Precision gas volume delivery

Get two good quality thermos flasks. Pour liquid nitrogen into one and liquid oxygen into the other. Seal each flasks with a rubber stopper with a hollow metal tube through it.

The tube just goes through to the bottom of the rubber stopper, and doesn't get anywhere near liquid oxygen/nitrogen.

Hook up both tubes to a Y- junction and connect the single tube to your patient.

Now get two wired-up resistors on sticks, and shove them through the corks, down to the bottom of the thermos flasks (under the liquified oxygen/nitrogen) and set up your electronics to boil off nitrogen and oxygen at the rate you want.

The gas coming off will be too cold (so heat it) and too dry (so evaporate some water with another resistor).

Don't ever seal off the system - the rubber bungs will blow out of the thermos flasks, if you haven't pushed them in too hard (they are your safety valves).

--
Bill Sloman, Nijmegen

Steve - try posting to sci.engr.mech

Not sure if it helps, I've seen some transducers that do 4/20mA or 0/5V to pressure conversion that have internal feedback sensors. Maybe there is a flow equivalent.

I think there are also flow control regulators that consist of more than just a needle valve. These may be cheaper/easier than mucking around with pics etc. rob

Needle valve are crap but have time proved reliability. Hate to imagine Rover's/Polly's/Spot's ... life-state being decided by an I-P valve in a complex servo loop with attendant fail-safe monitoring/voting systems and power back ups :-) regards john

[...]

Yes, they exist. They are called Mass Flow Controllers (MFC). In fact a combination of a Mass Flow Meter and a control Valve. I use them but in other ranges(mL/min)than what you require(l/min). Note that MFC's regulate MASS-flow, in fact they control gas flow in terms of WEIGHT (kg/hr, gr/min etc.), not in terms of VOLUME. However, since volume for most applications is the usual unit to work with, MFC's are calibrated in Ln/min (Liters-normal/min), i.e. a particular volume-flow(of gas) at 0°C.

[...]

You'll only need a power supply and a potentiometer to set the flow. A usefull addition is a digital voltmeter to readout the flow. For lots of explanation have a look at:

Rgds, Jan.

The standard flow control regulator used for welding shield gas is a fixed pressure regulator, a needle valve and a flow meter. These work pretty well. If the back pressure on your flow control valve is much less than the regulated supply pressure there is no reason this type of flow regulator should not be easy to adjust and rock steady once adjusted. Perhaps you need a better quality filter/separator, regulator, needle valve or flow meter? Does the pressure to the needle valve vary significantly in use? Do you have a flow meter, or do you have to wait and measure cycle rate of the ventilator?

Glen

My needle valve is indeed crap, and has more than demonstrated time proved *un-reliability*. You'd really struggle to do worse with an electronic control circuit than this basic valve achieves I fear. Tidal volume delivered can spontaneously switch to either negligeable or double target - I have learned to use the term "target" rather than "set volume" in this context. Crap in an op theatre context equals non-viable IMNSHO. The safety aspects of ventilation when using cheaper kit are actually dealt with using both an inflation pressure limit to catch overventilation, and visual/non-visual assesments of respiratory parameters to catch opposite problem. So the life state is essentially watched by a human, not the machine,this is not an un-attended life support piece of kit. It's just supposed to make our job easier and free our hands to do more valuable things than squeeze a bag.

Steve

On Sat, 21 Jan 2006 18:32:09 +0100, "J.Koning"

Looks very promising, thanks!

Steve

The supply is 60psi from a high quality regulator. It is tidal volume (ie one single breath) not cycle rate that is the issue here. It's not the same as welding or indeed setting stable anaesthetic fesh gas flows. The flow rate never really stabilises at all, we are talking seconds total of flow here, one breath, just a "squirt of air". A plot would give a curve. So you can't use a human readable flowmeter in any meaningful way. A micro doing integration on flow to derive volume in real time maybe, but as I said that's getting a bit complex for this application.

Steve

It seems to me that a simple way to do this would be to use a good filter upline of a regulator to regulate in the 5-50PSI range, and an orifice.

-

Many medical device manufacturers (Puritan-Bennett, Siemens, Bird, etc.) build ventilators that do a very good job of ventilating all manner of patients and so would very likely be suitable for other types of animals. The PB 7200 uses a much more sophisticated mechanism to control flow than the needle valve you describe - a proportional flow control valve. The solenoid, or rather the outer housing, of this device is about 4" in diameter. Inspired and exhaled flow are measured with a hot-wire anemometer. You can purchase a used 7200 for $6000 . No doubt there are less expensive ventilators available that do a reasonable job. Why not spend a few bucks and save yourself a lot of headaches.

You need a mass flow controller. You can get complete programmable systems, controllers only, or controllers with interface to computer systems from MKS Instruments. Look at

You'll find exactly what you need. For example, an 1179A, maybe the 20 liter/mn or 50 l/min,

plus a 246C single-channel controller, connected with a CB259-5-10 cable makes a complete programmable system. As I recall, the whole package is less than $1000 (they seem to have removed pricing from the site}.

John Perry

Well, I seem to have misunderstood. I had thought that all you wanted was a well-regulated flow rate - that's trivial: go to a weld shop, and look at their flowrate meters on their TIG and MIG rigs.

But now, you're talking about designing a machine that can substitute for human perception, and yea verily, compassion, since you are now apparently having to maually bag your patients.

There's no way I would trust any machine to do active bagging of patients, human or not, unless you want to go full-on medical respirators, (the kind that they use in those "should we pull the plug" cases), but you seem to be wanting to save a buck first.

Get a gas flow regulator, set the flow rate, and continue to bag them by hand - that's the only really safe way to do it. (other than spend maybe $100K.)

Don't short-change your patients.

Good Luck! Rich

Post this on a scuba diving group. They may have some ideas taken from mixed gas breathing systems that should be relatively reliable from a life safety point of view.

-- Paul Hovnanian mailto: snipped-for-privacy@Hovnanian.com

------------------------------------------------------------------ Matter cannot be created or destroyed, nor can it be returned without a receipt.

Might be the chosen solution ;-). Just exploring the options at the moment. It's a bit irritating when you have paid good money for something that doesn't really cut it!

Steve