Oxygen Cylinder Duration Calculator
Oxygen Cylinder Duration Calculator
How long an oxygen cylinder will last at a given flow, worked from the gauge pressure and the cylinder’s own conversion factor, with a reserve margin you set and a plain yes or no against the transfer time you actually need. The conversion factors are BOC’s published nominal contents divided by BOC’s published nominal pressures, size by size, and the page shows that division rather than hiding it — because the same letter means a different cylinder in a different country, and a borrowed factor is a wrong answer that looks right.
This calculates a rate or a volume from a prescription that has already been written. It does not decide what to give. Check every figure against the prescription and your local policy, and have it independently checked before administration.
Usable minutes at this flow, and whether that covers the time you need
A BOC size CD cylinder reading a full 230 bar, running at 4 L/min, holding back a 25 per cent reserve, with 90 minutes of transfer to cover
One division, one multiplication, one subtraction — and where each number has to come from
- C_nominal
- the nominal contents of a FULL cylinder of this size, in litres of gas at atmospheric pressure, as the supplier publishes it
- P_nominal
- the pressure a full cylinder of this size is filled to, as the supplier publishes it. 137 bar for BOC’s UK pin-index medical oxygen, 230 or 300 bar for the lightweight integral-valve sizes
- factor
- litres of gas per bar of gauge pressure. Not a constant of nature and not a property of oxygen: it is a property of one cylinder size from one supplier, and it is the quotient of the two numbers above. CD and HX both come out at exactly 2.000 and 10.000 L/bar, which is a useful check that the published pairs are consistent
- P_gauge
- what the gauge reads now. Gauge pressure, not absolute: a gauge reading zero still holds a cylinder’s worth of gas at atmospheric pressure, which is one atmosphere out of 137 or 230 and is the reason this approximation is good to better than a per cent
- r
- the reserve margin as a fraction of a full cylinder. The margin is taken against the FULL contents, so r = 0.25 means the calculation stops when the gauge reaches a quarter, which is how the gauge is actually read
- flow
- the continuous flow in litres per minute. Valid for a flowmeter into a mask or cannulae. Not valid for a demand or conserving valve, and not valid for a ventilator
Worked example
A BOC size CD cylinder reading a full 230 bar, running at 4 L/min, holding back a 25 per cent reserve, with 90 minutes of transfer to cover
The conversion factor first, because everything else is built on it. BOC publishes the CD as 460 litres nominal contents at 230 bar nominal pressure, so the factor is 460 / 230 = 2.000 litres per bar exactly. The HX comes out at 2,300 / 230 = 10.000 exactly on the same chart, which is a reassuring sign that the two published columns were derived from one another rather than measured independently.
At a full 230 bar the cylinder holds 2.000 × 230 = 460 L. At 4 L/min that is 460 / 4 = 115 minutes run to empty. BOC's own duration table prints 1 hour 55 minutes for a full CD at 4 L/min, which is the same 115 minutes, and prints it with no reserve allowance at all — the supplier's table is a count of the gas, not a plan.
Now the reserve. A quarter of a full cylinder is 0.25 × 460 = 115 L held back, leaving 460 − 115 = 345 L usable. At 4 L/min that is 345 / 4 = 86 minutes, which is the number in the large type.
And the answer to the question is no. Eighty-six usable minutes does not cover a 90-minute transfer. Run to empty it would have been 115 minutes and looked comfortable, which is exactly why the reserve belongs in the headline rather than in a footnote: a full CD at 4 L/min is not a 90-minute cylinder. The flow that 345 L would just sustain for 90 minutes is 345 / 90 = 3.83 L/min, and 90 minutes at 4 L/min needs 360 L, so it is 15 L short.
Change one number and watch it move. The same CD at 2 L/min gives 345 / 2 = 172 minutes usable, comfortably over. At 15 L/min it gives 23 minutes, and BOC's table for a full CD at 15 L/min prints 30 minutes run to empty — the entire cylinder, at the highest flow a mask reservoir needs, is half an hour.
The same arithmetic on a half-full CD: the gauge reads 115 bar, so 2.000 × 115 = 230 L in the cylinder, less the same 115 L reserve, leaves 115 L usable and 29 minutes at 4 L/min. Note that the reserve is unchanged while the contents have halved, so the usable duration falls much faster than the gauge does. At a quarter of a cylinder the usable duration is zero by definition.
If you have a psi gauge, convert before anything else. 230 bar is 230 × 14.5037738 = 3,336 psi; 137 bar is 1,987 psi. The page does that conversion itself and prints both, because the one thing worse than reading the wrong gauge is reading the right gauge in the wrong unit and getting an answer 14.5 times out in a direction that looks plausible.
BOC UK medical oxygen cylinders: published contents, published pressure, and the factor that follows
| Size | Nominal contents | Nominal pressure | Litres per bar | Run to empty at 4 L/min | Usable at 4 L/min, 25 % reserve |
|---|---|---|---|---|---|
| C | 170 L | 137 bar | 1.241 | 42.5 min | 31.9 min |
| D | 340 L | 137 bar | 2.482 | 85 min | 63.8 min |
| CD | 460 L | 230 bar | 2.000 | 115 min | 86.3 min |
| ZD | 605 L | 300 bar | 2.017 | 151.3 min | 113.4 min |
| E | 680 L | 137 bar | 4.964 | 170 min | 127 min |
| F | 1,360 L | 137 bar | 9.927 | 340 min | 255 min |
| HX | 2,300 L | 230 bar | 10.000 | 575 min | 431.3 min |
| ZX | 3,040 L | 300 bar | 10.133 | 760 min | 570 min |
| G | 3,400 L | 137 bar | 24.818 | 850 min | 637.5 min |
| J | 6,800 L | 137 bar | 49.635 | 1,700 min | 1,275 min |
Checked against BOC’s own duration table — a full cylinder, run to empty
| Size and contents | At 2 L/min | At 4 L/min | At 10 L/min | At 15 L/min |
|---|---|---|---|---|
| CD, 460 L — BOC prints | 3 h 50 | 1 h 55 | 0 h 46 | 0 h 30 |
| CD, 460 L — contents divided by flow | 230 min | 115 min | 46 min | 30.7 min |
| HX, 2,300 L — BOC prints | 19 h 10 | 9 h 35 | 3 h 50 | 2 h 33 |
| HX, 2,300 L — contents divided by flow | 1,150 min | 575 min | 230 min | 153.3 min |
| ZX, 3,040 L — BOC prints | 25 h 20 | 12 h 40 | 5 h 04 | 3 h 22 |
| ZX, 3,040 L — contents divided by flow | 1,520 min | 760 min | 304 min | 202.7 min |
Why the letter on the cylinder is not enough
| Size letter | BOC United Kingdom | BOC Australia | Ratio |
|---|---|---|---|
| C | 170 L | 490 L | 2.9 times |
| D | 340 L | 1,600 L | 4.7 times |
| CD | 460 L | 630 L | 1.4 times |
| E | 680 L | 4,000 L | 5.9 times |
The conversion factor, the reserve, and the three ways this arithmetic goes wrong
A compressed-gas cylinder is a pressure gauge attached to a volume of gas, and the only thing connecting the two is a number that belongs to that cylinder. Gauge pressure falls in proportion to the gas left, so the litres remaining are the gauge reading multiplied by a conversion factor in litres per bar — and that factor is the cylinder’s nominal contents divided by its nominal filling pressure. Both of those are published by the gas supplier. The quotient generally is not, which is why this page prints the division rather than presenting a factor as though it were a property of oxygen. For BOC’s UK medical oxygen the pin-index sizes C through J are filled to 137 bar and the lightweight integral-valve sizes to 230 or 300 bar, so the factors run from 1.241 L/bar for a C to 49.635 L/bar for a J. A CD works out at exactly 2.000 L/bar and an HX at exactly 10.000, which is a useful internal check on the published pairs.
The first way this goes wrong is a borrowed factor. Cylinder letters are not an international standard. BOC’s own Australian data gives a size C as 490 litres and a size D as 1,600, against 170 and 340 for the same letters in the United Kingdom — a factor of 2.9 and a factor of 4.7. North American practice uses a different set of letters again, with gauges marked in psi and factors conventionally quoted in litres per psi. A factor carried from one chart to another is undetectable in the answer: it produces a plausible duration in plausible units, and nothing about it looks wrong. The page therefore names the supplier and the country for every preset and offers a field for the two numbers off your own chart.
The second is the reserve. Contents divided by flow is a count of the gas, and it is what the supplier’s own duration tables print: BOC’s table for a full CD at 4 L/min says 1 hour 55 minutes, which is 460 divided by 4, with nothing held back. Nobody plans a transfer that way. A cylinder is changed while it still has gas in it, the usual convention being the bottom quarter of the gauge, and the difference is not small — a full CD at 4 L/min is 115 minutes of gas and 86 minutes of usable time. The 90-minute transfer that the run-to-empty figure comfortably covers is one the cylinder does not actually cover. That is why the reserve is in this page’s headline and the run-to-empty figure is a row underneath it. Note also that the reserve is a fixed number of litres while the contents fall, so the usable duration drops much faster than the gauge does, and reaches zero while the gauge still reads a quarter.
The third is the delivery device. Everything here assumes a constant continuous flow from a flowmeter into a mask or nasal cannulae, because that is what the flow setting means. A demand or conserving valve flows only on inspiration and extends a cylinder by a factor of two to four that depends on the patient’s respiratory rate and pattern, so this page understates its duration and cannot tell you by how much. A ventilator is worse: its consumption includes driving gas and bias flow as well as the patient’s minute volume, and is a property of the ventilator. High-flow nasal oxygen at 60 L/min empties a CD in under eight minutes. If the device is not a plain flowmeter, the answer on this page is a lower bound at best.
What the page deliberately does not do. It does not suggest a flow rate, a target saturation or an oxygen prescription — those are clinical decisions and oxygen is a prescribed drug. It takes a flow that has already been set or prescribed, a gauge you have read and a cylinder you have identified, and it answers two arithmetic questions: how long have I got, and is that long enough. The second answer is a plain yes or no against a time you type in, because that is the form the question is actually asked in at the lift door.
Frequently asked questions
Where do the conversion factors come from?
They are not taken from anywhere as factors. For each size the page carries BOC’s published nominal contents in litres and BOC’s published nominal filling pressure in bar, both from BOC’s Medical Gas Cylinder Data Chart, and divides one by the other. The result is shown on the page as a row in its own right so that you can see what it was built from. Two independent cross-checks were applied before the figures were used: BOC’s own patient information leaflet for compressed medicinal oxygen lists the same contents for every size, and BOC’s published duration tables for the integral-valve cylinders reproduce exactly as contents divided by flow. If your cylinder is not a BOC UK one, use the last option in the size list and enter the two numbers from your own supplier’s chart.
Why does the headline hold back a reserve when the supplier’s own table does not?
Because the supplier’s table answers a different question. Contents divided by flow tells you when the cylinder stops, which is a fact about the cylinder. What a transfer needs to know is how long it can be relied on, which stops earlier — a cylinder is changed while it still has gas in it, conventionally at a quarter of the gauge, and BOC’s own tables print a column at that 25 per cent mark. The margin matters most in the cases that look safest: a full CD at 4 L/min is 115 minutes of gas and 86 minutes of usable time, so the same cylinder passes a 90-minute transfer on one figure and fails it on the other. Set the reserve field to zero if you want the run-to-empty figure in the headline; it is printed as a row either way.
My gauge is in psi. Does that change anything?
Only the first step. Choose psi in the unit selector and the page divides by 14.5037738 to get bar before doing anything else, and prints the reading in both units so you can check it against the dial. The conversion factor itself stays in litres per bar on this page. North American charts conventionally quote factors in litres per psi instead, which are the same physics with a different denominator: a litres-per-bar factor divided by 14.5038 gives the litres-per-psi equivalent. If you are using a North American chart, take its contents and its service pressure from the same chart and enter them in the last option rather than mixing a factor from one source with a pressure from another.
Does this work for a demand valve, a conserving device or high-flow nasal oxygen?
No, and it will understate every one of them differently. A demand or conserving valve only delivers gas during inspiration, which typically extends a cylinder by a factor of two to four — but the factor depends on respiratory rate, tidal volume and the device’s own trigger behaviour, so there is no number this page could apply. High-flow nasal oxygen and ventilators consume gas the patient never breathes, so for those the answer here is not even a lower bound unless you know the total gas draw. Use this page for a flowmeter into a mask or nasal cannulae, which is what the flow setting on it means.
Why is the gauge pressure treated as if the cylinder were empty at zero?
Because the error is smaller than the gauge can be read to. A gauge reads gauge pressure, so at a reading of zero the cylinder still holds its own volume of gas at one atmosphere — about 2 litres for a CD. Against 460 litres that is 0.4 per cent, and against a 137 bar cylinder the one atmosphere is 0.7 per cent of the fill. Oxygen’s slight departure from the ideal gas law over this pressure range is of the same order and of the opposite sign at the top end. Both are far below the resolution of a dial you are reading in a corridor, and the published nominal contents already account for the real behaviour of the gas.
The page says the gauge reads above the cylinder’s full pressure. What does that mean?
It means the size selected and the reading entered do not belong together, and it is a genuinely useful error. A BOC pin-index C, D, E, F, G or J is filled to 137 bar and physically cannot read 230. If you have selected CD or HX and are reading 137, that is a cylinder at 60 per cent, which is fine. If you have selected C and entered 230, either the size is wrong or you are looking at a psi gauge and reading it as bar. The page caps the contents at a full cylinder rather than reporting gas that cannot be in there, but the number above the cap is not trustworthy and the selection needs fixing.
Does this page tell me what flow to give?
No. Oxygen is a prescribed drug and the flow, the device and the target saturation range are clinical decisions made for a patient this page knows nothing about. It takes a flow that has already been set and answers how long the cylinder will support it. The one flow figure it does produce — the flow that the usable gas would just sustain for the time you need — is arithmetic in the other direction and is there so you can see how tight the margin is, not as a recommendation to turn the flowmeter down.
Related calculators
References
- BOC, Medical Gas Cylinder Data Chart (reference HLC/604620/CST/0422), published at boconline.co.uk, for the nominal contents in litres and the nominal cylinder pressure in bar of every medical oxygen size used on this page: AZ 170 L and C 170 L at 137 bar, D 340 L at 137, CD 460 L at 230, ZD 605 L at 300, E 680 L at 137, F 1,360 L at 137, HX 2,300 L at 230, ZX 3,040 L at 300, G 3,400 L at 137, J 6,800 L at 137. The litres-per-bar conversion factors on this page are not reproduced from anywhere; they are those two published columns divided, size by size, here. Contents cross-checked against BOC’s own patient information leaflet for compressed medicinal oxygen, which lists the same contents for all of them, and against the NHS Business Services Authority dictionary of medicines and devices, whose BOC oxygen cylinder entries give the same pack sizes in litres (C 170, D 340, E 680, F 1,360, G 3,400, J 6,800, CD 460, HX 2,300).
- BOC, Medical oxygen: integral valve cylinders (CD, ZD, HX, ZX) instruction sheet, for the nominal contents and nominal pressures of the integral-valve sizes and for the published duration tables used as the verification in the second table. Every published duration in that table was re-derived here as nominal contents divided by flow, truncated to the whole minute, and agrees for CD, HX and ZX at 2, 4, 10 and 15 L/min and at the full, half and 25 per cent gauge levels — 36 figures checked, 36 matching. One mismatch is recorded on the page rather than smoothed over: the ZD’s own duration column corresponds to 600 L, not to the 605 L the same supplier publishes as its nominal contents.
- BOC’s Australian medical cylinder data (document code HCD130, BOC South Pacific), for the cross-country comparison: size C 490 L, CD 630 L, D 1,600 L, E 4,000 L, G 8,075 L. These are the same company’s published figures for the same letters in another country, and they are between 1.4 and 5.9 times the UK values. They are quoted here only to make the point that the letters are not international; they are not offered as presets, because this page’s preset list is UK BOC and mixing the two would be the exact error the comparison warns about.
- Not sourced, and therefore not shipped. North American practice quotes cylinder factors in litres per psi, and a set of values for sizes D, E, M, G and H circulates widely in respiratory-care teaching material. No gas supplier’s published data could be found that states both the nominal contents and the service pressure for those sizes, which is what would be needed to derive the factors the way the BOC ones were derived. Airgas publishes contents for its medical oxygen cylinders (size D approximately 15 cubic feet, about 425 L; size E approximately 24 cubic feet, about 680 L) but not the service pressure on the same product data. Rather than guess a pressure and present the quotient as a factor, this page ships no US presets: use the last option in the size list and take the contents and the service pressure off your own supplier’s chart or the cylinder label.
- Unit conversion: 1 bar is 100,000 Pa and 1 psi is 6,894.757293 Pa, so 1 bar = 14.50377377 psi. That exact figure is used, not a rounded 14.5. The ideal-gas proportionality between gauge pressure and remaining contents, and the decision to neglect both the residual atmosphere at a zero gauge reading and oxygen’s compressibility, are quantified in the FAQ above at 0.4 to 0.7 per cent — below the resolution of the dial and below the tolerance on the fill.
- Nothing on this page prescribes oxygen, suggests a flow rate or interprets a saturation. It converts a gauge reading into a duration at a flow that has already been set. Oxygen is a prescribed drug and those are clinical decisions.
CalcEngines health calculators are for education and for checking arithmetic that has already been decided elsewhere. They are not medical advice, they do not decide what to give, and they do not replace the judgement of a doctor, nurse, midwife or dietitian who knows the person in front of them. Every figure depends on the values you enter and on the assumptions stated on the page — check it against the prescription, the product label and your local policy before acting on it.