriented question, the more community the better.  Spitball, brainstorm -- let your hair down.  If a few options emerge, I'll start a poll.

Situation:

Ventilators need flow meters that operate at 0-80 lpm.  One common item in use is the Sensiron FM3200.  (22mm OD input/output).  However, there are now less than 100 of these in the supply-chain, according to Octopart.  Additionally they are expensive at ~$170/each.

======

Housekeeping:

Analysis of Open Source COVID-19 Pandemic Ventilator Projects by R. Read, et al at Public Invention. (shared by Dave)
spreadsheet: https://docs.google.com/spreadsheets/d/e/2PACX-1vTYAfldxoIiO46VAWH1NlhrwFBn9mguqS2bh1spnLEu4AVVN1cj1vaEm6vOp5Z6UnaAbUwd8dslCXdM/pubhtml
homepage: https://www.pubinv.org/project/ventilator-verification-project/
Git: https://github.com/PubInv/covid19-vent-list
Git for Inline Ventilator Test Fixture and Monitor: https://github.com/PubInv/ventmon-ventilator-inline-test-monitor
Git for A collection of validation test for for emergency covid19 ventilators (clinical usability and suitability as per RMVS): https://github.com/PubInv/covid19-ventilator-validation-tests
Forum (for Public Inv, not project specific): https://groups.google.com/forum/#!topic/pubinv/

This is the principal resource for determining project progress.  The author (Read and team), have attempted to quantify each project's progress.   I'm a participant in this project, so feel free to make suggestions.

Lastly, a few rules that I ask each individual to abide by for success.

- This sprint format isn't for everyone.  It's fast-paced and goal-oriented.
- Please consider using your real first name.  So...I'm Dan, nice to meet you!
- Before you post, ask yourself how contributory what you're about to post is to the current sprint.  Could it be better?  If it's obvious, it's probably already been thought of.  If you're unsure, DM me and I'll help. 
- If you have any great ideas that have to do with the future sprints you think we need to prepared for, also please DM. 
- This goes without saying but, no politics.  We can all scroll our social media feeds and get enough of that.
- Success will probably include a project t-shirt.  That's it.  Oh and you get a free ventilator if you build one.  So there's that.



Previous Day Sprint archives:

Sprint Day 1
====
Hey, all!

The previous day's Sprint was a smashing success.  There were many wonderful, diverse ideas.  Dave even stopped in to share his wisdom.  I've archived that thread, and it'll be up on the project Github shortly.  Here: https://github.com/EnginerdingDan/Open-Source-Ventilator

Thus, Sprint Day 1 begins.

The next critical path item is:

Does a fully operational open source ventilator already exist w/ a suitable BOM for the developing world?

At the bottom of this thread, I'll catalog, for reference, existing projects.  Be forewarned, most of the "open source" projects touted in media *will be hype*. So I ask that if you post a project; DON'T post a media link.  Run it down, and post the actual link to the project's mainpage, for example their Github.   Projects that DON'T have a locatable homepage can safely be assumed to be wank.  (I'll accumulate media-only links at some point).  If they don't have a specifications document, a BOM, a working prototype -- wank!

IF, you have other succinct criteria that will define the critical path question, please post it in BOLD.  And I'll update this main post accordingly.

Critical Path Criteria (not to be taken as an exhaustive list):

Truly open-source? (for example, are all the project files open and available)
Working prototype?
Design files including BOM?

=============

Known Open-Source Projects:

Motivational Resource: Open Source COVID19 Medical Supplies (in conjunction with Hackaday) <--- 1000s of successful PPE projects being delivered to hospitals.
https://www.facebook.com/groups/opensourcecovid19medicalsupplies/

Mechanical Designs that Adapt a Bag-Valve-Mask: (BVM: https://en.wikipedia.org/wiki/Bag_valve_mask)

MIT E-Vent (shared by floobydust), "Emergency ventilator design toolbox", homepage: https://e-vent.mit.edu/  Behind a login.  Registration is instant though.  The specifications documents are detailed and from a medical providers point-of-view, similar to the UF specifications but less detailed.  The electronics work is very basic, enough to essentially "bag" a patient on a timer.  Uses aluminium extrusions.

OpenBreath (shared by awallin) This Italian project has a notice from March 22d that it's now a 3D printed project due to the economic shutdown in that country. 
homepage: https://www.openbreath.it/en/
Git:  https://www.ohwr.org/project/openbreath

Motor-driven Designs:

 
University of Florida, Medical School, "Open Source Ventilator Project"
homepage: https://simulation.health.ufl.edu/technology-development/open-source-ventilator-project/
Engineering Specifications:  This is produced by the Medical School, led by Anesthesiology, so the specifications are from the clinical point-of-view of the feedback/control a physician is looking for.  They're well detailed, and high-quality. here: https://simulation.health.ufl.edu/technology-development/open-source-ventilator-project/engineering-specifications/
BOM:  Incomplete.  Mostly "home depot" type stuff.  No electronics other than specifying the BMP280 (Phillips) and an NXP pressure transducer.  here:  https://simulation.health.ufl.edu/technology-development/open-source-ventilator-project/bill-of-materials/
GitHub:  sparse.  here: https://github.com/CSSALTlab/Open_Source_Ventilator
Forum:  sparser.  here: https://simulation.health.ufl.edu/forums/forum/open-source-ventilator-project/

Bellows Design:

VentilAid (shared by Dave). (info soon) https://www.facebook.com/groups/ventilaid.community/


Not Open Source:

Medtronics (info soon, not true open source, seems like wank, "eevBLAB #72").  They say it includes schematics and manufacturing docs  https://www.medtronic.com/us-en/e/open-files.html

Mercedes F1 "Project Pitlane" (cannot find design files or homepage, just media)

uncharacterized:
https://twitter.com/VentilatorU
https://twitter.com/Selcuk/status/1244661929777532930

[coromonadalix]

https://techcrunch.com/2020/03/30/medtronic-is-sharing-its-portable-ventilator-design-specifications-and-code-for-free-to-all/

??

sorry already posted

[awallin]

this project was posted on ohwr forums maybe a week ago: https://www.ohwr.org/project/openbreath/wikis/home

edit: also check links on this page https://opensource.com/article/20/3/open-hardware-covid19

[Enginerding]

this project was posted on ohwr forums maybe a week ago: https://www.ohwr.org/project/openbreath/wikis/home

That's a new one!  Do you think you can run down some of the critical criteria listed above?  And I'll add it to the list.

Thanks awallin!  You're the man!

[MK14]

Making daily, new threads like this, doesn't seem to be a good idea.
This is a forum, NOT a personal BLOG.
The various people, "subscribed" to the previous thread will disappear. Ideally, you wanted them to still be in the thread, as they would have been notified and/or could be informed via the "Show new replies to your posts." link, of further progress in the thread.
Also, forum users, sometimes need or want to see what has already been mentioned/discussed.

If you continue to make new threads like this, you may find it rather counter-productive, and see many forum users lose interest in the thread(s).

[magic]

Also, forum users, sometimes need or want to see what has already been mentioned/discussed.

That's not agile. What happened in the past stays in the past. Today we concentrate on the task at hand

many forum users lose interest in the thread(s).

Never

[Enginerding]

MK14, don't consider a sprint "day" definitively to be one 24-hour time period.  "A task isn't done, until it's done", as the saying goes.

The best piece of collaboration would be to identify a promising open-source project OR flesh-out details for one of the projects listed above.

Particularly, whether a project has a BOM is going to be a good indicator of its viability.

Much appreciated,  Dan.

[EEVblog]

Making daily, new threads like this, doesn't seem to be a good idea.
This is a forum, NOT a personal BLOG.

Correct, we won't tolerate the projects section being filled with a new thread daily.

If you want a comprehensive list of venitlator projects and their status, some people have done the hard work for you:

https://docs.google.com/spreadsheets/d/e/2PACX-1vTYAfldxoIiO46VAWH1NlhrwFBn9mguqS2bh1spnLEu4AVVN1cj1vaEm6vOp5Z6UnaAbUwd8dslCXdM/pubhtml#

[Enginerding]

That's a killer list Dave!  Who sent it to you?  Brilliant. Added above.

My intuition was indeed correct; none of the projects are very far along as verified by this 3rd party (R. Read, et al).  For instance, the UoF project is ranked in the top 5, but only a 2.9 of 5.  Ventilaid is a nice project also, but most of these projects will start hitting the usual project decay bottlenecks (I'd guess 1 standard deviation, if not 2).

The biggest problem they're going to have in the developed world is getting the BOM together; it may be critical to identify commercial and household items as BOM substitutes.

(Also, it's quite easy enough to modify this thread for new sprints.  Doesn't really make a difference to me.  Thanks for the input.)

[Enginerding]

Day 1 Sprint Complete!  ✔️

Absolute success.

The next sprint will begin shortly.

[EEVblog]

This Aussie one looks great!
https://twitter.com/tripleeightaus/status/1244767611365126145

[Enginerding]

Bump for new sprint day.

[senso]

Dont edit the first post like that, it ends being confusing AF..
Grab the smallest MAF from a car, but remember any name brand like Bosch will run you down 180-250€, and those cheapo crappy ones that cost 50-80€ fail in a couple months and report erroneous values..

[Enginerding]

What are "good" OEM brands in automotive?  For small cars that would be popular in the developing world? (e.g. Japanese, SK).

Do you think you could find an example of a MAF sensor you mentioned?

Thanks, you're the man.

Dont edit the first post like that, it ends being confusing AF..
Grab the smallest MAF from a car, but remember any name brand like Bosch will run you down 180-250€, and those cheapo crappy ones that cost 50-80€ fail in a couple months and report erroneous values..

[AndyC_772]

This Aussie one looks great!
https://twitter.com/tripleeightaus/status/1244767611365126145

Cool, I like that. Here in the UK the Mercedes F1 team have done one too.

https://www.formula1.com/en/latest/article.mercedes-f1-help-develop-coronavirus-breathing-device-for-health-service-use.6Yx9lt3ADUMLwXr7UKDfYT.html

[ebastler]

Particularly, whether a project has a BOM is going to be a good indicator of its viability.

What kind of criterion is that?? Where does your initiative stand on the "viability" scale according to your own benchmark? 
Releasing a BOM is one of the last steps once your design is complete, verified and validated.

"All unfinished projects are worthless, so let's start another one!"

[TheHolyHorse]

Particularly, whether a project has a BOM is going to be a good indicator of its viability.

[...]

"All unfinished projects are worthless, so let's start another one!"

Yeah lets start more projects that will never be finished

[Simon]

If you are buying serious quantities from Bosch MAF meters are like £50, the smallest ones do 0-3l/m. attached is a datasheet of two models that are obsolete but will give you an idea. The other method of measuring flow is to create a small restriction (which by definition any length of hose is) and measure the pressure drop across it but that will need fully characterizing and ideally you want a sensor with the right output curve.

Yes if you create too many of these threads you will just loose the people you thought were on board (laughing at you more like), eventually piss everyone off and get banned.

[Simon]

Particularly, whether a project has a BOM is going to be a good indicator of its viability.

What kind of criterion is that?? Where does your initiative stand on the "viability" scale according to your own benchmark? 
Releasing a BOM is one of the last steps once your design is complete, verified and validated.

"All unfinished projects are worthless, so let's start another one!"

Because in a simpletons mind if he has a list of parts that's it, just attach them all together to the drawings and hey presto. Remember this guy is a medic, not an engineer so he hasn't a flipping clue what he is doing.

[EEVblog]

And it was smart of him not to mention the open Medtronics specs, I'm sure that's very deliberate.

[ArthurDent]

Dave, That video was a great overview that explains in layperson terms what would actually be required to build a ventilator that actually works. There is also the short-term reliability, life expectancy of the unit, and the whole legal side to consider. Bottom line is this isn't a sprint of days but a long-term expensive project well beyond my, or most people's abilities, and wallet.

[EEVblog]

Dave, That video was a great overview that explains in layperson terms what would actually be required to build a ventilator that actually works. There is also the short-term reliability, life expectancy of the unit, and the whole legal side to consider. Bottom line is this isn't a sprint of days but a long-term expensive project well beyond my, or most people's abilities, and wallet.

Yes, I think almost all of these hack rapid development projects, as well intentioned as they are, will not amount to a usable product.

[thatguy1988]

Adafruit sell some very cheap flow meters. Amazon do as well but they don't ship them to all countries.

https://www.adafruit.com/product/833

https://www.adafruit.com/product/828

[Simon]

Those are both liquid flow meters. If they are useful your patient is already dead!

[Simon]

The easiest thing to "make", "cheap" is a slight restriction in your tube with a very sensitive differential pressure sensor either side that has a log output. This counteracts the exponential characteristic of the restriction that follows a square law to give a linear output to flow. It would need characterizing. You can get a 125 Pa sensor for 10's of pounds.