Do refrigerants passing through a venturi act like steam?

[Carno]

TL;DR Summary

Can injector technology be used with refrigerants to avoid compressors for air con/refrigeration

To get water into a pressurised steam engine boiler, an injector passes high pressure steam through a converging/diverging nozzle, sucking cold water in at the throat. The resulting steam/water mix is at a higher pressure, but lower temperature, and so is able to return to the boiler. Heat is needed to restore the system to its original pressure and temperature.

My question is whether this system would work with a high latent heat refrigerant such as ammonia?

If so, high pressure vapour from a gas cylinder passing through the nozzle could draw in liquid gas at the throat, the liquid gas partially expanded so very cold, the resulting mix being returned to the cylinder, wetter and cooler, but at high enough pressure to re-enter the cylinder.

If this worked, the temperatures involved would be low, so that heat input required to restore the system could come from ambient air/water.

The resulting effectively cooled ambient air/water might be cold enough for air con/refrigeration etc.

It would need a small pump to get the process running, but possibly needing less power than a conventional Carnot cycle compressor?

 

[russ_watters]

To get water into a pressurised steam engine boiler, an injector passes high pressure steam through a converging/diverging nozzle, sucking cold water in at the throat. The resulting steam/water mix is at a higher pressure, but lower temperature, and so is able to return to the boiler. Heat is needed to restore the system to its original pressure and temperature.

I'd like to see a diagram/source for that, because it doesn't sound right; a C-D nozzle utilizes a large pressure drop to work.

Anyway, a pump and a compressor are similar devices working with different phases. The reason it's a compressor in an AC unit is you are compressing a gas before condensing it. For it to be a pump, you'd condense first, which loses the heat rejection advantage of condensing at a higher temperature.

Either way, diagramming this out would help.

 

[Carno]

Thank you for your reply. Appreciate your interest. The Giffard injector intrigues me because the ability to 'pump' water from atmospheric pressure into a high pressure boiler with no apparent power input seems counter-intuitive. Here are a couple of links with diagrams/descriptions of how it works

https://en.citizendium.org/wiki/Injector
https://en.wikipedia.org/wiki/Injector

-- so they don't defy any laws as they require energy input in the form of heat.

My question is instead of superheated steam leaving a boiler at maybe 140C, and returning at 120C (and therefore needing a lump of coal to restore the temperature), could a jet of high pressure refrigerant at say ambient of 20C leave a gas cylinder and return, carrying partially expanded liquid gas from the same cylinder at -20C? (ie as the Giffard Injector but much colder). If so, heat from ambient air (warm air in a house that needs air con etc) could maybe provide that heat to restore the initial temperature?

The high latent heat of steam is apparently quite important to the process, and not all refrigerants have this property, but ammonia does.

Also the Giffard injector whilst free running once started, requires feed water to be bled off to start the flow, and wouldn't be practical with a refrigerant, which is why maybe a small pump might be needed to start the flow, which could perhaps be switched off/by passed if the system worked as the Giffard Injector?

 

[russ_watters]

I really don't understand what you are trying to do there. Is it supposed to be an air conditioning unit? Where's the heat rejection? Maybe more labels would help: temperatures, phases, and processes.

For the injector in a boiler; wasn't that an open process, with some steam discarded?

 

[Carno]

Thanks for reply, goal would be free air conditioning and refrigeration.
The steam injector pumps cold water into a high pressure boiler with no work input other than heat. The heat comes from burning fossil fuels
My query is whether the same principle could be applied in a closed system using high pressure refrigerant vapour from a gas cylinder to pump cold, partially expanded refrigerant vapour from the same cylinder back into the cylinder. Heat input would still be needed, but as the temperatures are very much lower, this time the heat input could be from ambient air (like a warm room or refrigerator).

You're right about the steam injector, to start the system some water needs to be released - then the system runs continuously. Not practical with refrigerant clearly, I wonder if a small pump would force the process into cycling which could then be switched off and bypassed.

(Apologies for crude sketches attached)

 

[berkeman]

(Apologies for crude sketches attached)

They came out sideways and kind of dim and hard to read. Do you have access to a document scanner (like an all-in-1 printer)?

 

[russ_watters]

Attached, rotated and cropped.

 

[russ_watters]

Ok, I think I get what you are after. There's a few things you are missing though:

1. Your diagram of the steam injector contains a contradiction in that you are both returning all the steam back to the boiler and adding make-up water. You only need make-up water if your cycle is open (you are losing steam). Maybe that is just omitted from the diagram as a separate loop? Regardless, I doubt that for a mostly closed/slightly open cycle there's an advantage to be gained over a pump to get the condensate back to the boiler. This isn't exactly pertinent to what you are after....though it does play a role...

2. Your proposed refrigeration cycle is missing heat rejection. It doesn't seem to show the main purpose of a refrigeration cycle: to remove and reject heat. Whereas a steam engine produces a thermal (temperature) gradient and then harnesses it to create mechanical work, a refrigeration cycle does the opposite: it uses mechanical work to create a thermal gradient to reject heat. In other words, without a pump you don't even have a thermal gradient to harness. I realize it isn't intentional, but this violates the 2nd Law of Thermodynamics. More specifically, you start with a liquid/vapor mix at 0C, then go through the evaporator, converting the liquid to superheated vapor at 20C while still having liquid, which is a contradiction. Then you expand this liquid to -20C and still a liquid, which is another contradiction (it would be low pressure vapor). Here you'd need another evaporator, or, rather, that's where the evaporator would really go. What you are missing on the left side is that you absolutely need the pump(compressor), and you'd need a condenser to convert the vapor to liquid. Basically, you've reversed where you'd need/have the liquid and vapor.

 

[Carno]

Thanks for amending the sketches, I'm travelling at the moment and don't have access to a ruler let alone a scanner. Thanks for comments above, I'll read and absorb when I get back,