Liquid Nitrogen Vaporizer Design 1

[victorpbr]

Hello,

Does anyone have experience on designing vaporizers for cryogenic applications?

I'll be using fin tubes, on the input I have liquid nitrogen at -196°C and on the output I want gaseous nitrogen somewhere between 10 to 20°C.

The mass flow is known as it is pumped by a cryogenic pump, and the pressure on the output side is going to range from 0 to 220 bar as it is used for cylinder filling.

One thing to remember is the formation of ice on it, prejudicing the heat exchange, is there any rule of thumb to deal with that?

Thank you a lot for any inputs.

 

[Caloooomi]

For a high pressure vaporiser you would normally have aluminium starfin extrusion which is lined with stainless steel tubing for the higher design pressure. You would have excess surface area on the vaporiser to deal with the reduction of surface area as the unit ices up during operation. With cylinder filling though, you are typically running in short batches which means the ice growth is not too severe. Depending on where you are, I can recommend suppliers for you

 

[EdStainless]

These are fairly standard items, let Calooomi point you to guys that already make these.
The oversizing is a bit of an art, but there are general rules that work well.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[victorpbr]

Hi guys

Thank you a lot for the answers. I am aware this is a fairly standard product and we even got a few on our plant, I also hame some drawings of them on certain models, my goal is improving my understanding about the subject, so I'm more interested in learning rather than getting one.

If someone could help on that I'd aprecciate.

 

[Caloooomi]

There's not that much to them to be honest! They can be very complex to model if you want to account for every possible variable, but it's best just to stick with general rules of thumb as at the end of the day, no one can control the weather. If you want to control the ambient conditions, e.g. with fans, heaters etc., you may as well just get a boiler and a different cryogenic heat exchanger as this would be less energy intensive. I've seen sites where they've tried to control humidity using sprinklers - it's just daft. The ice loading on the vaporiser becomes massive and they end up almost ripping out the anchor bolts and tipping over, haha.

Their main advantage is that they are using essentially "free energy" for vaporisation, with little to no operating costs. You may need to remove ice manually if in a persistently cold region, but with your outlet temperature requirements, I suspect not.

Speaking of which, an outlet temperature of 10 to 20 degC is unlikely to happen unless you are in a very warm climate, or have an additional heater downstream of the vaporiser,. You are relying on ambient weather for your heat input - if the weather outside is 5degC, then your outlet temperature can't be above this. You can have a good approach temperature, but with the ice growth it will eventually widen over time. It's why you would typically aim for a specific approach temperature after a given period of time. A brochure would normally have a model, its nominal capacity for a given fluid, which is based on a run time, fixed weather conditions and an approximate approach temperature.

For cylinder filling though, you don't really want a warm gas outlet though as the cylinder may get too hot...

 

[victorpbr]

Hi Caloooomi,

Thank you a lot for your inputs, I agree with everything you said. Btw, in the region I live the mean temperature is around 25C throughout the entire year

What I thought was using some proportion between the length of the tubes and their nominal capacity, and stabilish a relation to use on new projects, say I need to double the nominal capacity I'd simply double the length, some like a reverse engineering situation. What do you think?

Also, considering a simple model (ignoring ice formation), I could only calculate the energy flow required to vaporize and heat all the liquid nitrogen, and then calculate the equivalent length using basic heat transfer theory and apply some design factors to take into account the simplifications, do you think it is suitable?

 

[Caloooomi]

Typically, the rating of a vaporiser is based an a ratio between a nominal or standard flowrate and the surface area, e.g. Nm3/hr / m2 or SCFH/ft2. You know the mass flowrate, and you should have the surface area of the vaporiser to work out what you have. From there, if you want to increase the flowrate and are happy with the current performance, then you could look at increasing the number of extrusion, length of extrusion etc.

 

[Compositepro]

It seems that a 100 ft length of coiled copper tubing maybe all you need. You do want to use a high pressure pump for the liquid nitrogen and then vaporize it, or else you will need far more power to compress the gas. Targeting a final vapor temperature of 10C is unnecessary and will require far more surface area to superheat the nitrogen after it has been vaporized.

 

[EdStainless]

Victor,
There used to a guide published by Air Products that gave a general sizing method. I would assume that other bulk gas suppliers have similar references.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[victorpbr]

Thank you for the inputs, it is way clearer now. I'll try to find this guide you mentioned EdStainless, should give me what I need!

 

[georgeverghese]

For small vaporisation loads, an ambient vaporiser may be okay,while for larger loads, a steam or electric heated unit may be required. For an ambient finned vaporiser, would suggest doing the thermal calcs to derive an outside fin surface temp which is just above 0degC to prevent icing, with ambient temp being the minimum hourly normal at low or zero cross wind speed. Read the chapter on extended surfaces HX design in DQ Kern's book on Process Heat Transfer for calc procedure.

 

[Caloooomi]

For small vaporisation loads, an ambient vaporiser may be okay,while for larger loads, a steam or electric heated unit may be required. For an ambient finned vaporiser, would suggest doing the thermal calcs to derive an outside fin surface temp which is just above 0degC to prevent icing, with ambient temp being the minimum hourly normal at low or zero cross wind speed. Read the chapter on extended surfaces HX design in DQ Kern's book on Process Heat Transfer for calc procedure.

In theory it is possible, but the size of the ambient vaporiser to do that would be monstrous. I've read about ambient vaporisers with insulation around a stainless steel liner to achieve this, but all it is doing is adding an additional layer of resistance. Icing is not a bad thing on an ambient, it just shows that the unit is working! With cylinder filling, it is intermittent use typically depending on the size of the buffer vessel. The reciprocating pump would run for a fixed period time, say 10 - 15 minutes, then shut off for 10 minutes or so while the gas in the buffer is being used. In a place where it is averages 25 degC ambient temperature, with this sort of operation any ice formation that does occur would just melt off quickly. The fouling would be more like fresh fluffy snow, rather than hard ice. As long as they have drainage under the vaporiser, there shouldn't be any concerns over small amounts of icing.

 

[EdStainless]

I have seen systems in dry climates where they have catch trays set up to collect the water off of vaporizers.
Ours often ice ~40% of the way if we go to full flow on a nice hot humid day. They are sized for that.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Caloooomi]

Yup. I've also specified drip trays in LHe installations in order to catch the liquefied air!

 

[georgeverghese]

The liquid air purge finned ambient vaporiser in the plant I used to work at always had a thick layer of hard ice all over it, and I often wondered how it worked. We must have overloaded it with liquid air I suppose. It didnt really bother us - the exit from this purge vaporiser lead into a large liquid air disposal tank and most likely this tank did what the vaporiser could not.

 

[Caloooomi]

Was it an ASU? I've seen disposal vaporisers were you dump the liquid from an ASU through a vaporiser and vent it later. Typically these are designed just to vaporise and not to add too much superheat in the disposal stream

 

[georgeverghese]

Yes, this was in an ASU - this vaporiser was continously fed with a purge LO2 stream from the reboiler in the upper - LP column operating at about 30-40kpag. Feed temp would have been approx -180degC.