Printable View
I dont think I have ever seen the sightglass clear, there are bubbles rushing up the center if it and its about 95% full.
The bottom of the compressor is cool and the side is cold where the suction line enters. The top is hot but not too hot, I can hold my hand there for aslong as I want.
The load should be around 120 Watts
Feel free to jump in. We are stumbling around in the dark also. :D
At the location of the sight glass, the temperature/pressure is such that the R23 cannot possibly condense into liquid. Thus a clear sight glass tells us that there is complete separation of the two refrigerants.
Each evaporator must flow enough refrigerant through to handle its heat load. We know this is happening if its exit temperature is very cold, just slight above its evaporating temperature (SST).
This is not happening in the R290 evaporator.
This is happening too easily in the R23 evaporator, thus we can lengthen it, which will reduce its temperature while still handling its heat load.
At this point, it looks like the R23 cap tube is too short, and the R290 cap tube is too long.
I have no idea what the proportions of anything should be at this point. Autocascades are the new kid on the block, proportions are proprietary and super secret. On commercial autocascade systems, you evacuate the system, and the manufacturer sends you a premixed refrigerant charge for the system, for a mere $1000 or so.
The way I see it. You need enough R23 to remove the heat input at the evaporator and enough R290 to condense the R23. If you have too little R290 then you condense less R23 thus evaporator temperatures rise until the specific heat of the R23 is enough to cope with the load on the evaporator.
We might try increasing the airflow through the condenser to both reduce high side pressure and improve separation. I would assume that any bubbles in the sightglass are a result of incomplete separation, and are making this more difficult. I believe this is R23 vapor going through the R290 evaporator.
If the separator is very warm, a fan might help there also.
Got a deskfan blowing over compressor, separator and discharge line. Condensor air out temps are slowly dropping. The separator is warm to the touch but only slightly. The R290 has got to be more than 30ºC subcooled. Just one more thing, the R23 evap exit temp is taken just 2inches from the evap exit and the delta-T is around 15ºC from SST. Just wanted to run that by.
Keep in mind that the mixed refrigerants in the condenser completely distorts the pressure temperature relationship. The actual R290 SCT is probably a couple degrees above condenser air out temp, thus the R290 subcooling is far less than you might think.
If the r290 is exiting its evaporator at a few degrees above ambient, dosnt that mean it isnt evaporating enough inside its evaporator i.e. there is too much charge of r290?
Regards
John.
More specifically, not enough refrigerant flowing through the evaporator, which could mean insufficient R290 at the entrance to the cap tube, and/or cap tube too restrictive. Insufficient R290 at the cap tube entrance could indeed be R290 undercharge, not overcharge.Quote:
If the r290 is exiting its evaporator at a few degrees above ambient, dosnt that mean it isnt evaporating enough inside its evaporator i.e. there is too much charge of r290?
Given the suction pressure, if R290 refrigerant were flooding completely through the evaporator in liquid form, the bottom of the compressor would be a ball of ice.
Gary, shall I try and get rid of all the bubbles. I will continue to add R290, but the more I add the higher my LP is getting. Is there a value for one of the system parameters i should not pass when adding the R290 or can I keep adding till the bubbles go?
But I dont understand sir:D
Could you take me round Russels system a component at a time?
We start with a charge of r23 and r290 in the compressor, both gasses at ambient switch on the compressor and both gasses are compressed and the heat is increased too. And then?
Regards
John.
Lets see,
Once the compressor is turned on, the R290 starts to condense in the air cooled condensor. The R23 stays as a gas. This fluid flows to the phase separator where the gas is taken off the top and the liquid R290 is taken off the bottom.
The R290 is then metered to the heat exchanger via the cap tube. The R290 evaporates in the heat exchanger and this condenses the R23. Note the refrigerants are separated by the heat exchanger walls. The liquid R23 is now metered to the evaporator where it removes the heat load from the cpu. The R23 gas is now returned to the compressor. Before it gets there it meets and mixes with the R290 gas from the heat exchanger and then enters the compressor.
Thus the cycle begins again.
Just a thought: Russell, are you absolutely certain that you are measuring the R290 exit and not the R23 entrance?
Good question Gary :D
I've doubled checked this many times, checked that the sensor is still in place and even felt the pipes with my hand. They are warm, as the pipe gets closer to the R23 junction the pipe gets colder as expected.
One other thing. If the R290 is evaporating at -2ºC and the compressor inlet is colder than that, will I be getting liquid slugging? I dont know how to tell.
Thanks Russel
So if there was too much liquid r290 flowing threw its evap than it wouldnt be able to evaporate and this would have the same effect as too little r290? And if this is the case how do you know which is the correct determination?
Regards
John.
By measuring the evaporator exit temperature and comparing it to the saturated temperature we get an idication of what is going on.
Im my system the exit temperature is 28ºC above the evaporating temperature for the system pressure. Thus if there was any liquid at that point in the system, there would be enough heat to boil it. If there was too much liquid then the exit temperature would be alot closer to the saturated temperature.
Okay,
So you have been increasing the r290 charge and shortening the r290 cap tube, and this will increase the flow rate of the r290 through the evaporator?
Each time you increase the flow rate of the r290 the cpu temp increases, so that means the r23 isnt being cooled as much as it was with a slower r290 flow. How could that be?
Regards
John.
I am not sure we can trust the R290 SST with a stream of R23 bubbles mixed in. We need a clear sightglass.
Somehow I get the feeling that we are missing something here. Possibly with complete separation, we would not have needed to shorten the cap tube? Can it be that the stream of bubbles is what is throwing the whole thing off?
Have you tried increasing the condenser airflow?
The condensor fan is at maximum speed. Will putting a desk fan infront of the condesor help? Is that the best place for it?
The fan is setup infront of it. I can open a window and bring air temps down. Its around 0ºC outside.
Just noticed the air temp in is around 24ºC and out temp is 30ºC. Ive opened the window and im bringing the temperatures down. Im gonna nail this beast if its the last thing i do. :mad:
How does the phase seperator work? I assume that the liquid r290 settles to the bottom under gravity and is thus filtered off the bottom with the r23 gas filtered off at the top.
If the correct ratio of r23 to r290 isnt charged wouldnt the r290 be saturated with r23 vapour and then eventually form large bubbles of r23 vapour some where down the line?
Regards
John.
Its just a pipe where the mix enters and there are 2 exits, one at the top and one at the bottom. Gravity does its job and the 2 phases separate.
Ive been adding more R290 Gary and im still seeing bubbles. The bottom of the compressor is starting to get cold now. With the extra fan on the condensor the HP is dropping.
Also the r290 evap(heat exchanger?) works by dropping the pressure f the r290 allowing it to evaporate and hence remove heat from the r23? If there are large bubbles of r23 expanding in the heat exchanger wouldnt that force out the r290 before it had done its work?
Sorry for all the questions guys:(
Regards
John.
Somehow I keep thinking that if we add enough R290, the bubbles must disappear, but I'm not entirely sure of this.
Stop right there. Shut off the machine. Our problem is separation.Quote:
The bottom of the compressor is starting to get cold now.