'Free' Air Con from LPG vapourisation old uk.rec.lpg topic

[SimonHobson]

1. Maybe get around the vapour pumping by drawing vapour from the bulk tank you use to fill a vehicle instead of liquid? The high pressure side of an R12 system is liquid isn't it (see liquid pumping through sight glasses on the HP side anyway)...?

It will take an awful lot of power. Think about it, you'll be vapourising the gas from the supply tank which will rapidly cool it and lower the pressure, then you'll be condensing it in the vehicle tank which will rapidly heat it and raise the pressure. Vary quickly you'll find frost round your supply tank and a high pressure differential. This, of course, exactly how the aircon system moves heat.

2. Ahh yes I was aware that all aircon systems have a lube oil mixed with the gas. Not much getting around that problem but I wonder with all the talk of a bit of oil in all LPG etc how long an aircon pump would last if operated at reduced rpm...

Well it will last <some period of time>, you just have no idea how log it'll last until it gives out.

[LPGC]

Yes I realise about the cooling / heating effects when pumping vapour..

I wonder if the aircon pump will pump liquid - The output of an AC pump can certainly be liquid. Would have thought there would be plenty of situations that the inlet side to an AC pump could contain liquid (e.g just before switching on the aircon on systems where the pump is mounted low in the engine bay).

Simon

[SimonHobson]

I wonder if the aircon pump will pump liquid

Well yes it will - it's just a small piston or roller pump. The Sanden ones on Classis RR etc is a swash plate piston pump with simple flap valves.

Would have thought there would be plenty of situations that the inlet side to an AC pump could contain liquid (e.g just before switching on the aircon on systems where the pump is mounted low in the engine bay).

There shouldn't be. At most, there will be a small quantity in the vapour return line.
But when teh system is turned off, the liquid will collect in the coolest point of the system - the temperature of the coolest point with mixed phase will set the system pressure, anywhere where the temperature is higher, any liquid will evaporate and move to the cooler section. That's normally going to be in the condenser while driving - certainly not anywhere in the engine bay ! I think you'd need a situation like this to get significant liquid in the vapour return line :
The vehicle is parked overnight in cold conditions; the next day, the sun comes up and shines on the front of the car - warming the condenser; due to the "cold soaked" lumps of metal in the engine bay, the condenser is briefly warmer and some refrigerant will migrate to the colder areas. Given the energy transferred through this mechanism, I can't see a significant amount of liquid condensing in the engine bay.

While the system is running, you should never get liquid in the vapour return line. If you do then there is something badly wrong with the system setup (such as metering valve stuck open on evaporator). Even then, it will still bring oil back to the pump so it's not the same as pumping liquid to refuel.

[LPGC]

I can see all your good points mate, I'm still thinking about this though!

If there is potential for liquid to be in the evaporator lines, then if the pump is mounted low in the engine bay the liquid will trickle down towards the pump? At what point does the gaseous refrigerant from the pump inlet become a liquid? My line of thinking - You turn on the aircon on a cold morning, engine bay and pump internals cold. Under these conditions there must be the potential for the gas to become a liquid actually inside the pump instead of only once it has cooled in the condensor...? But the pump still works and is undamaged even though it has had liquid in the compression parts of the pump? If the pump did not function as a pump under these conditions then in certain situations you could turn the AC on and it would never start to get cold, until you turned it off, waited for the engine bay and hence by heat soak, pump to warm up, and turned it back on again? If I tried to pump liquid using my air compressor I'm sure it would quickly be wrecked. The AC pump, however, must be able to cope with liquid and still function as a pump, albeit maybe not as efficiently or properly until it has warmed up? Would have thought that if AC pumps couldn't cope under such conditions we would all be aware of it being a bad idea to defrost windscreens on winter mornings with the aircon switched on, and vehicles wouldn't be designed to automatically turn on the aircon whenever the heater mode is set to windscreen / defrost?

[SimonHobson]

Yes, there is the scope for a small amount of liquid to go through the pump - but only a small amount and only for a short time. Lets take a few of those points :

Liquid in the vapour line.
Some of this will be drawn into the pump. But immediately the pressure in the line will drop and the remaining liquid will start to boil off (taking heat energy from the pipe walls) and the pump will soon see a two-phase mix. I strongly suspect that the pressure drop from drawing in the liquid through the (relatively) small orifices of the intake and valve system will also cause vapourisation. So the pump cylinders won't see all that much liquid.
Ditto if there is liquid in the cold pump.

As an aside, when I was making my hand pump (http://www.diy-lpg.co.uk/articles/files ... ed-up.html and earlier linked articles) I found that the pressure drop through inlet valves to be a significant problem. Before I made my own, I had borrowed one thanks to a chance encounter at the local forecourt pump. The guy who made it had used some small non-return valves with a significant pressure drop - with the result that it was hard to actually pump any liquid ! I took the springs out and it worked much better.

Vapour condensing in the pump.
Yes, if the pump is cold enough, some vapour may condense as it's compressed. But the act of compressing it raises the temperature of the vapour, and the time available to lose heat to the cylinder walls is limited - so the amount will be small. The pump itself will soon get quite warm - which will conduct back to the inlet end (see above about liquid in the inlet). Any condensation that takes place after the cylinder outlet valves won't matter anyway.


So yes, there is scope of the AC pump to pump a small amount of liquid, for a very brief time, when starting up. I assume the pump designers will have considered this. I cannot think of situations where it would be a significant amount, or happen for more than a second or two during pump startup. Once the pump has done a turn or two, I'd expect the amount of liquid to be minimal or none.

If you wanted to try pumping liquid then it would work. You'd wash the lubricating oil out fairly quickly, but it would work. I have no idea how long it would last without lubrication. Incidentally this is a problem with diesel systems where the injection pump is lubricated by the fuel which is a poor lubricant - made worse by modern low-sulphur diesel being even worse in that respect !
You'd need to keep the speed down I think to avoid too much pressure as liquid is forced through the outlet ports/valves, but not too slow so as not to give the liquid too much time to absorb heat and vaporise as it gets drawn through the inlet ports/valves. The pump shouldn't overheat as it'll be liquid cooled (as long as you don't run it dry), but I would expect it to wear relatively quickly.


Mind you, that's all assuming it's a piston pump with flap or poppet valves. A roller or vane pump could be a different matter as there is a period between the point where the pumping space is reducing (ie a roller/vane is on it's "compression stroke") and where the outlet port opens. Where they are pumping liquid, the ports will be arranged to only close just before (inlet) and open just after (outlet) the point of maximum volume. But where they are designed to pump vapour/gas then the outlet port closes later - to allow the gas to be compressed before it's expelled. If the port opens just after max volume - then there's a backflow of gas from the high pressure outlet port into the roller/vane space which means the pump is pumping against full outlet pressure through the whole cycle and efficiency drops.
As I've no idea what sort of pumps are used (other than the multi-cylinder Sanden types on old Range Rovers and Discoveries), I can't comment further.