OT: CO2 evaporator or vaporizer

Meter the liquid with a liquid pressure regulator.

LLoyd

I expect something that simple would not work. The pressure differential doesn't reflect the mass flow, because of the lagged phase change on the low side. You would draw a volume of gas, and it would be replaced by almost that volume of liquid. As the liquid boiled off into gas, you have an overpressure condition (and let's hope a good blow-off overpressure device).

The resemblance to refrigeration system stability suggests that one needs something like a combination of TXV and pressure differential, possibly with an accumulator to damp the instability from phase change.

Why do you need to meter the liquid input? Doesn't the 100 psi output meter the input automatically?

I'd use a coiled line like you suggested, and regulate the output with a paintball regulator.

Because if too much liquid feeds into confinement, you will reach the saturation vapor pressure of CO2 which is upwards of 1000 psi at these temperatures. The rest of the system is max 300 psi, and we'd like to plumb for that maximum pressure (with overpressure reliefs).

That and you can't have the possibility of liquid slugs flowing into the terminal regulator.

That's still not a problem, simply double regulate. Warm your CO2 with a copper or aluminum coil, then regulate down to 300 psi. You can't have liquid CO2 at room temperature if your pressure is that low. Then, you can regulate again down to your 100 psi. A couple hundred bucks worth of regulators and lines, and you're ready to go.

I don't follow you.

*Something* has to regulate the mass of liquid entering the vaporizer so that it *equals* the mass of the gas exiting. Given that the volume of these equal masses differ by a factor of 500, you can't just use pressure regulators. Without this regulation, the vaporizer will either flood or starve.

It's like an A/C evaporator. You need an orifice (crude open-loop control) or TXV (crude feedback control) to approximate the equal-mass regulation. But this is more complicated, because the output flow switches on and off.

starbolin

When I worked in homecare the plant that filled our cylinders took liquid O2 from a supply vessel, ran it through a huge vaporizer and then used a piston pump to compress the gaseous O2 to whatever pressure the plant needed. Hospitals use a similar system except the pump is replaced by a holding tank from which the O2 is dispensed. In both cases, the liquid output isn't tightly controlled- the gas is.

HTH- Carl

In a bottle, at room temperature, liquid CO2 is about 850 psi. If you draw gas off the top of the bottle, and release it into the air, then shut the valve, the gas is still at 850 psi if the bottle is still at room temperature. That's because the liquid boils at room temperature. It will continue to boil until the gas reaches a pressure that causes it to stop. This act of boiling causes the CO2 to get very cold, which drops the pressure in the tank.

If you can add enough heat to the tank to keep it at room temperature (higher temps can be dangerous), and aren't using a siphon tank, you just get gas out of the tank. If you drop the pressure to 300 psi, the temperature of the CO2 needs to be about -1 deg F. before you get back to a liquid. Therefore, if you insure that your tank and plumbing stay at room temperature, there cannot be liquid in your lines at 300 psi. Obviously 100 psi would require an even colder temperature to return the CO2 to it's liquid state. If you control the temperature and pressure, you control the liquid. Just remember, at each place that you reduce the pressure (the tank, and at each regulator) you also reduce temperature. Long coils of tubing, possibly running through warm water are your friend.

It seems a bit glib to just say "insure". That's what a vaporizer does.

Somehow 2700 BTUs have to go into 20 lbs of liquid CO2 to make it vaporize each hour. Long coils of tubing won't do that.

The tank is insulated and refrigerated. You do not draw vapor as you suggest for bulk applications. You draw liquid and vaporize it external to the tank. This is not a little trickle of gas for a MIG welder. We're talking multiple cubic feet per minute. And whatever volume of liquid you draw from the cryo tank, must be replaced by a backfeed of warm gas through another circuit.

I know finned-tube vaporizers are used on LN2 systems. Even there you have to have a pair that you switch between via a manifold, because they periodically frost up and quit working, by design.

Industrial CO2 vaporizers, which start at about 10 times this scale, are complicated gadgets involving PLCs and the like.

On what principle is the liquid prevented from flooding through the system?

Gravity.

Make the liquid up-flow from the tank. Make the high-pressure gas side of the system as short as possible.

LLoyd

The trick, here, is that the vaporizer is at a temperature where liquid cannot exist at these relatively low pressures. As soon as the regulator starts to admit liquid to the vaporizer, some of it flashes to gas and raises the pressure in the vaporizer. The vaporizer will be filled with gas almost 100% unless you draw so much CO2 through it to freeze the whole thing. As soon as the first spurt flashes to gas, the regulator stops admitting liquid to the vaporizer. The only way liquid can remain at 300 PSI is under very cold temperatures. That's why the tank has to be refrigerated. In fact, you may not need ANY regulator at all on the input to the vaporizer! You only need a gas regulator on the outlet to get your 100 PSI. You would need a cryogenic-rated valve on the inlet to prevent a surge of cold liquid rushing in when you first connect to the tank.

Jon

Yeah, but nothing regulates the size of this spurt. When it flashes to gas, it will pressurize to upwards of the saturation vapor pressure (900+ psi) at that warm temperature, and vent out the overpressure relief.

Let's say you had 20 feet of 1/2" refrigeration copper tubing coiled in a barrel of warm water for a vaporizer. Tubing volume is all of 0.018 cu ft. Liquid to vapor expansion is about 500-to-1, so you need an initial "spurt" of 0.000035 cu ft, or about 1 cc.

Now tell me, when you crack that valve with 300 psi liquid behind it, feeding into 20 ft of empty tube, do you really think that 1 cc of liquid is all that is going in?

I'm thinking tiny orifice followed by a check valve. Orifice sized to pass the liquid at 20 lbs/hour from a differential pressure of 50 or 100 psi.

This describes a "liquid cylinder regulator", which despite its name does not regulate liquid, but the vapor side of a liquid vessel, like a low- pressure Dewar tank. In other words, it's designed for input pressure about 300 psi or less, versus 1000 psi for high-pressure cylinders.

1) The liquid is coarsely throttled and there is most likely a liquid trap in the line. 2) The vaporizer is huge- it is outdoors and is about 8 feet square and 15 feet tall. It is made up of hundreds of vertical fin-tube arrays. That is a lot of heating area. 3) If you have enough heating area in your vaporizer, the outlet gas regulator should be enough control so liquid can't surge through the system.

I can't speak with absolute authority on CO2, but an liquid oxygen system that meets your flow rates would be very expensive. The homecare company I worked for investigated having a small liquid-to-gas setup installed on site (to fill portable tanks) and when all the costs were in, it was clearly cheaper to use large gas cylinders on staggered manifold (later, we just paid a company to fill the small tanks for us). The site work alone was a fortune- cryogenic tanks have to be installed on concrete pads with security fencing.

Can't you carbonate the product with a cylinders?

-Carl

Unlike O2, high-pressure CO2 cylinders are vapor-over-liquid, like the Dewar tanks. Same problem of requiring heat input for vaporization.

Thanks for the comments regarding the oxygen version via finned tubes.

It seems like you must choose between a bulky apparatus of simple operating principle, or a compact one with complex controls.

Oh- I didn't know that.

Isn't that always the way?

-Carl