Injector instead of TX Valve

[charlie n]

I'm working on a design of a 400 kW water chiller using R407c. The chiller has to be as absolutely simple as possible and it has to be explosion proof. It will have to work in a coal mine. I have an idea to use an injector instead of a thermal expansion valve. (Phillips in the US is one of the better known suppliers)
Reason: more simple and one less thing to break.
Anybody have any thoughts on this?

[RANGER1]

Using a injector would require critical charge with a surge drum if I understand it correctly .
Then you have to worry about oil return .
Some pluses and minuses , I'd stick with TX valve myself or something you are most familiar with .
Its probably pretty uncomfortable down there at the best of time .
Can you get a TX valve with that capacity ?
Is the water quality OK for both condensor / evaporator ?

[cool runings]

I'm working on a design of a 400 kW water chiller using R407c. The chiller has to be as absolutely simple as possible and it has to be explosion proof. It will have to work in a coal mine. I have an idea to use an injector instead of a thermal expansion valve. (Phillips in the US is one of the better known suppliers)
Reason: more simple and one less thing to break.
Anybody have any thoughts on this?


there is nothing special about what you call an injector.

In the crudest form a capilary is what you describe, but possibly fragile.

The restrictor is only that, a restrictor to restrict flow and create a pressure difference. The only thing with that type of metering device is the charge is critical and it will not deal with differance in water temp very well. That is why thermostatic expansion valves are used. They give a lot of flexibility when loads vary.

Why do you think a TEV is prone to failure?
TEV's are as reliable as anything else you put on the machine.

cool runnings.

.

[Lowrider]

why build one when u can buy a chiller and just do some alt's to get it explosion safe. I have done some in the past for LNG-plants. the only thing to do is get the places where there is a chance of sparks on an over pressure.
so u may want to alter the control and power supply panel to be able to put it on overpressure and maybe get the feedboxes of the compressors on overpressure. Move the LP and HP switch inside the overpressurized panel. Easy, simple and reliable.

[charlie n]

Thanks for the comments,
- we build instead of buy because that's what my company does.
- overpressure won't meet the Mining code requirements in China where we're using these.
- I don't thing a TX valve is prone to failure. These units get dragged around in a mine by a tractor & capillary tube breakage is a possibility. We can't braze in a new valve or tube in the mine atmosphere so a broken tube means a week or more downtime to bring the chiller up to the surface.
- Critical charge & a liquid separator is no big deal and I'll gain efficiency with a fully wetted evaporator surface.
- I can also omit the liquid solenoid valve this way & further simplify the system. The code approved solenoid valve will cost almost the same as the liquid separator so I'm actually not adding costs by going this way.
More comments or ideas are welcome.

[night_marios]

hello folks,i have to agree with the others that txv is beter when the loads vary!Prevents liguid refrigerant get back to the compressor.Today manufacters uses more acurate control systems to control the variety of loads like electronic expansions valve.If your load is standard then you can use "injector" as you said or beter "accurator"{CARRIER uses this term}
but it will be not proper if your chiller uses steper control!Thanks a lot for your consider, Marios

[michaelm]

B”H<O:p</O:p

Injector is an expensive pump and you will gain the efficiency. The only think, that injectors do not like high pressure drop in the evaporator. A receiver will be placed over your evaporator do you have the space constrain? Phillips has an arrangement for the oil return. Do not use any reducers after the injector.

[charlie n]

Thanks for the continuing discussion. The load is constant at 100% or more. I have space for the liquid separator/receiver. The evaporator will be a simple S & T water chiller so pressure drop shouldn't be a problem. I'll work with the Philips design for oil return.

[RANGER1]

charlie n ,
I have only seen them used on majonnier or baudelot chillers .

How would you pipe it up on a S&T chiller ?
Does phillips have an example that you know of so I can check it out .

Is refrigerant on shell or tube side .

[Magoo]

Something out of left feild. What about applying pneumatics, generally mines have truck loads of compressed air around the place, the whole thing could be air operated, ball valves etc., no electricity required.
Don't know if there are big enough air driven motors for open drive compressors.
Just thinking outside the circle again.
magoo

[michaelm]

B”H

If the refrigerant in the shell it could be sprayed on the tube bundle. Speak to Phillips I know that one of the application it is done and as far as I know it is efficient. I use to know who has this arrangement but it is too many ago. Good luck.

[charlie n]

All good ideas. I don't think I can get a pneumatic drive motor but I could probably replace most of the other electrics. I'll be contacting Philips this week about the application. I once saw a DX S&T chiller with an injector where the TX valve should have been but lets see where the spray chiller idea takes me.

[Gingerair]

Surely an expansion valve for that capacity would not come in one piece. Once the body was fitted the internals could be changed in the unlikely event there was a failure.
Otherwise flange the pipework either side of the valve & even add shut off valves..
As Ranger1 said, can you even get a tx that size, would they not be electronic at your capacity ?
If you went down the route of an electronic expansion valve this could also shut down on demand to act like a liquid line solenoid valve, giving you a pumpdown facility if required..

[Gingerair]

http://www.ra.danfoss.com/TechnicalInfo/Literature/Manuals/01/RD1AZ302.pdf

[charlie n]

Gingerair, I know this PHTX & yes, on the surface an electronic expansion valve would be an excellent idea but it will be very difficult and expensive to get this through the approval process to use it in the coal mine. I don't really need a pump down if I use an injector. The refrigerant charge is resident in the low side separator during the off cycle.

[RANGER1]

The simplest level control that is very reliable is high side Witt or Hansen float valve .
Also has no electrics and don't need a liquid reciever , just a drop leg to connect it too .

[Gingerair]

Would you not consider changing your gas to 134a ?
Could give you more options..

[charlie n]

134a is worth considering.