OT - how does a ventilator work?

[Glug]

Could one not be made with a linear actuator squashing a bellows.

Why get so fancy? Don't you have footballs?

Where there's a Gilligan, there's a way.

[boslab]

Along the lines of what blackforrest wrote, I made an analogue of one in the lab, we were testing something, it was basically a 6” pvc tube with a cap each end, inside was a pvc piston that was weighted, compressed air pushed the piston up until the top of the stroke was picked up by a proximity switch, the cylinder was then exhausted through a reg, when it got to the Bottom another proximity switch was triggered, the position of the switches controlled the swept volume, the whole lot was run on an old Siemens S10 plc, I replaced that with a mitsibushi fx and a HMI panel, there were no motors just air pressure, and return by weight, it was quite accurate as well, parts easy to sterilise, simple seems to be the way, loads of motors, bellows etc just make it unreliable, I was accused of building Dr who’s tardis as that’s what it looked like,
mark

[Glug]

it was basically a 6” pvc tube with a cap each end, inside was a pvc piston that was weighted, compressed air pushed the piston up until the top of the stroke was picked up by a proximity switch

Hmm.. I eagerly await the first steam powered ventilator on HSM. Or maybe the "Rupnow hit and miss ventilator"..

[Commander_Chaos]

Looks like there's any number of effective ways to make the pump. The issue, I'm told, is regulating it. Not volume, although that's important, but how often it cycles. Constant pumps are used but only when the patient is paralyzed. Otherwise, the pump needs to sense when the patient tries to breathe and the pump assists.

[alanganes]

So far as actual specs go, here is an actual specification for one issued by the UKK government for a rapidly developed unit for potential use in the current situation. There is a lot more to it than blowing air down a tube it seems:

https://www.gov.uk/government/public...-specification

For a more layman's version of a similar effort happening here in the US:

https://hackaday.com/2020/03/23/mit-...r-fda-testing/

The second link includes links to other similar efforts going on elsewhere.

Much of the terminology is beyond me, but it seems that something built around the existing manual bag type things makes a lot of sense. They already exist and are made specifically for the task, setrile, etc. As they are intended to be single use, I'd expect that hospitals have lots of them already, at least for a start.

I agree wit gellfex, I'd think designing around CNC capabilities makes way more sense than 3d printing for stuff like this. There are scads of shops with CNC capability that can crank out appropriately designed parts by the boatload and could start doing so tomorrow if the info and materials are available. There are far more of them and people who are skilled at using them than there are 3d printers, I'm sure

[Mcostello]

http://www.apneaboard.com/forums/Thr...9666#pid339666. This was sent to Me and I would like to pass it along if My feeble attempts at posting a link actually work.

[Glug]

https://www.gov.uk/government/publications/coronavirus-covid-19-ventilator-supply-specification/rapidly-manufactured-ventilator-system-specification

This is a specification of the minimally (and some preferred options) clinically acceptable ventilator to be used in UK hospitals during the current SARS-CoV2 outbreak. It sets out the clinical requirements based on the consensus of what is ‘minimally acceptable’ performance in the opinion of the anaesthesia and intensive care medicine professionals and medical device regulators.

[gellfex]

Along the lines of what blackforrest wrote, I made an analogue of one in the lab, we were testing something, it was basically a 6” pvc tube with a cap each end, inside was a pvc piston that was weighted, compressed air pushed the piston up until the top of the stroke was picked up by a proximity switch, the cylinder was then exhausted through a reg, when it got to the Bottom another proximity switch was triggered, the position of the switches controlled the swept volume, the whole lot was run on an old Siemens S10 plc, I replaced that with a mitsibushi fx and a HMI panel, there were no motors just air pressure, and return by weight, it was quite accurate as well, parts easy to sterilise, simple seems to be the way, loads of motors, bellows etc just make it unreliable, I was accused of building Dr who’s tardis as that’s what it looked like,
mark

I've made several similar items in my special effects career, notably a bellows using a DC motor, linear track, and heavy duty PVC/cloth spiral ducting. I think we all agree the real challenge is the controls, and standardising them. I would also think it's possible a semi-standard mechanical design could be developed from hacking inkjet printers, which have a very usable linear transport. But it might be too light duty, I have no idea of the actual mechanical power needed at this point.

McMaster has an endless variety of these ducts that can be easily made into a bellows with less issues than a piston system. I like the idea of a shaft from a cam arm to a bellows for smooth ramping at each end, rather than jerky linear compression. By adjusting the joint distance from the motor shaft the volume can be adjusted. But if volume can be adjusted in software on a linear track system, I guess that's moot. There's also a long history of double action forge bellows, that pump on stroke and reset.

[goodscrap]

This was on the local news last night. Uses mostly off the shelf medical items.

[gellfex]

This was on the local news last night. Uses mostly off the shelf medical items.

Clever to use the O2 supply for power, but that syringe piston ain't gonna last without lubrication. A real air actuator might be in order. And what about the volume-frequency controls?

[alanganes]

If anyone is interested in a bit more in-depth but still layman's terms explanation of how these devices work, here is another rundown. A bit longer read, but like many things there is a lot more to this stuff than meets the eye:

https://hackaday.com/2020/03/25/ventilators-101-what-they-do-and-how-they-work/#more-403759

[Hawkeye]

Here’s a different type of ventilator, developed in 165 by the US Army.

https://www.army.mil/article/234507/redditors_revive_interest_in_1960s_army_emergency_ventilator_invention

[Black Forest]

Breaking News! This is a Public Service Announcement. All Accordion and Bagpipe musicians are to report to their local hospitals for ventilator duty. Kilts optional.