/ Simple emergency ventilator. Medics view please?

Please Register as a New User in order to reply to this topic.
Rog Wilko 20:17 Tue

Mrs Wilko's nephew Jon (who is a successful inventor/engineer) has made a prototype ventilator. It's on a Facebook page, text below:

"If you know of anyone involved in the decision making process for acquiring emergency ventilators please forward this message. I have constructed a demonstration unit based on a 10 year old research paper I found online. This ventilator needs no electric power, just a cylinder of oxygen at 4 bar and a disposable "Ambu-bag". It may well be too crude to provide benefit to patients, but if it doesn't get presented to the right people I'll never know. Hope you're all keeping safe."

I can send a photo on pdf to anyone on here who knows what to do with such an idea. Also, can put anyone in touch with Jon.

Report
marsbar 20:25 Tue
MG 20:28 Tue
wintertree 20:43 Tue
In reply to MG:

Interesting document.  The final section appears to be unfinished "todo" notes that slipped through document control.

> Battery backup. Every current ventilator used inside hospitals has a battery backup, so users will expect it to be there and will behave as if it is, for example, unplug it from the wall in order to rearrange cables or while manoeuvring the patient. However, this needs very careful thought to balance the risks. Including this in the spec means instantly trying to source 30,000 large, heavy batteries. Specifying a DC voltage (ie 12VDC) may well be the most sensible for the machine working voltage. Need the advice of an electronic engineer with military/resource limited experience before specifying anything here. It needs to be got right first time.

This suggests to me that the plan is to rapidly get 30,000 units built.  Eight "ExCeL" style field hospitals and 30,000 new ventilators and you could push the whole country through the pandemic in 10 months without contact tracing etc.  Today the Navy shipped 7 qPCR machines from Newcastle University off to a testing centre in Milton Keynes; I imagine more are getting shipped from elsewhere; lab suppliers set them up immediately.  It looks like the logistics beast is slowly rumbling to life and will be quite the response.

Post edited at 20:43
Report
MG 20:49 Tue
In reply to wintertree:

There is a serious effort I think. Colleagues involved with getting major manufacturers involved. (And well spotted)

Post edited at 20:49
Report
wintertree 20:52 Tue
In reply to MG:

It seems to me an over-constraint to specify the voltage of the system; different manufacturers will have different product lines with plentiful supplies of batteries, 22.2 V for Dyson for example.

Report
MG 20:56 Tue
In reply to wintertree:

Yes not clear why it is “sensible”. There are plenty of car batteries around however.

Report
mik82 21:09 Tue
In reply to Rog Wilko:

Good effort for trying and getting involved

I do think the concentration on ventilators is a bit of a red herring to distract from the fact the government has left the NHS woefully underprepared. This isn't like polio where people's respiratory muscles are paralysed and you take over their breathing until things get better. These cases are really unwell. They have a lot of damage to their lungs and require special strategies to ventilate effectively, needing lots of specially trained staff. 81% of ventilated patients in China died anyway. Those that do survive are likely to be chronically breathless.

Post edited at 21:11
Report
nikoid 21:30 Tue
In reply to mik82:

Yes there was a respiratory consultant on here the other day saying respirators aren't the issue, numbers of specialist staff is the limiting factor. He seemed credible to me.

Report
jkarran 23:23 Tue
In reply to wintertree:

> Interesting document.  The final section appears to be unfinished "todo" notes that slipped through document control.

> This suggests to me that the plan is to rapidly get 30,000 units built.  Eight "ExCeL" style field hospitals and 30,000 new ventilators and you could push the whole country through the pandemic in 10 months without contact tracing etc.  Today the Navy shipped 7 qPCR machines from Newcastle University off to a testing centre in Milton Keynes; I imagine more are getting shipped from elsewhere; lab suppliers set them up immediately.  It looks like the logistics beast is slowly rumbling to life and will be quite the response.

The UK supply chain constraint requiring every component be UK sourced in that spec is the killer. There's not 30k of anything useful for building that sort of machine in freely available UK stock, to comply will take a shit ton of money and jets to bring non UK stock to UK warehouses asap but everyone in the world is thinking the same right now, export controls will be a hurdle soon, flexibility will bey key to success.

Jk

Post edited at 23:24
Report
nikoid 08:34 Wed
In reply to nikoid:

> Yes there was a respiratory consultant on here the other day saying respirators aren't the issue, numbers of specialist staff is the limiting factor. He seemed credible to me.

I meant ventilators not respirators.

Report
neilh 10:19 Wed
In reply to nikoid:

They just do what happens in Italy. The medical protocol means that instead of 1 person managing a unit,that person is allowed to manage 5 or more.

Its easy to change the riules when you have to.

Report
wintertree 10:24 Wed
In reply to neilh:

> They just do what happens in Italy. The medical protocol means that instead of 1 person managing a unit,that person is allowed to manage 5 or more.

> Its easy to change the riules when you have to.

Yes; there is a certain grim economy of scale to having lots of patients with the same problem in the same place. There are documented instances from past pandemics of an anaesthesiologist remote supervising medical students undertaking manual ventilation of patients whilst under quarantine.  It’s very far from ideal but it’s better than just leaving people to die.  

Report
Deleted bagger 08:25 Thu
In reply to Rog Wilko:

Can't imagine these vents will be very high tech. In-out, variable pressure and rate. I expect they'll be similar to the ones I used when i started nursing. Still fingers crossed they'll knock them out in time.

Report
MG 08:32 Thu
In reply to Deleted bagger:

They've now been ordered from Dyson and are a completely new design.  They will probably be see-through with purple knobs that fall off and cost five times normal ones.

BAe are apparently about to start production of an existing design, which sounds more promising to me.

Report
neilh 08:50 Thu
In reply to MG:

I would guess they are tackling it in various ways. Ramping up exisitng production, and also looking at new sources such as Dyson and another.

They will not put all their eggs in one basket.

Report
MG 08:53 Thu
In reply to neilh:

You would hope! 

Report
Ridge 09:02 Thu
In reply to MG:

> They've now been ordered from Dyson and are a completely new design.  They will probably be see-through with purple knobs that fall off and cost five times normal ones.

> BAe are apparently about to start production of an existing design, which sounds more promising to me.

BAe will be 50 times the price and will be a metal box fitted for (but not with) a ventilator unit.

Report
wintertree 09:05 Thu
In reply to neilh:

> I would guess they are tackling it in various ways

The NHS is repurposing animal ventilators in some size bracket.  I’m wondering what’s larger than a dog and smaller than a horse though, and whose owners would go so far as to have it ventilated?  Price pigs?  Companion sheep?  Ponies?

Edit: link - https://www.bbc.co.uk/news/uk-england-52026687

Post edited at 09:09
Report
jkarran 09:13 Thu
In reply to MG:

> Yes not clear why it is “sensible”. There are plenty of car batteries around however.

Wouldn't these normally run with a compressed air and O2 line? If so, there's most of your power, the rest runs a screen and a few solenoids.

Or would you expect to be able to run without an airline instead pumping ambient air with suplemental O2 added as necessary?

jk

Report
MG 09:14 Thu
In reply to Ridge:

Ah!  Difficult choice of suppler then!

Report
jkarran 09:38 Thu
In reply to Deleted bagger:

> Can't imagine these vents will be very high tech. In-out, variable pressure and rate. I expect they'll be similar to the ones I used when i started nursing. Still fingers crossed they'll knock them out in time.

Rapid prototyping tools like https://4dsystems.com.au/ mean a competent and familiar developer in a team with the people who understand and can clearly explain the requirements could put the interface and control software together in a matter of days while others test and validate it. Devices like these have almost all the hardware needed available in the module, just needing a bit of external support: power supply, any patient/electrical isolation, op amplifiers, and a sensor set. I can't remember what watchdog/lock-up protection is built into these but there is some and it can be bolstered with an external power manager chip anyway to improve resilience/reliability. Firmware can be updated with a new SD card so fixes and features can be pushed out to basic early units quite simply. Pressure sensors will need to be hidden behind diaphragms in a custom module or made disposable to facilitate disinfection, whichever is quickest to make in volume.

I have no idea what the supply chain is like for suitable solenoids and other pneumatic parts, the stuff I use is low volume and pretty niche but they're used extensively in factory automation so I suspect there are sizeable manufacturers and stocks to be found.

Finding 30k of anything like this in stock is tricky, near impossible in the UK but a few thousand at a time less so, pallets can be here in a day or two with the right resources available and software portability is relatively easy from one module in a range to another so you don't need thousands all the same, just closely related. Quite sophisticated products can go together very quickly these days at the cost of cost and neatness (wires, adaptor plugs and boards etc to make all the bits play nice).

jk

Post edited at 09:40
Report
Mark Edwards 10:30 Thu
In reply to Rog Wilko:

Watching Nadhim Zahawi on Newsnight last night, squirming his way out of answering the questions was pitiful. Reading between the lines, yes we’ve spent millions on machines that have no chance of arriving until way too late for many people.

Report
Rob Parsons 10:39 Thu
In reply to Mark Edwards:

> Watching Nadhim Zahawi on Newsnight last night, squirming his way out of answering the questions was pitiful.

Yes - wasn't he awful: avoidance of questions; and otherwise waffle. Shame on him.

Report
neilh 10:53 Thu
In reply to jkarran:

A very interesting excercise in lean manufacturing. The teams in the lean catapult centre at Coventry will be working 24/7 on this. Very sharp minds.

Report
Deleted bagger 11:26 Thu
In reply to jkarran:

I imagine these vents will be pretty brutal with patients flattened and paralysed. Hope they get a descent dose of Midazolam!

Report
drunken monkey 13:02 Thu
In reply to Ridge:

And no spares, and about 5 years too late

Report
jkarran 14:03 Thu
In reply to neilh:

> A very interesting excercise in lean manufacturing. The teams in the lean catapult centre at Coventry will be working 24/7 on this. Very sharp minds.

No doubt. Problem as I see it is the government's procurement plan looks to be calling heavily on Johnson/Brexit donors ahead of apparently more suitable options. I fear there'll be a few weeks of largess for the boys, little return then back round the loop again hopefully with some adult procurement and engineering specialists leading, not Cummings and his muppet. I really hope that's just my instinctive revulsion colouring my perception but JCB and Dyson? I'm sure they each have a competent electronics, prototyping and manufacturing team but they make tractors and noisy trinkets.

jk

Report
neilh 14:28 Thu
In reply to jkarran:

I suspect its your perception. I listened to a medical device valve manufactuer talking about the turning time of these parts saying it is weeks. I thought yes it probably is to a very high spec, but that sort of thing is out of the window, these will be stripped back basic units. There is nothing wrong with that.I am not sure he really got the message about switching from niche to volume.

I doubt Cummins has the time to even be involved.Adonis would be better.

Post edited at 14:29
Report
jkarran 14:31 Thu
In reply to Deleted bagger:

> I imagine these vents will be pretty brutal with patients flattened and paralysed. Hope they get a descent dose of Midazolam!

Why do you think that? What they need to do is well , just not widely understood. Fast delivery doesn't have to mean poor function What's the old saying? Fast - Good - Cheap, pick two. Something like that.

jk

Report
wintertree 14:38 Thu
In reply to jkarran:

> but JCB and Dyson

JCB design and make their own engines (a rarity these days) which involves motors, airflow, microprocessor control, modelling, fabrication and assembly.  A turbo charged common rail diesel is a lot more complicated than a ventilator, with many of the same sensor types.  A small (in auto volume terms) firm that can design and make their own such engine is really quite special.

 Dyson make electrically powered devices to move gas around and have a very good core motors team as well as experience of packing them in small, robust units with microprocessor control and battery operation.  Dyson in particular seem like a good choice to me for what a ventilator is.  

JCB can certainly achieve things when they want to - https://en.m.wikipedia.org/wiki/JCB_Dieselmax

Post edited at 14:41
Report
jkarran 14:50 Thu
In reply to neilh:

> I suspect its your perception. I listened to a medical device valve manufactuer talking about the turning time of these parts saying it is weeks. I thought yes it probably is to a very high spec, but that sort of thing is out of the window, these will be stripped back basic units. There is nothing wrong with that.I am not sure he really got the message about switching from niche to volume.

Yeah, I'm assuming if these happen (not a given) then knowing the necessary volumes the parts will mostly have to be drawn from stock in other industrial supply chains and/or spares bins then coupled to existing medical equipment.

I see today on the Isle of Man a request has gone out to the public for specific types of full face snorkelling equipment! I presume that's either to improvise reusable positive pressure PPE or cobble together ventilators.

> I doubt Cummins has the time to even be involved.Adonis would be better.

Well someone thought a tractor and a hoover maker might fancy a pop at developing medical devices, seems to me a very odd leap to make on technical grounds.

jk

Report
jkarran 15:10 Thu
In reply to wintertree:

> JCB design and make their own engines (a rarity these days) which involves motors, airflow, microprocessor control, modelling, fabrication and assembly.  A turbo charged common rail diesel is a lot more complicated than a ventilator, with many of the same sensor types.  A small (in auto volume terms) firm that can design and make their own such engine is really quite special.

I'm not unimpressed that they retain the expertise to develop and manufacture complex engines but they are engines and they do take years. I'm sure they could make a ventilator, I'm just not really seeing why they'd be on the first page of companies you'd call to do so.

If I was looking to auto makers I would have thought with the F1 season all but cancelled that might be a useful scene to tap into, very capable, very agile, well resourced, great logistics capability and well connected to companies with lots of skilled labour and manufacturing capacity.

>  Dyson make electrically powered devices to move gas around and have a very good core motors team as well as experience of packing them in small, robust units with microprocessor control and battery operation.  Dyson in particular seem like a good choice to me for what a ventilator is.  

You're apparently far more impressed by Dyson's product than I am, high frequency Brushless motors and centrifugal compressors are hardly radical technical innovations, I'm afraid I mostly see blue LED's designer plastics and world class BS.

> JCB can certainly achieve things when they want to - https://en.m.wikipedia.org/wiki/JCB_Dieselmax

It's a good advert, I'll give them that.

jk

Post edited at 15:12
Report
neilh 15:14 Thu
In reply to jkarran:

I doubt F1 has the capability to fast track large production as they are high value extremely low volumes etc.Not really ideal.You will be making more ventilaotrs in one hit than F1 has made any car etc in goodness how many years.Might be ideal for prototyping, but that is it.

Report
wintertree 15:23 Thu
In reply to jkarran:

You could well be right about F1 teams.

> You're apparently far more impressed by Dyson's product than I am, high frequency Brushless motors and centrifugal compressors are hardly radical technical innovations, I'm afraid I mostly see blue LED's designer plastics and world class BS.

I agree on their products but then there’s no technical innovation needed for these ventilators.  They do need motors, valves, microprocessors, plastic cases, batteries (I bet dyson can supply the required backup batteries from in-pipeline stock) and a production line.  Shame their production lines moved out East like...

Report
nikoid 15:34 Thu
In reply to all:

I don't know anything about ventilators, but I imagine the control system is the tricky bit. Anybody know how they are controlled, ie inputs, blood pressure,  oxygen saturation, heart rate presumably are all used to control air flow or pressure? How do you prevent lungs being over pressurised etc etc.

Or are the emergency ventilators going to be more akin to something from years back before we had complex software controls?

It would be interesting to learn more. Can anyone educate me?

Report
FactorXXX 16:13 Thu
In reply to jkarran:

> If I was looking to auto makers I would have thought with the F1 season all but cancelled that might be a useful scene to tap into, very capable, very agile, well resourced, great logistics capability and well connected to companies with lots of skilled labour and manufacturing capacity.

They're already involved:
https://www.bbc.co.uk/sport/formula1/51975690

Report
Archy Styrigg 16:17 Thu
In reply to nikoid:

Controllers which are a breeze to program have been available for decades.
I might be bold enough to say that that's the 'easy' part.

Report
Dr.S at work 16:49 Thu
In reply to wintertree:

> > I would guess they are tackling it in various ways

> The NHS is repurposing animal ventilators in some size bracket.  I’m wondering what’s larger than a dog and smaller than a horse though, and whose owners would go so far as to have it ventilated?  Price pigs?  Companion sheep?  Ponies?

Most dogs that get ventilated (and recall that dogs mass from 500g to 100kg) do so during general anaesthesia.  In general the type of ventilator vets have is designed for anaesthesia specifically - there are few occasions when long term ICU ventilation is appropriate from a cost or ethics perspective. That said there are maybe 4 or 5 veterinary hospitals in the country that have one or two 'ICU 'vents.

In my hospital when we do ventilate a patient outside anaestheisa we use a modern human anaesthetic machine with an integrated vent - these are actually pretty capable and do a good job on the occasions when we want what for us is long term ventilation (4 or 5 days is a long time for us, but nothing from a human ICU perspective).

Those machines obviously exist in the NHS in large numbers - and will be the initial fall back once ICU vents are all in use and the new ones have not rocked up yet.

AFAIK no modified vet specific vents have been utilised by the NHS, they would need to go through some serious regulatory checks first. As the BBC says the profession has audited what we have of suitable (i.e. currently regulatory approved human equipment) and shared that information with appropriate folk.

Report
Mark Edwards 16:53 Thu
In reply to Archy Styrigg:

> Controllers which are a breeze to program have been available for decades.

> I might be bold enough to say that that's the 'easy' part.

Getting hold of microprocessors isn’t difficult, but getting hold of the quantities of the exact MCU’s the software is designed for could be another matter (porting to a different part is possible but isn’t trivial).

Then you are going to need EMI/EMC compliance testing (once you get the PCB’S and they’re populated), but the PCB manufacturers and testing labs have been shut down and probably CE Certification.

Calling on other industries is a nice idea but if the government had scaled up existing manufacturers instead, then some of the above problems could have been eliminated although chip availability could still be the major problem.

Report
wintertree 16:59 Thu
In reply to Archy Styrigg:

> Controllers which are a breeze to program have been available for decades.

> I might be bold enough to say that that's the 'easy' part.

I think Nikoid was asking about control algorithms not implementation hardware.  There’s only so many things the machine can sense and humans breath about 2 million times slower than an early 1980s microprocessor thinks so I don’t think implanting it will be hard but implementing it in a regulatory compliant way could be amidst a shutdown.  

The aerospace and automotive people have different regulatory frameworks for reliability that could I imagine be accepted in a pinch, with military and automotive (MISRA?) both having high reliability hardware/software platforms.

Report
Dr.S at work 17:00 Thu
In reply to Mark Edwards:

I think they are doing both - very sensibly.

Report
wintertree 17:08 Thu
In reply to Dr.S at work:

Thanks!  I was hoping you’d chime in.

Report
Mr Lopez 17:24 Thu
In reply to jkarran:

> Well someone thought a tractor and a hoover maker might fancy a pop at developing medical devices, seems to me a very odd leap to make on technical grounds.

And astonishingly, that's what the lifes of many will have to depend on. FFS

https://www.bbc.co.uk/news/uk-politics-52052694

Report
neilh 17:31 Thu
In reply to Mark Edwards:

How long does emc testing really take?Put it in a chamber for an hour.its hardly difficult and you can prove it at design stage.   Hardly rocket science. Stuff ( just ignore it) ec certification , need is driven by other factors. 

Report
MG 17:36 Thu
In reply to Mark Edwards:

If the choice is dead people or potentially not dead people, I think compliance etc can be short circuited. If need be government could provide liability cover.

Report
jkarran 17:51 Thu
In reply to neilh:

> How long does emc testing really take?Put it in a chamber for an hour.its hardly difficult and you can prove it at design stage.   Hardly rocket science. Stuff ( just ignore it) ec certification , need is driven by other factors. 

Your experience appears to have been rather different to mine. 2-3 cycles: full 6-8hr day of testing, design changes, re-prototype, re-test the failures, that is more in line with my experience for sensitive instrumentation. Good practice, tools, parts and experience help but you never really know (from design alone) it'll pass first go.

EMC is compliance is hard to argue against, you can't have a mobile phone on the ward killing people or your new machines meaning other critical kit stops working properly. I presume ordinarily the software would be at least SIL2, that would cause any company not experienced in safety critical system development a much bigger headache than EMC.

jk

Report
jkarran 17:54 Thu
In reply to FactorXXX:

Good. The tooling capability alone would be extremely valuable but the whole package is incredibly powerful, they're very capable machines.

jk

Report
wintertree 17:59 Thu
In reply to Dr.S at work:

> I think they are doing both - very sensibly.

Quite - you back every horse in this race, and if it barks you stick a saddle in it on the off chance.

Any successful approach to making a high volume manufacturing line for building regulation compliant ventilators is going to have a global market of up to 2 million units in the next 6 months.  Could be a successful export...  Not something to gloat about (not the word I’m after but I can’t find it) but this is where people like Dyson of the defence/auto industry can really change the world - I doubt any medical manufacturers produce and sell complex kit in these volumes; cracking the problem of rapid mass production rather than simply leaning on suppliers to make some more could change the world in the next 6 months 

I hope they succeed.

Report
Archy Styrigg 18:02 Thu
In reply to jkarran:

When I'm lieing on an ITU bed, gasping for breath, and there's some Heath-Robinson ventilating machine sat by the bed ... Staff: "Can't use it, it hasn't ticked all the compliance boxes, sorry."  Me: "F*ck that, tube me up!"

Report
johncook 18:04 Thu
In reply to jkarran:

Many manufacturing companies are changing to manufacture the parts required. The response to the govt has been massive. It is believed that the process of manufacture and assembly can be carried through almost as quickly as transporting outsourced units from outside the country.  It will; also put us in the position to continue manufacture when the rest of the world has stopped exporting. That is one of the reasons the govt has put in place this requirement for UK sourced stuff.

Believe in the UK and it's ability to respond to a crisis, even if the man in the street can't manage to obey the rules of the lockdown. (which could result in marshal law if we don't get our collective public act together. There are even some on this forum who think the rules don't apply to them!)

Report
neilh 18:18 Thu
In reply to jkarran
 

I agree it’s critical. But it is not really a barrier. There are ways you can be clever to tick the requirement. Just needs thinking through. 

Report
jkarran 18:20 Thu
In reply to johncook:

> Many manufacturing companies are changing to manufacture the parts required. The response to the govt has been massive. It is believed that the process of manufacture and assembly can be carried through almost as quickly as transporting outsourced units from outside the country.  It will; also put us in the position to continue manufacture when the rest of the world has stopped exporting. That is one of the reasons the govt has put in place this requirement for UK sourced stuff.

We don't make this stuff, we haven't for a generation if we ever did. We don't make touch screens, regulators, micro controllers, pressure sensors or amplifiers, we buy them from all over the world and at most assemble them into products here (mostly that's done abroad too).

Trying to build an industry on this crisis ahead of pragmatism will get us killed in our hundreds of thousands.

> Believe in the UK and it's ability to respond to a crisis, even if the man in the street can't manage to obey the rules of the lockdown

I make electronic instrumentation and gas sampling systems for a living, I have a bit of an idea what's available. What we build has plent of UK bits but it would be nothing without irreplaceable Thai, Chinese, Malaysian, German etc parts. 'Believe in the UK' to make, ground up UK source, machines rings hollow over any timeframe let alone 2-3 weeks but then what would I know.

jk

Report
Dr.S at work 18:22 Thu
In reply to wintertree:

Quite. I’m reminded of the days we used to build three or more models of bomber to ensure we had one that worked.....

Report
jkarran 18:26 Thu
In reply to neilh:

> I agree it’s critical. But it is not really a barrier. There are ways you can be clever to tick the requirement. Just needs thinking through. 

Personally I'd just consider it one of those things that had to be got right. It's a delay risk in any plan but building Around a widely available existing safety critical design from a harsh environment would be a solid starting point. An electronic stability control module perhaps?

jk

Report
jkarran 18:28 Thu
In reply to Archy Styrigg:

> When I'm lieing on an ITU bed, gasping for breath, and there's some Heath-Robinson ventilating machine sat by the bed ... Staff: "Can't use it, it hasn't ticked all the compliance boxes, sorry."  Me: "F*ck that, tube me up!"

Sounds good but that's not how it works. If the other kit you need won't work with it or it resets whenever a phone rings nearby you're still stuffed.

jk

Report
nikoid 19:33 Thu
In reply to wintertree:

Yes I think I was asking for a basic description of the control system at a "black box" level.

Report
Mark Edwards 23:33 Thu
In reply to neilh:

> How long does emc testing really take?

Between 1 and 3 days per Lab session depending on complexity. Getting it to pass is what takes the time. The normal procedure is the first test in the Lab you find out how bad it is, go back and try and isolate what is causing the problem, if you have the gear you can do this in house but without an isolation chamber the background RF means your mostly guessing. You try again and again and again, hopefully getting better each time. Eventually you’ve tried everything and it’s still a few dB over somewhere on the graph. That’s when the real work begins.

Report
Mark Edwards 23:39 Thu
In reply to MG:

> I think compliance etc can be short circuited.

And if your dirty circuit interferes with another piece of life support equipment and it results in people dying? Not a chance I would take. The testing regime is there for a reason. In a hospital, doubly so.

Report
wintertree 23:48 Thu
In reply to Mark Edwards:

There’s more to CE marking than the EMI/EMC testing that’s discussed up thread.  It means all relevant regulations are complied with.  As these devices take a pure oxygen feed, I imagine there are other regulations and tests about not causing giant fireballs...

Post edited at 23:49
Report
Dr.S at work 23:52 Thu
In reply to wintertree:

yes. airway fires are nasty.

Report
ben b 03:19 Fri
In reply to nikoid:

As mentioned elsewhere, most vents either pressure control or volume control. So, they can be set to blow e.g. 500mls of air in, and then allow it out again (against  bit of resistance to stop the lungs collapsing), or blow to a certain amount of pressure and then stop.

If you consider how lungs vary in size, shape and stiffness when packed full of COVID-19 related inflammation, if the lungs are really stiff and you have the vent set on delivering 500mls, the peak pressures may be catastrophically high (think overinflated balloon). We think the amount of shear force in the lungs also worsens inflammation. So often a pressure controlled ventilation strategy is used, and we accept that the lesser of the two evils is allowing small tidal volume ventilation at lower shear forces. In turn, this may mean worse ventilation - so lower oxygen and higher CO2 levels in the blood than we are 'happy' with. In turn, the high CO2 means a more acidic pH in the bloodstream (but for god's sake don't anyone ask me about apple cider vinegar now or I may blow a fuse!). This has knock on effects on muscle contraction, heart, brain, kidneys etc.

In peacetime ARDS ventilation is difficult to do well and needs to be as gentle as possible , for want of a better phrase. We are at risk of needing boy scouts and bike pumps, primarily because of governmental dithering and delay. We may get a bunch of vents through this route, but there are plenty of "non-invasive" vents out there in good supply that are capable of being repurposed. The FDA have a statement out allowing emergency legal modifications to allow this to happen.

I'm more worried about the lack of staff - even not particularly trained staff - and the lack of oxygen to drive the machines on. Oxygen cylinders contain a couple of hundred litres (standard D cylinders of the sort most often seen about 350l when full), vents often drive off 15l/min, so the maths is pretty compelling. Portable oxygen concentrators run about 1-5l/min tops. For an Excel-style field hospital for 2000 vents...

Ventilated patients need keeping asleep (drugs by infusion, 50ml syringe, maybe 12mls/hr) and turning (so pressure sores do not develop) and catheters emptying (assuming not on dialysis which needs constant attention) and bowel motions sorting out and a hundred and one other tasks. This is why ICUs in peacetime operate 1:1 nurse:pt ratios. We will almost certainly need to widen the ratio - and most ICUs run 1:2 ratios in extremis for short periods - but at 1:5 it is an unsustainable workload for a human being. 

So, please everyone, stay inside. Help people who need it only if you safely can. Support people staying out of hospital. Don't be the one who's bike ride results in a crash and a wrist fracture, or as an asymptomatic carrier ends up infecting an ICU nurse. Every one we lose from the workforce or who gets taken away from their core role is enough to tip multiple people over the edge. Flattening the curve is the one thing that you can all do, easily, and it literally makes a life saving difference. Good luck to us all - we are going to need it.

(Dr) B

Report
In reply to jkarran:

Apparently about half the ventilators in the world are made by a company in Ireland and they are setting up a couple of new production lines because lots of folk want to buy them at the moment.

https://www.irishexaminer.com/breakingnews/ireland/coronavirus-hse-and-ida-in-talks-with-ventilator-manufacturers-988985.html

So instead of dicking about with Dyson and JCB and untested designs that nobody has been trained on why not offer the Irish company a ton of money to make some for us, or if they absolutely can't scale up any more, license us to make an exact duplicate of one of their production lines somewhere in the UK.

I have a bad feeling that the Tories are handling ventilator procurement much the same way as they handled ferry procurement.

Report
Deleted bagger 08:25 Fri
In reply to nikoid:

Not deeply involved in the techie side. But I do know how to drive the Drager Babylog 5000 in all its modes. Looks like I'll be getting familiar with it again very soon. The unit I last on is currently down 13 nurses and 2 consultants. 

Report
Rob Exile Ward 09:25 Fri
In reply to ben b:

That seems all rather more plausibly complex than  Dyson or JCB knocking out Heath Robinson affairs that might or might not work - I have visions of 10,000 devices being held together by duct tape, a la Apollo 13 but without the supporting team of engineers focused on just one device.

Profoundly depressing though.

Report
Rog Wilko 09:38 Fri
In reply to Rog Wilko:

This project is going to official review today.

Report
HansStuttgart 09:45 Fri
In reply to Rog Wilko:

See here: https://twitter.com/aroradrn/status/1243091107149418496

I assume (hope) all government noise about british inventiveness in making new ventilators is just some contigency planning for when the main project of ordering more proper ventilators from the usual suppliers don't work.

Germany has a reasonable stock and is ramping the production to beyond the level it can actually use (the bottleneck is qualified personel). So the UK should just join the common European procurement and then the ventilators can be distributed to those countries who need them most.

Report
Hardonicus 09:57 Fri
In reply to Mark Edwards:

One consideration for developing control systems on microcontrollers will be stringent requirements on robust firmware development that will be required for medical devices operating in a life-critical capacity. Getting such firmware signed off is probably more complex than other aspects like EMI/EMC complicance in my opinion.

Report
neilh 10:19 Fri
In reply to Mark Edwards:

Well the first obvious thig is that these units are mobile ( they are on wheels) so they are easy to move around, which makes it a lot easier than a machine to test.. You could test the system before completion.You can use a " wand" to help determine if you are on the right track.If you speak to some really good EMC compliance people then can identify the issues from Day 1... they will know from experience what to go for..There are all sorts  of ways you can do this so its not adding to the time to build.

Lets put it this way, these units are going to have priority to go into a chamber- 24/7. No messing.

Report
johncook 10:40 Fri
In reply to jkarran:

There is the problem. We have chased the cheapest options, from countries with poor labour controls because we want to be seen with the latest gizmos, and it is cheaper to have them made by child labour in Korea/India wherever. For years the (hypocritical) British public have been saying that we should be making more stuff here, the buying cheaper foreign made options! Now the govt is trying to bring our production back home people like you are criticising them. British manufacturing should be pleased at this prospect, and the public should back them.

(There is one company who are now 'doing a China' and reverse engineering a ventilator by copying one which was made elsewhere. They are within a few days of production just so long as some busy body doesn't stop them or fine them for breaching copyright/patent. Everything they need is being produced in the UK, mostly by specialist small machine shops. I hope they manage to build up supply to a level where we don't need to be in the world queue for the chinese/korean/etc made stuff.)

As a manufacturer you should be pleased that production is being upped in the UK. Maybe your company will gain from some of the spin-off! Just so long as you are enthusiastic about avoiding cheap reverse engineered imports.

Report
jkarran 12:29 Fri
In reply to johncook:

> There is the problem. We have chased the cheapest options, from countries with poor labour controls because we want to be seen with the latest gizmos, and it is cheaper to have them made by child labour in Korea/India wherever. For years the (hypocritical) British public have been saying that we should be making more stuff here, the buying cheaper foreign made options! Now the govt is trying to bring our production back home people like you are criticising them. British manufacturing should be pleased at this prospect, and the public should back them.

The stuff we're talking about relies not on cheap labour or lax regulation but on decades of investment in cutting edge infrastructure and training. That ecosystem worth many billions developed over decades cannot just be magiked into existence because we wish it so. No amount of patriotic zeal can do that over the course of a single government let alone the 2-3 weeks we have to build capacity in order to avert a humanitarian catastrophe. I criticise the government not for their desire to diversify British industry (that's laudable) but for the appearance of cronyism their timing if building a new export industry is the game. Now, stood watching a tsunami piling up towards us is not the moment.

> (There is one company who are now 'doing a China' and reverse engineering a ventilator by copying one which was made elsewhere. They are within a few days of production just so long as some busy body doesn't stop them or fine them for breaching copyright/patent. Everything they need is being produced in the UK, mostly by specialist small machine shops. I hope they manage to build up supply to a level where we don't need to be in the world queue for the chinese/korean/etc made stuff.)

Oh goody, IP theft celebrated and legitimised, that'll help us as a knowledge economy valued for it's robust rule of law in the coming years. We don't make this stuff, we can make the plastics and the boards and the valves, the software but the silicon, the glass, the sensors, the passives, the batteries, they're all import if you need the right parts at any kind of scale.

> As a manufacturer you should be pleased that production is being upped in the UK. Maybe your company will gain from some of the spin-off! Just so long as you are enthusiastic about avoiding cheap reverse engineered imports.

Eh, sorry I don't follow most of that. Production isn't being upped, many businesses are throttled right back, some are hoping to diversify. We're ticking over thanks, we export globally, as the virus moves around so to does activity. That and we're picking up some work from the scaling up of the medical oxygen supply chain.

jk

Report
Aly 12:56 Fri
In reply to jkarran:

I suspect manpower, especially experienced ICU staff may be the limiting factor in trying to keep ventilated patients alive.  Even given that, the prognosis of a patient who needs to be put on a ventilator due to Covid-19 doesn’t appear to be great.  I’ve been struggling to find good numbers on this, I’ve heard mortality ranging anywhere from 50-80% so if you want to save lives social and behavioral changes are probably far more effective than producing a mountain of vents.  The CDC published a report yesterday and their ICU outcomes didn’t look so bad in the under 75’s but I think the data only went up to March 12th and a lot has changed since then in the States.
 

In terms of engineering companies making ventilators, whilst the device itself is not that complex (a basic transport vent like an Oxylog only needs pressurized O2, wall or cylinder, and power (battery lasts about 4hrs btw)), engineering in the redundancy that it’s not going to fail and the sensors so it will coordinate with the patients breathing would not be easy I suspect.  I assume this is what makes them so specialized and expensive, rather than just the “put air in, let it out” function.  
Having said that I’m neither an intensivist, an engineer, or a respiratory physician (lowly ED doc) so that could be rubbish.  And in the end we’ll all just have to work with whatever equipment we have available.  

Report
neilh 14:34 Fri
In reply to jkarran:

Turn it round. What does a ventilator normally cost. £20 k. Approx. 

it’s pretty low cost for the numbers typically  made in a year. Hardly high end  

indicating it’s possibly  not that difficult to make if you see what I mean .

Report
jkarran 15:05 Fri
In reply to neilh:

> Turn it round. What does a ventilator normally cost. £20 k. Approx. 

> it’s pretty low cost for the numbers typically  made in a year. Hardly high end  

> indicating it’s possibly  not that difficult to make if you see what I mean .

I'm not arguing they are inherently difficult to make, I have all the bits and skills needed in my office/workshop, I expect you might too. I'm arguing that achieving regulatory compliance is necessary and with the best will in the world it takes significant time. One could make a strong case for some corner cutting here but not complete abandonment. Also there is no way we can make these to the government's spec if UK sourced means UK made because we simply don't make some of the electronic parts and they're far far too complex to start now, let alone be up and running producing several lines of chips by the 10's of thousand within a week or two. 

Jk

Report
neilh 15:59 Fri
In reply to jkarran:

I was having minnor urgent surgery 2 weeks ago and in the nursing station you could overhear the nurses and a consultant review the medical protocals being scrapped in aniticpation of the workload.

If medics can do that engineers certainly can on a piece of equipment.

Govt approach on this is to source from a few companies, not tie them into one supplier, all sensible stuff.

These ventailitors are not going to appear overnight anyway and save the day.

A vaccine is, imho

Report
MG 16:08 Fri
In reply to jkarran:

> and they're far far too complex to start now, let alone be up and running producing several lines of chips by the 10's of thousand within a week or two. 

Is that correct?  I know little about electronics but the min spec the government has given for the ventilators sounds fairly basic to me.  Air in and out with certain pressure and volume.  Can't a control system for that be made from generic components?

Report
jkarran 16:12 Fri
In reply to neilh:

> I was having minnor urgent surgery 2 weeks ago and in the nursing station you could overhear the nurses and a consultant review the medical protocals being scrapped in aniticpation of the workload. If medics can do that engineers certainly can on a piece of equipment.

I don't disagree but to extend your analogy they're not dropping everything, your surgeon still scrubbed in even if his team may have been a little smaller, his hours a little longer, your discharge a little quicker. Some things are more important than others. EMC and electrical isolation is very important for example, software integrity and reliability can be addresed a number of ways while delivering a functional device to get us through this emergency. Material toxicity and ability to sterilise reliably, that's important. Oxygen compatibility is important...

> Govt approach on this is to source from a few companies, not tie them into one supplier, all sensible stuff.

Not arguing with that though as I've said already the first pass list of those approached stinks of cronyism.

> These ventailitors are not going to appear overnight anyway and save the day.

> A vaccine is, imho

A vaccine overnight or did you not finish the post?

Seems a vaccine is likely a year off if it comes, peak demand for ICU beds is maybe 4-6 weeks off. We don't need overnight but in the next 3-4 weeks assuming there are any capable staff left well and on duty by then anyway. As has been pointed out a few times, the staffing crisis is likely to be as acute as the equipment shortage.

jk

Post edited at 16:12
Report
jkarran 16:26 Fri
In reply to MG:

> Is that correct?  I know little about electronics but the min spec the government has given for the ventilators sounds fairly basic to me.  Air in and out with certain pressure and volume.  Can't a control system for that be made from generic components?

Generic like op-amps, potentiometers, switches and transistors? You absolutely could do it but I don't know why you would nor where you'd source those parts if they had to be UK made.

I suppose you could go old school and build it electro-mechanical with timer relays and pneumatic logic but it doesn't seem the optimal solution if the idea is quick, functional and reliable, nor the basis of a thriving new export business if that's the concern about 'UK sourced'*.

*I don't actually think it is, I think this is being misinterpreted. I think the aim is to ensure any prototype scales to a few 10k units without running into supply chain issues so secured stock from overseas should suffice. Realistically this is essential if you want electronics.

jk

Report
MG 16:30 Fri
In reply to jkarran:

> *I don't actually think it is, I think this is being misinterpreted. I think the aim is to ensure any prototype scales to a few 10k units without running into supply chain issues so secured stock from overseas should suffice. Realistically this is essential if you want electronics.

That's what I meant - components so common, supply isnt an issue, even now. 

Report
wintertree 16:32 Fri
In reply to MG:

> > and they're far far too complex to start now, let alone be up and running producing several lines of chips by the 10's of thousand within a week or two. 

> Is that correct?  I know little about electronics but the min spec the government has given for the ventilators sounds fairly basic to me.  Air in and out with certain pressure and volume.  Can't a control system for that be made from generic components?

Every car has pressure sensors in it - I bet there’s a load in the supply chain.  RS have over 500 types of pressure sensor in stock; I’ll bet good odds there’s several with 1000+ quantities in stock.  Valves can be and are made in the UK.  I’m more positive than JK on this...  A valve can be an injection moulded casing, a magnet and a voice coil.  You can get an injection mould made in days by protolabs from your laptop, and magnets can be made easily enough; coils are trivial.  Safety certifying it is another matter but there’s a lot of this stuff made here.  You can make a pressure sensor with a glass tube, a ball and a spring.  There’s endless solutions - the key is to pick some workable ones and work them, fast.

Post edited at 16:35
Report
jkarran 17:57 Fri
In reply to MG:

> That's what I meant - components so common, supply isnt an issue, even now. 

Thousands of a device say an amplifier: easy enough to find in UK stock if you plan ahead to use one you know is available and are relaxed about the exact version (often the same devices from the same production run, just factory sorted by performance and priced accordingly). Tens or hundreds of thousands of devices from uk stock, you have to start relaxing the spec, designing for a very broad range of possible devices sharing basic characteristics an footprint. It's doable but what you build is increasingly limited and verifying the performance of the myriad similar but different versions would be a nightmare, lots of devices have really odd quirks particularly at the edges of their operating envelopes!

For passive devices, diodes, transistors, a generic spec and a careful design is generally fine, op-amps less so, microprocessors and specialist chips, you start to run into limited choices that are 100% interchangeable with several other parts. There's an alarming amount of devices that are genuinely single source, poorly stocked and sporadically available, it's a battle I've been fighting for years to keep old designs in production. For years after 2008 there were all sorts of random shortages, partly the economic shock, partly a shift to reduced toxicity 'ROHS' parts by 2013 which caused various technical problems and obsolescence issue. It was quite a headache.

jk

Report
jkarran 18:18 Fri
In reply to wintertree:

My issue all along has been with the bizzare idea the parts must be UK made (which I actually think is a misinterpretation). If that means from UK held stock or stock which can be secured and delivered in days then it's a very different problem. I just find the idea we can make all this stuff at short notice without drawing on the existing capability of others nations a bit daft and frustrating, some of it is extraordinarily specialised and capital hungry, microchip fabrication being the most glaring example. I have no doubt we could tool up and produce: PCBs, software, housings, valves, pipes, mouthpieces etc in days if we made a really well coordinated effort to draw on industries and businesses with expertise. Or that things like pressure sensors, displays, solenoids and controllers couldn't be drawn from existing industrial supply lines and stock bins in their 1000's (dubious about 10'sK or 100'sK for all but the simplest parts), just that they wouldn't be UK made, they'd be Chinese, German, Turkish etc assemblies populated with chips from all over Asia and the US. That's just the world as it is.

As you say, a modern car contains almost if not all the key electro-pneumatic components: valves, sensors, reliable controller with a huge range of well protected IO lines, human interface via the infotainment unit, batteries. I don't think it's impossible.

Report
ben b 00:43 Sat
In reply to Hardonicus:

True, but CDC have opened this up dramatically. For obvious reasons life-preserving medical equipment tends to be very carefully regulated, but the CDC have basically issued guidance saying "do the best you can, and if you don't have what you need, feel free to improvise". It's essentially the only sensible advice I have seen out of the US for some decades!

The test will be if litigious US agree that in a national emergency suboptimal things happen because they are probably better than nothing, but not as good as normal care. I see the "why not sue your doctor - it's free" ads are still going in the UK.

I'm sure that's a huge reassurance to all the healthcare workers out there, to know that the industry of complaint and compensation is right behind them, popping on its rubber gloves and reaching for the lube...

b

Report

Please Register as a New User in order to reply to this topic.