PDA

View Full Version : Flooded Starts on Marine Air Conditioner



Mayball
08-04-2007, 10:07 AM
I am having trouble with refrigerant (R-22) migration to the compressor on a CruiseAir system in a motor yacht. The unit has sea water cooling and a oil separator and capillary tube expansion. It would be difficult to add a pump-down solinoid. I understand that overcharging with refrigerant can contribute to this migration. My question is whether some degree of undercharging can prevent the migration. I have had to replace the hermetic compressor three times.

NoNickName
08-04-2007, 10:36 AM
I don't think that one is an oil separator, but a liquid receiver. I think you need a suction accumulator and a suction riser (sort of pipe pointing upwards), or a suction filter with drainage.

http://img.alibaba.com/photo/51221155/Suction_Line_Filter.jpg

The Viking
08-04-2007, 12:39 PM
Well,
The obvious answer is to fit a crankcase heater, if the compressor is warmer than the rest of the system then there won't be any liquid in it when it starts.

But as you are on a boat, the question arises, have you got the power supply to do it?

Under charging?
Well 2 negatives makes a positive but adding another fault? Will that really sort the first problem out?



P.S
As you said it was migration, I have assumed that it happens when the compressor stands, not when it is running?

lana
08-04-2007, 04:24 PM
Hi there,
As Viking said "migration" is a term used for refrigerant movement when system is stopped. This is due to temp. difference.
Flooding is a term used for liquid refrigerant reaching the compressor.
The reasons :
1- The hight difference between the evaporator and the compressor.
2- If the system is off and the oil separator is in a cold place then the refrigerant gas inside could condense and through the floating valve enters the compressor.
3-Malfunction of your expansion device.If there is cap tube then maybe you overcharged it.
4-Any problem related to the evaporator that prevents good evaporation of the liquid refrigerant inside the evaporator. This can be verified by measuring the superheat. If it is near 0°C then this problem exists. Check these also, fan is working, frosting on the coil.
Good luck.
Cheers:)

monkey spanners
08-04-2007, 05:06 PM
Why not fit a solenoid just before the capiliary. No need to pump down. That way you've only got the volume of refigerant that was in the evaporator and suction line instead of the condenser and liquid line to condense in the compressor.
Is it migrating to the compressor from the evaporator, or leaking back from the highside?

Jon

lana
08-04-2007, 06:30 PM
Hi,
What monkey-spanners says is called "Minimum compressor protection". If there is a "flooded Start" then try it.

What is the relative hight between the evaporator and the compressor?:confused:

Cheers:)

Mayball
08-04-2007, 08:13 PM
The unit has an oil separator and a suction accumulator. The evaporator is two decks above the compressor and thirty feet forward. I think the refrigerant is migrating from the evaporator when the system is off, however I could fit a check valve downstream of the oil separator to prevent flow back from the condenser. The sequencing of a solenoid is difficult because of the electronic controls, and an interior crankcase heater is impossible on this hermetic compressor.

The Viking
08-04-2007, 09:32 PM
Ahh,

But the crankcase heater doesn't need to be internal, it's common practise to have external heaters strapped on to hermetic compressors.
If it was me than I would strap one to the compressor and another one to the oil separator.
(it would be cheap and easy to try)

Mayball
08-04-2007, 10:27 PM
The more I think about this, the more I think the refrigerant flooding is coming from the condenser back through the oil separator. The previous compressor failures were due to oil starvation and is why I added the oil separator. Due to space restrictions, I installed a Temprite 320 separator which does not use a float valve, but meters the oil back to the suction through a capillary tube. I think putting a check valve on the separator outlet will fix the problem. What do you think?

Brian_UK
08-04-2007, 10:34 PM
Surely a check valve in the oil separator outlet will defeat the object of the separator.

It might be safer to simply put a solenoid valve in the line connected in with the compressor contactor.

Mayball
08-04-2007, 10:39 PM
I don't mean the oil outlet from the separator, I mean the refrigerant outlet heading to the condenser.

Brian_UK
08-04-2007, 10:48 PM
OK, understood, but it will only stop the flow backwards towards the compressor discharge valves. If you are suspecting the compressor valves are letting by then that is the problem not the separator.

However if you are returning the oil back to the compressor crankcase through a capillary line what is stopping the flow of refrigerant through the separator after the oil has returned?

Also are there any oil traps in the system at all especially a loop by the evaporator to stop flow down through the yacht back to the compressor?

Mayball
08-04-2007, 10:49 PM
I didn't mean to install the check valve on the oil outlet from the separator, but on the refrigerant line from the separator to the condenser.

Mayball
08-04-2007, 10:55 PM
Nothing can stop the refrigerant flow through the cap tube after the oil, but when the compressor is running, this is not a problem. However, when the compressor stops, the only refrigerant return through the tube would be that which is trapped between the check valve and the discharge valves.

Mayball
08-04-2007, 11:03 PM
There are no oil traps between the evaporator and the compressor, but there are no vertical risers with upward flow. It has a steady downward slope all of the way.

Brian_UK
08-04-2007, 11:07 PM
Nothing can stop the refrigerant flow through the cap tube after the oil, but when the compressor is running, this is not a problem. However, when the compressor stops, the only refrigerant return through the tube would be that which is trapped between the check valve and the discharge valves.True, sorry I wasn't thinking straight :(

lana
09-04-2007, 09:32 AM
Hi Mayball,

Try an inverted trap before the condenser to prevent liquid refrigerant going into the compressor. Also install the check valve after the oil separator (on the refrigerant line), as mentioned before.
Good luck.:)

Mayball
09-04-2007, 07:48 PM
Hi Lana

Because this installation is also a heat pump, the oil separator discharges through a reversing valve which, in effect, has both an oil trap and an inverted oil trap that the refrigerant must go through when traveling backwards from the condenser to the oil separator, so I think that part is covered.

Mayball

Brian_UK
09-04-2007, 08:26 PM
Reading back though this thread and it seems that the installation of the oil separator is the cause of the floodback problems as you suspect.

I will therefore stick with my previous suggestion:- fit a solenoid valve into the oil line from the separator that will shut down when the compressor is idle.

It can be fed via a relay which is triggered by the power line to the compressor.

Mayball
09-04-2007, 09:15 PM
Certainly, the solinoid on the oil line will eliminate any chance of refrigerant migration to the suction side, and there is already a spare reley to drive it, that used to be used on a DC cooling water pump. The question in my mind is which is more reliable (least chance of failure): the electro/mechanical solinoid-relay combination, or a mechanical check valve in the hot discharge gas environment coming from the oil separator.

Thanks for your input, Brian

Mayball

Brian_UK
09-04-2007, 10:49 PM
A check valve will impart some form of restriction in the refrigerant line but obviously has its merits.

The modern solenoid is a pretty reliable item nowadays and using one with a 'timed cycle' is one of the Temprite options with the 320 separator.

I have got a little confused as to where the oil return capillary is connected to but if it is connected into the suction to the accummulator then it would give you another 'reservior' for any refrigerant that gets through.

Mayball
09-04-2007, 11:34 PM
Hi Brian

I did have the oil return line connected to the process tube on the suction of the compressor. There is no room for a suction accumulator on the condensing unit. Cruisaire did have what looks like a 2" D copper strainer on the suction line before the compressor, but when I took it off, you could see right through it, so I think it was only a muffler. I replaced it with a suction filter/dryer, since that was all the space I had. The filter/dryer has inlet and outlet pressure taps, and I could attach the oil line to the inlet tap, if that seems appropriate.

Don

Brian_UK
09-04-2007, 11:43 PM
Sorry Don, you said that you had an accumulator in an earlier post that's why I suggested that route.

I am surprised that being a heat pump the unit does not have an accumulator as on changeover there is always the risk of liquid going straight into the compressor, especially on defrost, however....

Yes, why not connect onto the filter/drier.

My admiration of you doing this work on board at your respectable age is growing Don, keep it up:)

Mayball
10-04-2007, 12:29 AM
Hi Brian,

The only reason I keep working here is it is 80 Degrees here in Puerto Vallarta. By the way, how is the weather there in Dorset?

Don

lana
10-04-2007, 05:54 AM
Hi Don,

I suggest that you draw a simple schematic of your system and attach it here. This would clear so many questions and help to find a solution ASAP.
Cheers:)

Gary
10-04-2007, 06:10 AM
You mentioned that there is a steady downward slope in the suction line, from the evaporator to the compressor. It is entirely possible (and not uncommon) that liquid refrigerant gathers in the evaporator (coldest point) during the off cycle and is then delivered by gravity (as opposed to migration) to the compressor inlet. This can be eliminated/minimized by looping the suction line above the evaporator and then down to the compressor, thus trapping the liquid in the evaporator.

Mayball
10-04-2007, 11:55 AM
Hi Lana,

In the attachment, the following legend applies:

SA - Compressor, BC - Oil Separator, DE - Proposed Check Valve, FG - Reversing Valve, H - Solenoid Valve Proposed by Brian UK, IJ - Water Cooled Condenser, KL - Liquid Line Filter/Dryer, MN - Cap Tube, OP - Evaporator, QR - Suction Filter/Dryer. The condenser is actually coaxial and is located above the compressor.

Mayball

Mayball
10-04-2007, 12:03 PM
Hi Gary,

The suction line does leave the evaporator at an elevation above the evaporator. This fact, and the fact that the migration did not happen until I added the oil separator, leads me to believe that the migration is not coming from the evaporator.

Thanks
Mayball

lana
10-04-2007, 12:17 PM
Hi MAyball,

In the diagram the reversing valve is not right. It must be connected to both heat exchangers. Can you correct it and show the actual one? Also where is the accumulator?
The suction line must be connected to the reversing valve also.
Lana

Gary
10-04-2007, 02:55 PM
I am concerned that an oil separator was added without resolving the original oil return problem. There should be a trap at the evaporator outlet.

Is there an oil sightglass on the compressor?

What makes you believe the compressor is flooded on startup?

Is it possible that the compressor is flooding with excess oil rather than refrigerant?

hfc134a
10-04-2007, 03:25 PM
hi sounds like you have a flooding back problem rather than migration. Is there sufficient air flow across the evaporator/ evaporator fan running?
I have installed/worked on many of these systems some times the boat manufactures do not allow for air return or squash the air outlet hose considerably restricting flow. Is the fan fuse ok?? hope this helps.

Mayball
10-04-2007, 05:37 PM
Hi Gary,

The oil separator was installed to resolve the oil return problem. The previous two compressor failures were due to oil starvation. I believe the oil was trapped in the evaporator. I measured the oil in the compressors that failed. The first two were about 1/3 low with dirty oil. The last one was almost exactly right with clean oil. However, when I depressured that last compressor to check the oil, a very large amount of refrigerant boiled off. There is no sight glass on this Tecumseh 1 HP compressor. That last compressor failed after the longest shutdown period in its service; previous shut downs were normal cycling.
mayball

Mayball
10-04-2007, 05:44 PM
Hi Lana

I only showed the high side part of the reversing valve because that is the part that is involved in my problem. The suction line is also connected to the reversing valve, but since I am not a great artist, I left that part out. I misspoke when I said there was an accumulator, only the suction filter/dryer, which I installed as there is no room for an accumulator. The original condensing unit had only a muffler.

Mayball

Mayball
10-04-2007, 05:53 PM
Hi HFC134a,

There is plenty of air flow through the evaporator. In fact it has twice the fans of an identical unit on the same boat, that is running fine. The evaporator is sandwiched between inlet and outlet plenums and use grills from and to the air conditioned space rather than hose ducts.

Mayball

lana
10-04-2007, 07:29 PM
I misspoke when I said there was an accumulator, only the suction filter/dryer, which I installed as there is no room for an accumulator. The original condensing unit had only a muffler.


Hi Mayball,

I think your problem is 100% for not having an accumulator. In a heat pump unit, an accumulator is a must.:eek:
When changing the mode (cooling & heating) there is definitely liquid flood back to the compressor.
I suggest that you make some room to put an accumulator.
Good Luck.

Mayball
10-04-2007, 07:39 PM
Hi Lana,

I totally agree that this is a poor design, but this is Cruisair's design. Also, this unit has never been run as a heat pump and the reversing valve has never been exercised. Further, it probably never will be used as a heat pump, because the cabin has a very adequate diesel heater and the owner said he would never use the heat pump. I will, however, caution him on its use, and, if he wishes will disable the reversing valve.

Mayball

lana
10-04-2007, 07:52 PM
Hi,

I think after 36 posts we still don't understand the actual problem :mad: .

Please answer the following questions very carefully and clearly.

1- Is there only one compressor on the unit or you have two parallel one?
2- Is there an accumulator?
3- Is this a heat pump or not? If this is not going to be used as a heat pump then take away any unnecessary parts like the reversing valve and make the system simpler to handle.

Cheers

Mayball
10-04-2007, 08:29 PM
Hi Lana,

Sorry I seem so vague.
1. There is only one compressor on the unit. There is, however another system on the boat identical to this one.
2. There is no accumulator.
3. This is an air conditioner that can function as a heat pump. This condensing unit/heat pump was ordered by the customer and he hired me to install it and connect it to his existing air handler (replacing an existing 134a Glacier Bay feed to that air handler). I cannot remove the reversing valve because the customer wants the capability available when he sells the boat.

Mayball

Gary
10-04-2007, 09:32 PM
What was the fault with the three compressors? Valves broken? Burnt windings? Other?

Gary
10-04-2007, 09:35 PM
There is, however another system on the boat identical to this one.

Not really. It sounds like this one has twice the airflow. Does it also have twice the evaporator? Perhaps twice the refrigerant charge?

Mayball
10-04-2007, 11:15 PM
Hi Gary,

I only opened one of the failed compressors: one of the two that were oil starved. It had wiped the main bearing and the connecting rod bearing. The third compressor which had not lost oil and, I believe, failed due to a flooded start had close to a dead short in the windin(s). On energing the compressor, it drew 67 amps and tripped the breaker. Its valves appeared to be OK.

The second system is identical to the failed system except it has one fan instead of two and and the failed system has about double the distance and elevation between the condensing unit and the air handler. It therefore has a larger refrigerant charge, having twice the length of liquid line.

Mayball

Gary
11-04-2007, 12:40 PM
During the off cycle, given an open path between the high side and the low side, most of the refrigerant charge will migrate to, and condense in, the coldest part of the system. The larger the charge, the more refrigerant to migrate.

The purpose of the liquid line solenoid is to close the path between high side and low side, and the purpose of the crankcase heater is to ensure that the compressor is never the coldest part of the system.

You need to install a liquid line solenoid, the closer to the cap tube the better, which closes when the compressor stops... and you need a crankcase heater.

Mayball
11-04-2007, 07:17 PM
Hi Gary,

I agree with you on the advantages of a crankcase heater but I disagree on the need for a solenoid at the entrance to the evaporator cap-tube. In fact, that solenoid would exacerbate the problem. Consider the following:
1. There was no problem with refrigerant migration or flooded starts, until I added the oil separator to solve the oil starvation problem.
2. The addition of the oil separator to the compressor discharge does not affect the suction line or cause refrigerant migration down the suction line to the compressor when the unit is shut down.
3. The addition of a solenoid before the evaporator would eliminate the off cycle equalization of pressure through the capillary tube, leading to harder starts, and worse, would trap the liquid between the condenser and the solenoid. This liquid would then move backwards through the condenser to the oil separator and thence out the oil outlet of the separator to the compressor suction, driven, not by temperature difference, but by pressure difference.

This is why both Brian UK and I see an advantage of placing the solenoid on the oil line between the separator and the compressor suction. To my mind, however, a more simple solution is a check valve added to the line between the oil separator and the condenser. Both the solenoid and the check valve would stop any backward movement of refrigerant from the condenser to the compressor suction. This is, of course, not a problem when there is no separator because back flow is prevented by the compressor valves.

Mayball

lana
11-04-2007, 08:34 PM
3. The addition of a solenoid before the evaporator would eliminate the off cycle equalization of pressure through the capillary tube, leading to harder starts, and worse, would trap the liquid between the condenser and the solenoid. This liquid would then move backwards through the condenser to the oil separator and thence out the oil outlet of the separator to the compressor suction, driven, not by temperature difference, but by pressure difference.



In my opinion this is completely wrong.



This is why both Brian UK and I see an advantage of placing the solenoid on the oil line between the separator and the compressor suction. To my mind, however, a more simple solution is a check valve added to the line between the oil separator and the condenser. Both the solenoid and the check valve would stop any backward movement of refrigerant from the condenser to the compressor suction. This is, of course, not a problem when there is no separator because back flow is prevented by the compressor valves.


Do that.

Also do whatever Gary said in his last post.

Cheers:)

Gary
11-04-2007, 11:17 PM
1. There was no problem with refrigerant migration or flooded starts, until I added the oil separator to solve the oil starvation problem.

On the contrary, I believe the oil was washed out of the first two compressors by flooded starts from migration.



2. The addition of the oil separator to the compressor discharge does not affect the suction line or cause refrigerant migration down the suction line to the compressor when the unit is shut down.

Migration to the separator does not affect the suction line. Rather it dumps liquid directly into the compressor via the oil return line. Oil separators (particularly those mounted below the condenser) are a target for migration from the condenser. That's why the check valve after the separator is a very good idea.



3. The addition of a solenoid before the evaporator would eliminate the off cycle equalization of pressure through the capillary tube, leading to harder starts, and worse, would trap the liquid between the condenser and the solenoid. This liquid would then move backwards through the condenser to the oil separator and thence out the oil outlet of the separator to the compressor suction, driven, not by temperature difference, but by pressure difference.

It doesn't work that way.



This is why both Brian UK and I see an advantage of placing the solenoid on the oil line between the separator and the compressor suction. To my mind, however, a more simple solution is a check valve added to the line between the oil separator and the condenser. Both the solenoid and the check valve would stop any backward movement of refrigerant from the condenser to the compressor suction. This is, of course, not a problem when there is no separator because back flow is prevented by the compressor valves.

Congratulations. You now have two migration targets.

Mayball
12-04-2007, 04:13 AM
Hi Lana,
Are you saying that everything I said in item 3 is wrong? Are you saying that it could not happen that way? If so, could you be more specific in explaining why?
Mayball

lana
12-04-2007, 04:34 AM
Hi Mayball,

Gary Said it all.



The addition of a solenoid before the evaporator would eliminate the off cycle equalization of pressure through the capillary tube, leading to harder starts,


Why would be a hard start? This is the whole point of Pump-Down.


and worse, would trap the liquid between the condenser and the solenoid. This liquid would then move backwards through the condenser to the oil separator and thence out the oil outlet of the separator to the compressor suction, driven, not by temperature difference, but by pressure difference.

If every piping and trap is installed correctly then why would the liquid move? You say it moves because of pressure difference:confused: . Difference between what points? I guess you mean between compressor crankcase pressure and pressure in the condenser. BUT there is the valve reeds there. Of course if we leave a system in pump-down for a very long time this might happen but for start and stop this will not. Believe me.

Installing a solenoid valve and implementing pump-down is the best way to protect the compressor from liquid refrigerant during off cycle. If you want read the Copeland manual (the 5 volume).

Hope this clear some points
Good Luck
Cheers

Mayball
12-04-2007, 04:45 AM
Hi Gary,

If 2/3 of the oil in the system washed out during flooded starts, why has none of it ever come back? Of course, 1/3 of the oil did leave each of the two compressors, and didn't come back, but I think it resides in the evaporator. If it blew through the compressor on flooded starts it would seem more likely to return, since it is passing through the system in slugs, sort of like the oil trap effect. It seems to me more likely that the oil passed into the system, in the normal fashion (the unit ran 24/7 all summer last year in the tropics before it failed) and is still in the evaporator and it isn't coming back. I can't imagine it lasting very long with a flooded start each time it cycled.

Mayball

Mayball
12-04-2007, 05:03 AM
Hi Lana,
The pump down protects the system on the suction side, but it does this by pumping into the high side. If it can then move from the high side to the low side by pressure differential through the oil separator, then the flooded start possibility is aggrevated. That is why I thought it would be better to install the check valve before the condenser. I still think the flooded start came from the condenser refrigerant moving backward through the oil separater rather than from the evaporator moving forward.

Mayball

lana
12-04-2007, 05:19 AM
The pump down protects the system on the suction side, but it does this by pumping into the high side. If it can then move from the high side to the low side by pressure differential through the oil separator, then the flooded start possibility is aggrevated.

As we said before, if the oil separator body gets colder then refrigerant gas would condense in the oil separator and goes into the crankcase. This is why we suggested to install an inverted trap and check valve on the discharge line.
If there is no liquid refrigerant in the oil separator then it can not move to the compressor.

You agree with us on the solution but you don't agree with the reason :D .

Cheers:)

Mayball
12-04-2007, 11:13 AM
Hi Lana,

You're right, I do agree with you on the solution. Also, I agree with you on the reason the solution works. The part I don't agree with is the reason the unit sustained the flooded start(s) in the first place. If I understand you and Gary, you think the migration came from the evaporator during the off cycle. I think the flooding came from the condensor, through the oil separator during a shut down period.

Mayball

Gary
12-04-2007, 01:54 PM
Hi Gary,

If 2/3 of the oil in the system washed out during flooded starts, why has none of it ever come back? Of course, 1/3 of the oil did leave each of the two compressors, and didn't come back, but I think it resides in the evaporator. If it blew through the compressor on flooded starts it would seem more likely to return, since it is passing through the system in slugs, sort of like the oil trap effect. It seems to me more likely that the oil passed into the system, in the normal fashion (the unit ran 24/7 all summer last year in the tropics before it failed) and is still in the evaporator and it isn't coming back. I can't imagine it lasting very long with a flooded start each time it cycled.

Mayball

I wouldn't be a bit surprised to find that the evaporator is oil logged. It will probably all come back when someone switches the system to heat mode.

Refrigerant migrates to the coldest point. During a normal cycle, the evaporator is the coldest point and the compressor is still warm. Migration would only occur during shutdown periods, when the evaporator warms up and compressor cools to a lower temperature than the evaporator. It would not have a flooded start each time it cycles.

Gary
12-04-2007, 02:00 PM
Hi Lana,

You're right, I do agree with you on the solution. Also, I agree with you on the reason the solution works. The part I don't agree with is the reason the unit sustained the flooded start(s) in the first place. If I understand you and Gary, you think the migration came from the evaporator during the off cycle. I think the flooding came from the condensor, through the oil separator during a shut down period.

Mayball

Or both. Migration follows whatever open path is available. You now have two open paths and you need to close both of them (check valve between the separator and condenser, and liquid line solenoid near the cap tube).

BTW, is the evaporator top feed (as in your drawing) or bottom feed? In heat mode, the indoor coil is the condenser and the flow is reversed. If it is top feed in cooling mode, then how does the liquid get out of the condenser in heat mode, when the flow is now upwards?

Mayball
13-04-2007, 09:43 AM
BTW, is the Evaporator top feed (as in your drawing) or bottom feed? In heat mode, the indoor coil is the Condenser and the flow is reversed. If it is top feed in cooling mode, then how does the liquid get out of the Condenser in heat mode, when the flow is now upwards?
Good question. It is top feed. How is that accomplished in a normal heat pump? They must be bottom feed as an evaporator so they can be top feed as a condensor.

Gary
13-04-2007, 08:41 PM
Good question. It is top feed. How is that accomplished in a normal heat pump? They must be bottom feed as an evaporator so they can be top feed as a condensor.

Yep... and this would help wash the oil out of the indoor coil, too.

church2k
13-04-2007, 11:34 PM
Lo! I think that the most easy u can do, is to put a non-return valve before the evaporator for avoid the liquid could return to the oil separator, or even best, to build a gooseneck(siphon), at the top of the oil separator. I use to build with a meter up, meter down, bend and to the top again, that ensures u that the condensated liquid in the avaporator is not mixed with the oil,cos in the the satr up of the compressor there`s a change in the pressure of the system that makes the liquid vaporise and drift the oil which by its higher weight is the first leaving the oil separator. Anyway, the resistance in the separator is not even a good idea, is essential in most cases.

Mayball
14-04-2007, 11:41 AM
As of yesterday, the system is back up and running with a check valve between the oil separator and the condenser, and with the heat pump capability disabled. Ironically, the only problem so far has been a flooded evaporator tray, because of the lack of a vent in the existing gravity condensate drain.

Thanks so much to all of you for your valuable suggestions on this matter. I will re-post if any further problems occur on this system.

Mayball

Brian_UK
14-04-2007, 08:28 PM
Thanks Mayball, keep us posted.