Heat pumps freezing

[dkalasz]

I have a problem that has us stumped over here in afghanistan. we have many (2000+ on this base alone)

heres the conditions before i get to the problem

5000ft to 9000ft above sea level

low humidity usually 40% or less

Identical units in identical conditions

problem

SOME of the units freeze lets say 15% of the total as a matter of fact you can pump down 2 units weigh in a charge to both and one may freeze and the other not.

any ideas? i'm tired of defrosting this crap lol

[Tesla]

Hi
Just a stab in the dark with info provided it could be a restriction in the liquidline. If the prob.s got you stumped use of dataloggers would help provide more to go on. Good luck

[Toosh]

Hi Dk are the units rated for high altitude is the first question I would ask

[nike123]

Who is the manufacturer? Also, which is the model of these units?

I suppose that they are window type.
Many of Chinese manufactures have different compressors in same type of their units, because they have buying all stocks of technically obsolete compressors in entire world and installing them in their unit as they arrives in containers. :mad:
Check that all components are same in these units. That could be indication of possible problem.;)

[Andy]

I have a problem that has us stumped over here in afghanistan. we have many (2000+ on this base alone)

heres the conditions before i get to the problem

5000ft to 9000ft above sea level

low humidity usually 40% or less

Identical units in identical conditions

problem

SOME of the units freeze lets say 15% of the total as a matter of fact you can pump down 2 units weigh in a charge to both and one may freeze and the other not.

any ideas? i'm tired of defrosting this crap lol

Hi dkalasz:)

are they freezing on heating or cooling.

If it is heating, what about the pipe runs do they exceed the manufacturers recomended.

Please post some temps and pressures and we can see where the problem is.

Kind Regards Andy :)

[Gary]

Upon checking the weather in Afghanistan, I see that it is winter there, so let's assume heating.

The temperature drops well below freezing at night, so the question is not "Why are the coils freezing?" but rather "Why are the coils not automatically defrosting themselves?".

Does that sound about right?

[Andy]

Upon checking the weather in Afghanistan, I see that it is winter there, so let's assume heating.

The temperature drops well below freezing at night, so the question is not "Why are the coils freezing?" but rather "Why are the coils not automatically defrosting themselves?".

Does that sound about right?

Yep:)

I have often seen this problem associated with incorrectly placed defrost termination stat phails or probes.

Kind Regards Andy:)

[dkalasz]

ok sorry tried to edit this earlier but for some reason was unable, these uniots are split systems rangeing in size from 1/2 ton to 2 ton most are chigo but we have the same problem with at least 8 other manufacures units carrier trane and several misc chinese brands they all have the problem but the real stumper is you can take 2 identical units pump them down weigh in identical charges and one will freeze the other wont. you can take multiple new units some will some wont freeze and this is all over the country. yes i realize altidude is the main issue here but what i am investigateing is why they don't either all freeze or not.. and yes we have the problem both summer and winter in summer its obviously the evap. while another unit can run continuesly for weeks without a spec of ice.

there are no seperate controls for the defrost its all operated by a single board and your dead on about the defrost seems each one of these has a mind of its own no one has been able to figure them out aside from the fact it seems chigo at least changes there cycles with every shipment ( we have upwards of 20,000 of these in country)there are no sensor on the condesor what so ever the only 2 sensors on the unit are a air temp sensor (same type as a window unit ) and an evap temp sensor. one unit in the office i have been keeping an eye on seems to have the defrost on either a 12 or a 24 hr cycle as it defrost the same time each morning . i am sleeping at night thus i cannot confirm a night cycle .(loggers and such are not available in country i finally broke down and bought the tools to measure superheat on my own)

overcharging the unit will help the problem in the winter. allthough it tendds to take out compressors once summer comes and the cycle is reversed.

[nike123]

yes i realize altidude is the main issue here but what i am investigateing is why they don't either all freeze or not..
Why you, or anyone else, think that altitude has anything to do with this symptom? Please, explain me, if you (or anyone else) want, because I don't see or know any reason for that.:confused:

Ups, I see now. High altitude = thinner air = lower exchanger efficiency:o

That mean that on higher altitudes heat exchangers (or/and air volume) need to be bigger.

[dkalasz]

the main reason we believe this is that the very few units that we have that are designed for high altitude a work Alot better and B we have zero problems with them if you take out altitude then i'm at a loss yto expalina the efficency differences between when i am in bagram (5000') and when i am in l'wara (8000') l'wara being much more troublesomealmost all the units needed constant defrosting even when they did run through there cycles. these are units of less than 6 months age at the time. unfortunatly many many tech have been trying to solve this problem for the last 5 years. I only really started investigateing it this year simply because i am tired of melting ice its.....cold

[Brian_UK]

It really does sound as though there is a major flaw with the controls of the various units regarding defrosting functions.

[Gary]

Here is a chigo manual:

http://www.unionaire.us/fileAdmin//Technical%20Manual.pdf

Check out page 9-3

Heat pumps require an outdoor temp sensor.

[Gary]

Pressure is pressure, temperature is temperature and humidity is humidity. The only difference I can think of for high altitude would be lower air density. Quite possibly these units need more airflow to achieve the same heat transfer. Have you measured the delta-T's (air in and air out temp differences) across both coils to see if they have sufficient airflow? Have you compared delta-T's between units that freeze and units that do not freeze? The higher the delta-T the lower the airflow, the lower the airflow the colder the coil, the colder the coil the more likely it is to freeze.

[nike123]

there are no seperate controls for the defrost its all operated by a single board and your dead on about the defrost seems each one of these has a mind of its own no one has been able to figure them out aside from the fact it seems chigo at least changes there cycles with every shipment ( we have upwards of 20,000 of these in country)there are no sensor on the condesor what so ever the only 2 sensors on the unit are a air temp sensor (same type as a window unit ) and an evap temp sensor. one unit in the office i have been keeping an eye on seems to have the defrost on either a 12 or a 24 hr cycle as it defrost the same time each morning . i am sleeping at night thus i cannot confirm a night cycle .(loggers and such are not available in country i finally broke down and bought the tools to measure superheat on my own)



When there is no sensor/thermostat in outdoor unit, than controller estimate need for defrost based on temperature sensors of indoor unit (heat exchanger and ambient) and complex algorithm whose link to flowchart I attached here.
http://www.mediafire.com/?antn2z2yytc

[nike123]

the main reason we believe this is that the very few units that we have that are designed for high altitude a work Alot better and B we have zero problems with them if you take out altitude then i'm at a loss yto expalina the efficency differences between when i am in bagram (5000') and when i am in l'wara (8000') l'wara being much more troublesomealmost all the units needed constant defrosting even when they did run through there cycles. these are units of less than 6 months age at the time. unfortunatly many many tech have been trying to solve this problem for the last 5 years. I only really started investigateing it this year simply because i am tired of melting ice its.....cold

Does this apply to your observations:
1. Same make and model behave differently on different altitudes
2.Different make and model behave different on same altitudes
3.Same make and model behave similar on same altitudes

If answer is yes for most cases, then you could say that nothing is wrong with most of air conditioners.
Wrong is their use in conditions for which they are not intended.

[Thermatech]

If the split systems are working in heating mode but only have indoor coil & air temperature sensors then the indoor unit must make an estimation of the outdoor coil temperature ?
So the outdoor coil temperature is actually getting very low & frost is building up on the outdoor coil but in this case the indoor unit cannot see this condition based on the indoor coil temperature ??

1/ what about seperation distance between indoor & outdoor units ?
These very small systems often do not have any suction line accumulator & can be overcharged when mounted back to back.
If the pipe lenght is excessive then the indoor unit may be seeing good hot gas temperature but because of the long pipe lenght & pressure drop the evaporating temperature may be lower than the indoor unit is estimating.
2/ What is the humidity / wet bulb temperature at your high altitude sites?
I would suspect very low humidity.
You wont be able to get any sensible info from the manufacturer but most of these units are made for comfort cooling in parts of the world where humidity is high.The units are not primarly made for heating operation & you can be sure any crude defrost program will be based on higher humidity & wet bulb temeratures than your high altitude site.

I would suggest you retrofit a heat pump defrost timer so that you can set your own defrost time & temperature. You can then ensure that the unit must do a defrost at the time interval you consider appropriate & a coil temperature that will ensure full defrost.

Years ago Ranco did a good model E15 which always worked well.

[Thermatech]

Do you check the power supply voltage at the units ?
As you work on remote sites is the power supply from generator ?

The manufacturer will design & rate the units for a voltage & will base the defrost stratagy program on the expected refrigerant temperatures with the compressor operating at the rated power supply voltage.
For example.
The unit is rated by the manufacturer at 230v.But on your site the generator kicks out 250v. The compressor is running faster & pressure / temperature of hot gas is higher. The indoor unit gets hotter gas then defrost stratagy is designed for so it does not need defrost but meentime at the outdoor unit the actual evaporating temperature is low partly due to the low outside air humidity & the outdoor coil is frosting up.

So different power supply voltage may be part of the reason why some units defrost ok & others do not.

[philjd26]

hi,
i was wondering,you say that your charge two identical units and get one ok and the other freezing up? are these units serving the same room?..if the humidity is low,install a humidifier into room, would imagine itchy eyes and dry throat...will give you more of a latent load on one side a least getting it more to normal design(if you think it might be an issue)...
but by the sounds of it some arent defrosting have another look for something on out door coil if der is nothing i would retro fit...

rgrds phil

[collo]

Hi Blokes,
I have a Midea 7kw wall split at my home, which I suspect has a problem deep in the internals of the compressor. That aside, the MO for these systems in a heating cycle is that the condensor fan cycles by a combination of pipe temps :ie: indoor and outdoor. Defrost cycle is then chosen both by time and evap. pipe temperature. The problem that I have is that the darn thing switches on the condensor fan at the start of the heating cycle, switches the fan of about 10 secs later and will not switch on again. The end result of course is that the condensor coil freezes. As I have tried everything from suck down and regassing to replacing capiliaries, replacing reversing valve and replacing PC the board. In the end I fitted the condensor fan with a pipe temp thermostat through a NO relay, with the fan call signal being from the Reversing valve terminal. The relay NC terminal fires the fan in the cooling cycle. Voila works like a charm.

[collo]

Oops seems as though I lost my last post in the ether. I have a similar problem with two identical machines, identical that is as from same maufacturer same size, same house, same conditions, same same same. One heats one wont. I have traced it to the condensor fan not cycling, so the outdoor coil freezes over. If the fan runs all the time, the head pressure goes too high. :mad:After trying everything and then some I fitted a pipe thermostat to the inlet to the Condensor via a NO contact relay. The relay/thermostat is then activated by the Reversing valve terminal. The NC contact fires the fan in summer. Voil;a works like a charm. :D

[collo]

Sorry fellows, both posts have now shown up, any comments?. yes I figured that I had an internal blockage hence my rebuild of the system, many times.:rolleyes:

[nike123]

Oops seems as though I lost my last post in the ether. I have a similar problem with two identical machines, identical that is as from same maufacturer same size, same house, same conditions, same same same. One heats one wont. I have traced it to the condensor fan not cycling, so the outdoor coil freezes over. If the fan runs all the time, the head pressure goes too high. :mad:After trying everything and then some I fitted a pipe thermostat to the inlet to the Condensor via a NO contact relay. The relay/thermostat is then activated by the Reversing valve terminal. The NC contact fires the fan in summer. Voil;a works like a charm. :D

You said that they are same and one work OK.
Did you try to swap PCBs and/or thermistors?
If you did, what is your findings on that?