Non condensables??

[SteinarN]

I guess it depends on which ones you are reading.

You can't make broad statements and expect them to hold true for every case.



Not all the time. I have modified my answer in this edit.

SteinarN is getting to the heart of the problem. Where the non-condensables appear also depends on how the condensers are piped. If you have a liquid seal (P-trap) in the condenser drain line the non-condensables get trapped on top of the liquid seal. If the liquid seal disappears or the drain line is not trapped at all, the non-condensbales do flow down into the receiver. That makes a lot more sense!

Thank you for some support

I felt i was taken on heavily.

I agree in what you said about liquid seal. I'm not into the industrial stuff. I dont know anything about the construction of the condensers there. I'm talking about normal comercial fin and tube condensers.

Older condensers had larger diameter tubes and consequently lower pressure drop and lower refrigerant speed in the tubes. If the ricers? also had a large diameter, then i can imagine some non condensibles accumulate in the condenser.

However, with new modern designed high efficiency condensers the tubes has a much smaller diameter and consequently larger refrigerant speed through the tubes. Also the ricers is smaller. This makes it next to impossible for any non condensables to accumulate in the condenser. When i design the liqiud line i aim for a refrigerant speed of roughly 1,5m/s. With such speed i doubt any non condensables can avoid beeing swept along to the receiver regardless of any traps or the direction of the liqiud pipe.

[Gary]

Obviously, the subcooling cannot exceed the TD, but rarely are TD's that low. The exception does not invalidate the principle.

[SteinarN]

Obviously, the subcooling cannot exceed the TD, but rarely are TD's that low. The exception does not invalidate the principle.

It was not you Gary that i defended my self against. It was Refrigerologist.

But that said, when i design a system, i design it with no more than 10K dt condensing at full load summer time. The additional cost for a larger condenser is saved back in a couple years due to lower power consumption to the compressor. In some cases i can get away with the next smaller compressor also. The actual load at night and with low humidity for instance is easily only 50% of full design load. Thereby a dt of only 5K.

[Gary]

Just to make sure we are on the same page:

Condenser dT is air out minus air in.

Condenser TD is saturated condensing temperature (SCT) minus air in.

[SteinarN]

Just to make sure we are on the same page:

Condenser dT is air out minus air in.

Condenser TD is saturated condensing temperature (SCT) minus air in.

Ahhh. Seems that we have some different abbreviations on oposite sides of the Atlantic

I'm talking about TD. SCT minus air in.

[Gary]

It was not you Gary that i defended my self against. It was Refrigerologist.

You should not feel that you are under attack. We are simply discussing an important aspect of our trade. Don't take it personally.

[Gary]

There seems to be a great deal of confusion on both sides of the pond.

Throughout the scientific world, a delta-T (dT) is a change in temperature of a single substance or flow of substance, while a TD is a temperature difference between two different substances or flows of substances.

[SteinarN]

You should not feel that you are under attack. We are simply discussing an important aspect of our trade. Don't take it personally.

Maybe i owe you to attack me some time

But i did actually feel i had to defend my self against Refrigerologist. But now i have don it

And yes, it is very interesting to discuss various aspects of our trade. I apreciate very much when i get substantiated answers. I do get respect (not to much tho) for people with a high grade of knowlegde. I know i maybe appear stubborn sometimes, but i hope you guys in here can take me down if i get to stubborn

[SteinarN]

There seems to be a great deal of confusion on both sides of the pond.

Throughout the scientific world, a delta-T (dT) is a change in temperature of a single substance or flow of substance, while a TD is a temperature difference between two different substances or flows of substances.

Hmmmm. I use delta T as a difference in temp of a single substance as you says. But i think that delta T is also used in catalogs here in Norway for the difference of SCT and entering air of condensers and evaporators also. Have to have a look tomorrow for sure.

[Gary]

I have several systems with floating head running with condenser dt as low as 5K in low load condition. What is your recomended subcooling in such a case?

In low ambient conditions, if there is no fan control and the SCT is below 90F/32C, I block off part of the condenser, raising the SCT to about 100F/38C before checking subcooling and superheat. I would do the same on a floating head system.

I would not consider 15F/8.5K to be the norm for receiver outlet subcooling, but rather an upper limit, beyond which liquid is backing up into the condenser... and I always recommend charging until the sight glass is clear or this limit is reached, whichever comes first... whatsmore I always check superheat before charging, as it is a mistake to add refrigerant if the superheat is low.

[US Iceman]

I agree in what you said about liquid seal. I'm not into the industrial stuff. I dont know anything about the construction of the condensers there. I'm talking about normal commercial fin and tube condensers.

Normal for me is the big evaporative condensers and thousands of pounds of ammonia.

One of the good things that came out of this discussion is the fact that each industry has it's own unique requirements, although they are based on common physics.

Air cooled systems do not normally have equalizing lines between the condenser and receiver, so everything flows by pressure difference alone essentially.

On ammonia systems the receiver is a basic gravity drain with the receiver pressure equalizing with the discharge pressure. This would require much larger pipes (risers) as we try to design these so the condenser does not have liquid backed up in it for subcooling. Why use condensing heat transfer surface for subcooling???

These simple differences dictate some different piping practices also, so what happens in a "*****" system may be somewhat different than an ammonia system.

Several other points about non-condensables, subcooling, etc were also discussed. I hope everyone reading this thread can follow it since we covered so many related topics.

BTW, you were not alone in the thread. I was watching the discussion for awhile.

[Tesla]

Guys the methods of measurement described in threads is old hack and just an estimated guess. There is only one correct way to measure the gas - that is direct temp and press measurement. There is a tool from yellow jacket ritchie which does this and I think a lesser one from robinare. So the old way of measuring the pipe temp is not accurate.

[SteinarN]

Guys the methods of measurement described in threads is old hack and just an estimated guess. There is only one correct way to measure the gas - that is direct temp and press measurement. There is a tool from yellow jacket ritchie which does this and I think a lesser one from robinare. So the old way of measuring the pipe temp is not accurate.

Actually i have that kit, two of them.

I plan to do some serious testing soon. Insert that
thermometer into the scrader on the receiver outlet rotalock valve and measure the exact pressure at the same time.

Wonder if i should soften my position ever so slightly first. Admit it may be a slight degree of subcooling there

Have to take some pictures as vell. Just for the records.

[Josip]

Hi, Mike

Normal for me is the big evaporative condensers and thousands of pounds of ammonia.

....for me too .....btw...with compressor's electrical motor usually bigger of 100kW


Best regards, Josip

[nike123]

Hi, Mike



....for me too .....btw...with compressor's electrical motor usually bigger of 100kW


Best regards, Josip

Womans said, that size is not only thing what is matter!

[Springbok]

Hey chilltechnician,since you were the original poster of this thread,does this answer your questions?

[SteinarN]

Hi, Mike



....for me too .....btw...with compressor's electrical motor usually bigger of 100kW


Best regards, Josip

The largest system i have had a look at was on some sort of a trawler, i dont know the english word, they fished pelagic fish, like herring. They take large amounts aboard in a short time, lets say 100 ton fish plus 100 ton water. Or maybe it was several hundred tons, i dont remember. Then that mass must be chilled down from sea temperature of maybe 8C to close to 0C in no more than a couple hours. This was a smal boat but they had a system with 2 large screws with combined cooling capasity of 1950 kW. The electric generator delivered some 2000 kW. The system was actually rather simple. But it was very interesting to have a close look at it and se it running.

[Gary]

The largest system I worked on was a 3600 ton/12660 kW centrifugal chiller for a skyscraper. You could walk around on the top of the motor. They had 3 of those and a little 1250 ton/4400 kW. We had to notify the power company two weeks in advance before starting any of these, so they could divert power for it.

[ChillTechnican]

Yes , well that all brought up a few things to think about and yes my question has been answered , thanks to all. i now own a MA Line dual probe thermometer , and it has been really useful from day one and i am now confident that i am getting accurate readings and there fore a much clearer idea of what the system is doing.
cheers guys and look forward to getting info when i next have a question.
i hope to get back to the chiller in question next week for relocation, will take readings again and let you know.

[ChillTechnican]

the weather here has been alot cooler this last 10 days and had no problems!

[Refrigerologist]

The correct way to determine any existence of non condensables is to measure the liquid subcooling at condenser outlet or receiver outlet. Assuming only minor hight differences and not any measurable pressure differences between condenser/receiver and not overfilled system there should be no measurable subcooling at condenser/receiver outlet. Any subcooling here is a certain symtom of non condensables in the system.

When there is sufficient refrigerant in the system, any non condensables will accumulate in the receiver bc only liquid is drawn out of the receiver and the non condensables (gasses) cant find the way to the bottom of the receiver where the outlet pipe is.

An example:
Lets say the liquid temperature (condenser outlet and receiver outlet) is 30 degres C, but you measure a condensing/receiver pressure of 15,1 bar. At 30 condensing/liquid outlet, the pressure should have been 13,1 bar! In this instance you have 2 bar additional pressure (15,1-13,1) in the receiver caused by non condensables. Lets say the gas phase volume of the receiver is 5 litres, then you have 10 litres of noncondensables at atmospheric pressure in the system. (10 litres at 1 atmosphere equals 5 litres at those 2 atmospheres additional pressure.) This is assuming that the temperature of the liquid surface in the receiver is the same temperature as the condenser outlet/receiver outlet. This assumtion should be valid in a system with short pipes between condenser and receiver, not any excess pressure loss between condenser outlet and receiver outlet and not considerably temperature difference between the condensing temperature and the surrounding air of the receiver.

When the system is stopped, the non condensables have a possibility to escape from the receiver and expand/dilute into a much larger volume of the system. That 10 litres of non condensables, when expanded/diluted in, lets say, 20 litres of the system, will cause an additional pressure above the saturated pressure of only 0,5 bar. (10 litres at 1 bar equals 20 litres at 0,5 bar) Thus it is far more accurate to measure the receiver outlet pressure and temperature when the system is running than measure those values at stand still in order to determine the presence of any non condensables.

Please refer to Daltons Law regarding gas pressure. You will see that different gases excert their own partial pressure. Therefore, if any air is present in a closed system, that has been switched off and allowed to cool down or warm up to the ambient temperature, then by consulting your saturated refrigerant pressure/temperature relationship chart for that refrigerant if the pressure is 1bar or 14.7lbs per sq in higher then it must contain air.

[Refrigerologist]

Why do we need sub-coolinmg then? Firstly it increases the refrigerating effect by reducing adibiatic expansion after the expansion device. Secondly it ensures that there is no flashing of the liquid refrigerant to vapour in the liquid line due to pressure drop within the line. Therefore Gary must be correct that sub-cooling is a good thing and it is what manufacturer's do ask for and they design their condensers specifically for it, hence many condensers actually have a sub-cooler coil.

[Refrigerologist]

The correct way to determine any existence of non condensables is to measure the liquid subcooling at condenser outlet or receiver outlet. Assuming only minor hight differences and not any measurable pressure differences between condenser/receiver and not overfilled system there should be no measurable subcooling at condenser/receiver outlet. Any subcooling here is a certain symtom of non condensables in the system.

When there is sufficient refrigerant in the system, any non condensables will accumulate in the receiver bc only liquid is drawn out of the receiver and the non condensables (gasses) cant find the way to the bottom of the receiver where the outlet pipe is.

An example:
Lets say the liquid temperature (condenser outlet and receiver outlet) is 30 degres C, but you measure a condensing/receiver pressure of 15,1 bar. At 30 condensing/liquid outlet, the pressure should have been 13,1 bar! In this instance you have 2 bar additional pressure (15,1-13,1) in the receiver caused by non condensables. Lets say the gas phase volume of the receiver is 5 litres, then you have 10 litres of noncondensables at atmospheric pressure in the system. (10 litres at 1 atmosphere equals 5 litres at those 2 atmospheres additional pressure.) This is assuming that the temperature of the liquid surface in the receiver is the same temperature as the condenser outlet/receiver outlet. This assumtion should be valid in a system with short pipes between condenser and receiver, not any excess pressure loss between condenser outlet and receiver outlet and not considerably temperature difference between the condensing temperature and the surrounding air of the receiver.

When the system is stopped, the non condensables have a possibility to escape from the receiver and expand/dilute into a much larger volume of the system. That 10 litres of non condensables, when expanded/diluted in, lets say, 20 litres of the system, will cause an additional pressure above the saturated pressure of only 0,5 bar. (10 litres at 1 bar equals 20 litres at 0,5 bar) Thus it is far more accurate to measure the receiver outlet pressure and temperature when the system is running than measure those values at stand still in order to determine the presence of any non condensables.

I am not attacking you personally, just having a debate. I think we all need to go back to the fundamentals: The compressor should not be thought of as a pump. It draws vapour from the evaporator, compresses it, and adds heat, by friction. This superheated gas enters the condenser and by being cooled in the first third (approx) of the condenser it is desuperheated. In the next third it begins to condense. (When I was at Willesden Tech, circa. 1975, they had a fully operable glass tube fridge system and you could actually see this condensation taking place, it was more like rain). At this point any non-condensibles due to the density being lower than the liquid refrigerant rises back to the top of the condenser and remains trapped. The refrigerant carries on being cooled/sub-cooled. The liquid refrigerant carries on as normal, to, (if fitted the receiver), and then on to the expansion device. Remember that the liquid refrigerant is a medium temperature subcooled liquid at this point and due to the pressure differential, which is created by the expansion device, and the compressor withdrawing the vapour, and compressing it, is forced into the evaporator where its' pressure drops. The compressor draws the vapourized refrigerant back into its' bores to be compressed and the temperature is raised. And so the cycle continues. You need to visulize what is occuring in the condenser: the superheated gas does not just flow through and collect any air pushing it in front. As the refrigerant condenses, the air is released and returns to the top of the condenser taking up valuable room. This is why it is called non-condensibles. If the air were to condense it would be carried around the system in a continual cycle. I hope that this rather poor description helps.

In order to clarify some of the above, if a cylinder of refrigerant were piped to an open ended evaporator via a simple capilliary expansion device, and the liquid valve opened, then liquid refrigerant would pass though capilliary and as in a closed system refrigeration would take place. Obviously an expensive excercise! The compressor and condenser arrangement is used to complete the cycle and to retain the refrigerant. Remember we do not need a compressor to obtain refrigeration, we can use absorbtion systems where the initial driving force is heat. Both systems use the application of energy to operate the cycle.

You are correct on one point though if the discharge pressure has risen due to the presence of non-condensibles it is possible to have sub-cooling. But the refrigerant would be at a higher temperature than would be normal for a given ambient. I had not actually thought that part through. Conversely, if there is enough air present and full condensation cannot be acheived then it is possible that there would be no sub-cooling.

[SteinarN]

Please refer to Charles Law and Boyles Law regarding gas pressure. You will see that different gases excert their own partial pressure. Therefore, if any air is present in a closed system, that has been switched off and allowed to cool down or warm up to the ambient temperature, then by consulting your saturated refrigerant pressure/temperature relationship chart for that refrigerant if the pressure is 1bar or 14.7lbs per sq in higher then it must contain air.

I'm fully avare of the part pressures from different gasses in a closed system. There is no need for the pressure in a closed system to be 1bar higher than the saturated temperature if it is air in the system. A smal amount of air will increase the pressure less than one bar, lets say 0,1bar while a large amount of air will increase the pressure several bar over the saturated pressure.

A system containing a smal amount of air will show a far less increase in the saturated pressure at standstill than will be the case if the air could be collected in a smal part of the system like the receiver. Hence my sugestion on how to establish the eventual presence of air in the system.

[SteinarN]

Why do we need sub-coolinmg then? Firstly it increases the refrigerating effect by reducing adibiatic expansion after the expansion device. Secondly it ensures that there is no flashing of the liquid refrigerant to vapour in the liquid line due to pressure drop within the line. Therefore Gary must be correct that sub-cooling is a good thing and it is what manufacturer's do ask for and they design their condensers specifically for it, hence many condensers actually have a sub-cooler coil.

I have never argued subcooling isnt a good thing. I'm fully avare of and i'm able to calculate the result in detail of the "free" increase in cooling capasity if the condenser outlet temperature is lowered or alternatively the receiver outlet is routed through a subcooling coil.

On the other hand, subcooling caused by an increase in condensing pressure without actually lower the liquid temperature is a loss no matter how large the resulting subcooling is.

[SteinarN]

I am not attacking you personally, just having a debate. I think we all need to go back to the fundamentals: The compressor should not be thought of as a pump. It draws vapour from the evaporator, compresses it, and adds heat, by friction. This superheated gas enters the condenser and by being cooled in the first third (approx) of the condenser it is desuperheated. In the next third it begins to condense. (When I was at Willesden Tech, circa. 1975, they had a fully operable glass tube fridge system and you could actually see this condensation taking place, it was more like rain). At this point any non-condensibles due to the density being lower than the liquid refrigerant rises back to the top of the condenser and remains trapped. The refrigerant carries on being cooled/sub-cooled. The liquid refrigerant carries on as normal, to, (if fitted the receiver), and then on to the expansion device. Remember that the liquid refrigerant is a medium temperature subcooled liquid at this point and due to the pressure differential, which is created by the expansion device, and the compressor withdrawing the vapour, and compressing it, is forced into the evaporator where its' pressure drops. The compressor draws the vapourized refrigerant back into its' bores to be compressed and the temperature is raised. And so the cycle continues. You need to visulize what is occuring in the condenser: the superheated gas does not just flow through and collect any air pushing it in front. As the refrigerant condenses, the air is released and returns to the top of the condenser taking up valuable room. This is why it is called non-condensibles. If the air were to condense it would be carried around the system in a continual cycle. I hope that this rather poor description helps.

In order to clarify some of the above, if a cylinder of refrigerant were piped to an open ended evaporator via a simple capilliary expansion device, and the liquid valve opened, then liquid refrigerant would pass though capilliary and as in a closed system refrigeration would take place. Obviously an expensive excercise! The compressor and condenser arrangement is used to complete the cycle and to retain the refrigerant. Remember we do not need a compressor to obtain refrigeration, we can use absorbtion systems where the initial driving force is heat. Both systems use the application of energy to operate the cycle.

You are correct on one point though if the discharge pressure has risen due to the presence of non-condensibles it is possible to have sub-cooling. But the refrigerant would be at a higher temperature than would be normal for a given ambient. I had not actually thought that part through. Conversely, if there is enough air present and full condensation cannot be acheived then it is possible that there would be no sub-cooling.

I've never seen a glass tube condenser. One thing that comes to mind is that such a condenser without fins must have an exceptional low capasity in relation to the area of pipe surface and pipe volume. It makes sence for me that the gas has to flow very slowly through the pipe and therefore show the liquid at the bottom and gas at the top in the pipe.

In a real modern condenser the speed of flow is much greater and not allowing that separation of the gas and liquid as you saw. The flow inside a real condenser is more like the flow through a liquid line sight glass on a system short of gas where the sight glass is showing heavy foaming, a relatively uniform flow of opaque foam.

In the beginning of the condensing section there is the highest velocity. There will be a thin film of liquid at the pipe wall, and fast flowing gas in the rest of the pipe. As the liquid film grews thicker the gas will tear off excess liquid from the wall. There will gradually arise a gas/liquid foam at the center of the pipe but still a thin film of liquid at the wall. Toward the end the film at the wall grew thicker and actually partially start to subcool. The velocity decreases gradually as the foam in the middle condenses fully. Only in the very last part of the condenser is the velocity low enough to allow some separation of the liquid at the bottom and gas at the top in the pipe. However any noncondensables will be sweept along with the refrigerant out of the condenser.

Non condensables in the receiver will cause increased condensing pressure and lower condenser outlet temperature. You can acheive a fenomenal amount of subcooling with non condensables in the receiver. If it is enough refrigerant in the system, then it is absolutely impossible not to have subcooling out of the condenser with non condensables in the system.

[Refrigerologist]

I'm fully avare of the part pressures from different gasses in a closed system. There is no need for the pressure in a closed system to be 1bar higher than the saturated temperature if it is air in the system. A smal amount of air will increase the pressure less than one bar, lets say 0,1bar while a large amount of air will increase the pressure several bar over the saturated pressure.

A system containing a smal amount of air will show a far less increase in the saturated pressure at standstill than will be the case if the air could be collected in a smal part of the system like the receiver. Hence my sugestion on how to establish the eventual presence of air in the system.

Not in a static system, it will only raise the pressure by 1 atmosphere, unless you have managed to compress the non-condensibles and the compressor kept sucking more and more air into the system and it became completely overload with air. Before that point is reached the system waould have failed on high pressure limit.
Switching the system off, allowing it to stabilse to the average ambient, and then reading the standing pressure and comparing the recorded pressure against a pressure temperature relationship chart for the refrigerant is standard industry practice, and the according to the laws of partial pressures the air will excert its own force ie 1 atmosphere above the refrigerant pressure. This is a scientific fact. I did not make it up and I am sure if you research it you will find it to be true!

Also you say the non condensibles end up in the receiver. Where do they go then in a system without a receiver?

I did also say that I had not thought it through and that subcooling could be high, but this would be based on the higher than normal pressure, but the actual refrigerant temperature would also be much higher than is usual due to the lack of condensing effect due to air remaining in the top of the condenser.

[Refrigerologist]

Why not look up KOTZA on the internet and get one of their training evalution disks. It covers this subject really well. It is free and provides tests based on the temperature value and pressure readings and uses an animated type display showing gauges etc.

[SteinarN]

Not in a static system, it will only raise the pressure by 1 atmosphere, unless you have managed to compress the non-condensibles and the compressor kept sucking more and more air into the system and it became completely overload with air. Before that point is reached the system waould have failed on high pressure limit.
Switching the system off, allowing it to stabilse to the average ambient, and then reading the standing pressure and comparing the recorded pressure against a pressure temperature relationship chart for the refrigerant is standard industry practice, and the according to the laws of partial pressures the air will excert its own force ie 1 atmosphere above the refrigerant pressure. This is a scientific fact. I did not make it up and I am sure if you research it you will find it to be true!

I dont se that any of what i said contradict what you says here. Of course if you charge refrigerant in a system while it contains one bar absolute pressure of air, the resulting pressure will be 1 bar higher than the saturated ref. pressure. But if it contained only, lets say, 0,1 bar absolute pressure of air, then the resulting saturated pressure would be 0,1 bar higher as you indicate. If that system is started and the air acumulate in the receiver, then the receiver/condensing pressure would be significantly more than 0,1 bar higher than the saturated pressure. Imo, small amounts of non condensibles is best diagnosed with the system running, especially in systems with small receiver volumes compared to the volume of the rest of the system.

[Refrigerologist]

Sorry guys', in one of my previous posts I quoted the wrong gas law. It should of course been Daltons' Law of Partial Pressures. I have corrected it to save confusion, but mainly to stop me looking any more of a plonker than I some of you may already think!

It's my age you know. senility is creeping up fast!

[SteinarN]

Sorry guys', in one of my previous posts I quoted the wrong gas law. It should of course been Daltons' Law of Partial Pressures. I have corrected it to save confusion, but mainly to stop me looking any more of a plonker than I some of you may already think!

It's my age you know. senility is creeping up fast!

I did not notice. I dont even know the name of the law tho i know the implications of it.

Maybe some of our different views comes from different experiences with different systems. My experience is only with smal and moderate size comercial systems.