Maximum withdrawal rate of LPG from cylinders @ 15 degrees C

[Oxyset Services]

thread135-111334 After reading this thread I wanted to add some basic information and ask for some help.
Pressure and contents in industrial gas cylinders is always calculated at 15 degrees C by the Industrial gas companies.
I am trying to work out the maximum withdrawal rate of LPG from different sized cylinders when used with oxygen/LPG equipment for cutting, brazing, and heating.
Usually for heating and cutting applications I would suggest the use of 45kg LPG cylinders with a "G" sized oxygen cylinder, but I have heard of people using 9kg and 18 kg LPG cylinders.
I would think this is dangerous because a lack of LPG pressure and flow could cause a flashback due to a lack of gas pressure and volume to keep the flame at the end of the tip or nozzle.
Here in Australia most cutting and heating tips use 100-150 kPa of LPG and anywhere between 200-1000 kPa of oxygen, depending on the size of the tip or nozzle.
From information gleaned from an LPG supplier here in Australia I have found out the maximum withdrawal rate from a 45kg cylinder is about 56.6 litres per minute, I am trying to find out the maximum withdrawal from a 9kg & 18kg cylinder at 15 degrees C.
Cylinders for LPG here go 4kg, 9kg, 18kg, 45kg, 190kg

The reason I ask this is due to the issue of withdrawal of acetylene from cylinders which is limited to 1/7th of the cylinder content per hour for intermittent use and 1/10th per hour for continuous use.
All oxygen/acetylene heating torches on the Aussie market grossly exceed the maximum withdrawal rate and often cause flashbacks, this is why I always recommend oxygen/LPG for heating over oxygen/acetylene even though the flame is colder, 2,850 degrees C v's 3,160 degrees C the heating tips for oxygen/LPG are much larger so the calorific energy available is much greater.
Can anyone out there help?

 

[LittleInch]

As the previous post says it's all about the heat input.

One source I found said max rate was

47 kg - 2.37 kg/h
19 - 1.32
13 - 1.05

[EDIT] - so for the 47kg one this equates to about 22l per minute so about half of your flow. But it is highly dependant on the air temp and also how much fluid is in the cylinder. The fuller the cylinder the more liquid area there is for heat transfer so the higher the rate of vaporisation.

It's not clear what temperature this was based on though.

But it sounds right - the smaller the tank the greater the surface are to volume is and so the greater the heat transfer per kg of fluid.

So you need to define the air temp and the heat input if it's external to find out max flow.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

Info / picture on the net says there is some kind of flame / flashback arrestor incorporated into the welding torch assembly, but it is not described nor is it shown as a discrete device in this torch.
What do you know about this device?

 

[LittleInch]

You could always add in a heating blanket for your cylinder though.e.g.

https://shop.metallisation.com/products/heating-blanket-for-47kg-propane-cylinders.aspx

https://www.ameritempgroup.com/stor...anket_(Propane),_420_lb,_Charcoal_Gray.html#/

They're not cheap but you will get a reliable flow.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Oxyset Services]

Let me see if I can answer everyone's questions and responses.
For my job I need to know the flow rates of the tips and nozzles of the gas equipment I am supplying and ensure the gas supply system and cylinders can give the required amount of gas, both in pressure and in flow.
Acetylene is simple, I have the details written in my thread at the top.
Oxygen is simple as it's a gaseous gas in the cylinders not a liquid.
I have the specifications of the regulators, hoses, flashback arrestors, blowpipes, cutting attachments, mixers and tips or nozzles, but the maximum withdrawal rates of LPG from the different sized cylinders is difficult to find.
The reason I need this information is if there is not enough gas pressure and flow from a cylinder, it could cause a flashback.
Remember a flashback arrestor won't stop you having a flashback, it will only stop the flame going back into the down stream gas equipment.
I need to be sure the cylinder I recommend is big enough to supply what is needed to the different tips and nozzles.
The ambient temperature for gases should always be measured and calculated at 15 degrees C

 

[chicopee]

Vaporizers are normally installed with propane tanks in the Northern States since ambient temperatures during cooler months will not vaporize liquid propane fast enough to meet the demands of the heat input of the equipment that burns propane. The size of vaporizers are based on the required heat input and to a lesser degree pressure losses and heat loss thru the piping.

 

[georgeverghese]

We can possibly infer that withdrawal rate is directly proportional to external surface area of the holding vessel (through which ambient heat leaks in, and latent heat of vaporisation in kJ/kg is more or less constant for a given mix of LPG, which is mostly propane). So if withdrawal rate permissible for the base case 45kg tank is Mo in kg/hr, and its external surface area is Ao, we infer that for any other tank with surface area A, that

M = (A/Ao) x Mo

 

[Compositepro]

It is the wetted surface area inside the tank that limits the heat transfer. This surface area gets rather small when the tank is near empty. This is why you do not see figures for maximal withdrawal rates.

 

[georgeverghese]

@compositepro
Theoretically, you are right. But wall resistance is much lower than inside htc, and hence inside htc is one of the limiting resistances to heat transfer, in addition to external htc. Of these 2 limiting resistances, the external natural convection htc would be much lower than the internal boiling htc. So I've approximated that heat transfer into the liquid from ambient is relatively constant regardless of level in the tank. This approximation helps to enable the linear relationship suggested. An approximation to reality is better than not knowing at all.

 

[Compositepro]

Only the heat that transfers through the tank wall directly into the liquid contributes to vaporization. Heat transfer above the liquid level may help warm the vapor a little, but almost none of it will get to the liquid.

 

[LittleInch]

The other variable is length of time for the flow. You will be able to get a very high rate as the liquid cools but only for a few minutes. Maybe that's all the OP wants it for?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MJCronin]

Nobody wants to talk about "choked flow" and how the inside size of the nozzle restricts maximum flow ?

MJCronin
Sr. Process Engineer

 

[LittleInch]

True - the pressure regulator max flow is also another variable...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[georgeverghese]

Both Simsci Pro II and Aspentech-Hysis process simulation programs have the batch depressure routine, which can model this constant flow operation to match input heat leak from ambient. You can also include the regulator in the model. Talk to an oil/gas engineering contractor or consultant - most of them would have either or both of these simulation programs.

 

[crshears]

The Electrical Maintenance Departments in my utility commonly convert empty 45 gallon drums into water baths for sulphur hexafluoride cylinders; the baths have thermostatically controlled electric heater belts wrapped around them to maintain the water temperature in the bath at the maximum permissible value so as to facilitate gas insulated switchgear compartment filling, as some of the compartments are HUGE by virtue of their length.

See here as an example:

https://www.google.com/maps/place/C...2980dbadfee0c6!8m2!3d43.7455004!4d-79.6142841

There are really long bus ducts that run from the brown roofed 230 kV GIS building to the right and then way, way up to where the line terminals' air-to-gas bushings are. These take multiple cylinders of SF[sub]6[/sub] gas to fill.

In a similar vein, we electrical operators used to keep our handheld propane torches inside the control building during the winter months, and take them out to our vehicles when going out on a switching run; these torches were invaluable for thawing frozen switchgear locks. One soon learned the value of this the first time one forgot to store them inside; one either had to thaw locks with one's bare hands, or warm the torch cylinder with one's bare hands to scare anything more than a very sluggish flame out of the torch due to near-nonexistent propane vapourization . . .

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]