Mitsubishi Ecodans ASHP issues...

[back2space]

Evening all, would appreciate your input and advice on the following issue from another forum: As usual I know your the best to ask and I have asked the op to sign up on here so any questions can be asked, I have quoted a few posts and replies etc.

There are also pictures of the unit included:

Hi guys, new to the forum, and please bear with me if I ask something here that's already been asked, I read everything up to post 830 and thought I'd skip ahead for some advice

Ok, quick run down of my install:
2 x 14kw EcoDans heating a 170m/2 house with 31 radiators, mostly doubles (!, yes I know), if you count a hall or a landing as a room then 17 rooms, no buffer tank fitted (as Mitsubishi say no need),
2 x 210 litre DHW tanks,
28mm, 22mm and some 15mm pipe, 4 x 15/60 Grundfos Alpha pumps (2 pull, 2 push).
No seperate zones, most rads have TRV, but all turned to "max on",
400mm loft insulation, 600mm solid stone walls uninsulated due to listing. Full double glazing.
SAP report said 15kw required, and walls have a U value of 1.9.
House is undergoing refurb and not lived in. Heating was set to 10'c, water from Ecodans 22'c @ 17'c ambient ramping to 38'c @ -10'c ambient. No DHW generating set at the moment.
House is an old croft in northern Scotland (KW11 6UA, near Altnaharra, you may have heard of it recently) Recent temps dropped to -16'c and hovered around -10'c for several days.

Last winter the units iced up and failed due to inadequate water flow. Replumbed and sorted that. Both Mitsubishi and the install company are now happy with the water flow. Last winter it got down to -22'c, but the heating was off due to the failures.

This year they have again iced up to the point that the air flow is completely blocked. They try to defrost but fail, go back to heating and build up more ice, try to defrost etc....
Honestly, the ice was a solid sheet 10mm thick covering the back and sides of both units, last year it did this so thickly the fan blades sheared off due to the build up. Caught it this year before that happened.

When we're there and the room stat is set at 18'c the Ecodans easily manage to hit target temp within an hour of a cold start (31 out, 31 return) and the house is up to temp in about 3 hours, so I don't think it's a problem of sizing. They get the DHW tanks up to 58'c in 30 mins when we need it.

When they do a defrost it can be as frequent as 10 mins apart, or as long as 45, I'm sure they've never run for 2 hours without one. It lasts about 2 mins, though I haven't timed it. Interestingly Mitsubishi website says 30 to 120 mins apart, definitely more frequent than that.

When it fails it's different: it says "defrost" on the indoor unit, the outdoor shuts down, then you can hear the compressor start, then it makes a huffing sound and stops. The indoor still says "defrost" but the outdoor is doing nothing. After a couple of mins they then go back to heating mode and the cycle starts again.....it doesn't throw any codes or warnings on the indoor units, and I don't want to take the covers off outside to check as I don't want them to say I've voided my warranty by doing so. Sometimes after several failed defrosts it does a good one and then dumps loads of water out. It seems to fail more over night than during the day for some reason.

While the failures are bad enough it's the electric consumption we're more concerned about. To keep the house at 18'c (overnight 14'c) when outdoors it's ~8'c took roughly 100 kwh / 24hrs. At our rate that's about £11. When people nearby in similar houses with similar insulation are spending about £6 / 24hr heating to 22'c and DHW we're more than a bit concerned! In the temps we've just been having I imagine it would have been substantially higher. At that sort of rate we'd be better off burning £5 notes to keep warm

The only thing I can think of is high humidity causing very frequent defrosts using more electric and dragging the COP right down as a result. We share the valley with a river and according to my weather station thingy humidity is always around the 90% area outdoors (30% indoors). Other than that I can't see any reason why our system doesn't work as promised, and neither can the engineers who have been.

The install company has passed all the info I give them to Mitsubishi, data from my weather station showing indoor & outdoor temps, humidity and wind speed, loads of photo's, and all my observations, and all they said was basically "they're working within parameters, stop bothering us, not our problem".
Real helpful.

The install company is baffled and we're getting a bit worried about our £14k system. I think the next thing they're going to try is to put a buffer tank in but I don't see how that's going to help.

The main problem as far as I can see is the failure to defrost. It's not the fact that the units defrost but fail to melt all the "frost" on them, when they do a succesful defrost they clear completely down to the fins.
It's just that sometimes they try to defrost but fail to complete the cycle, actually, they fail to start the cycle. The indoor unit says "defrosting" but the outdoor unit is completly shut down, doing absolutely nothing. It tries to start the defrost, makes a huffing noise and then totally shuts down. The indoor unit says defrosting for another two mins or so, then kicks back into heating mode. This then sucks more moisture onto the exchanger and it builds up a sheet of ice.
When the exchanger is completly covered, it keeps the ice well below freezing by trying to extract heat from it and any moisture that touches it then adds to the thickness.
Even if a succesfull defrost occurs it leaves a sheet of ice clinging to the frame over the exchanger but not actually touching it so it cannot be defrosted by the unit. This in turn leades to no airflow and extreme electric consumption. Add in ambient sub zero temps and the ice simply stays put. Mitsubishi initially said it's wind chill, there was no wind this time, and the ice was only on the heat exchanger end of the case, so I don't think that's the cause.

Thank you for reading this much, sorry, it did get a bit rambling in places, I tried to keep it short while putting all the info down

[back2space]

After seeing the pictures I can see that the ice build up on the coils is due to the snow as you also have ice on the surface on the centre of the fan inside which will be from snow blowing into the front and freezing on contact with sub zero temps.

The ice on the coils though I have witnessed on my own units, snow is sucked onto the unit and freezes on contact it then turns to ice when sun shines on it the next day etc. The water has also not drained from inside the unit probably because it has frozen underneath where it would exit from the holes. You need to perhaps put some salt or grit under neath the unit to try to stop this water freezing. My unit looked similar not as bad and only at the bottom inch of the coil due to snow fall behind the unit. The rest of the coil was clear though but you have had much worse temps up there. The blocked coil with ice has then caused air flow loss so the unit has had to work harder and draw more power. However ice should not form as thick as this and the defrost cycles dont look to be lasting long enough but it looks like the unit has been exposed face on into the snow as there is quite a lot on top of the unit as well. This depth of snow will have been sucked into the back of the unit at the same time and stuck straight away.

To me though as evidenced in the side view picture of the units you can clearly see that this is snow that has turned into ice. Snow fall and extremes of temps have been your issue here. The fact then that this hasnt been able to drain away out the bottom has caused these issues.

I was having to go outside and clear the build up of snow from the front and rear of the unit as mine isnt mounted off the ground like yours but my unit is sheltered at the front of the house so wasnt in the wind whereas at the rear we had big drifts of snow and at the front it wasnt too full on. If you was living at the property you perhaps could have been able to go out and assisted in removing some of the ice by pouring on warm water or something. I reckon the same will happen again if it snows bad again but I bet if it was -20C outside and dry you wouldnt have these issues.

I think snow is the culprit here. I dont think the defrost strategy can cope with such extreme conditions and full on snow.

[back2space]

There is suggestion that the defrost strategy is not working properly but mitsi say the unit is working ok and blaming wind chill which again is a possibility when the snow makes contact with the freezing cold casing and coils it quickly builds up then takes time to defrost.

As the op isnt living there at the moment he isnt able to go outside to ensure the drainage holes under neath the unit are clear of ice so that it can drain away any build up of ice etc.

Just to clarify a couple of things
We have two 14kw units so 28kw available to us, 15kw for the house according to SAP report so 28kw should be enough even in the very cold when the units drop off, say they drop 5kw each due to the cold and we should still be in the clear.

18'c is the temp of choice for us, it's nice and warm in the house at that so we're happy. The water outputs from the Ecodans are usually around 31'c and this works with our radiators fine. We have turned the stat up some times to 20'c when we're sitting watching telly (40'c output from them), by using the boost feature, this makes the rads hotter but uses more electricity, so we're happy with 18'c in general

- So it heats the house upto 18C & maintains it at this in a short space of time you say... This confirms that the system has enough capacity... this rules out it being undersized. Your flow temps are fine if they are achieving the indoor temps you are comfortable with.

Snow is not building up behind the units to any extent, we have them off the floor by 9 inches or so, the snow was only about 6 inches deep this year when they failed. Neither had snow fallen from the roof behind them to block up the breathing. There will obviously be some snow gets behind them and sitting on the outdoor pipe work, and snow will be sucked into them, but from what I have seen it's not snow that's causing this it's condensation on the heat exchanger that builds up.

Correct... snow being sucked onto the back of the exhanger can freeze it over in a short space of time - I have seen this with my own unit. It freezes on contact and in a short space of time it would just be recognisable as white ice like in your freezer.

grahamc2003,
nope, not foggy where we are, it was my thought initially as well. It's generally too windy for fog or mist to form, there is pretty much a constant steady breeze down the valley, though the humidity the weather station records is always very high. We have a large loch within the valley and the river draining it runs past my house to the sea 5 miles further up. We don't have salty air to contend with but this is northern Scotland, it only stops raining so it can snow!


The main problem as far as I can see is the failure to defrost. It's not the fact that the units defrost but fail to melt all the "frost" on them, when they do a succesful defrost they clear completely down to the fins. - When is this most noticeable? WHen its snowing, when the weather is cold but dry?

It's just that sometimes they try to defrost but fail to complete the cycle, actually, they fail to start the cycle. The indoor unit says "defrosting" but the outdoor unit is completly shut down, doing absolutely nothing. It tries to start the defrost, makes a huffing noise and then totally shuts down. The indoor unit says defrosting for another two mins or so, then kicks back into heating mode. This then sucks more moisture onto the exchanger and it builds up a sheet of ice. - This indicates a fault with the unit... it should complete its defrost cycle and shouldnt start another defrost for at least half an hour if it did it properly the first time. This is something that needs raising with Mitsi. The sensors on the coil and pipework inside the outdoor unit will detect certain conditions which will cause it to go into defrost... it will then stay in defrost until certain conditions (temps etc) are met and then switch back into heating. If it is below 0C the outdoor fan should not be running & only the compressor will be whirring away when it is defrosting to enable the heat to build up in the coil. If it is running during initial stages of defrost this suggests a fault with the outdoor air temp sensor.

When the exchanger is completly covered, it keeps the ice well below freezing by trying to extract heat from it and any moisture that touches it then adds to the thickness. - Spot on

Even if a succesfull defrost occurs it leaves a sheet of ice clinging to the frame over the exchanger but not actually touching it so it cannot be defrosted by the unit. - Ice should never build upto this level anyway... if defrost cycle worked properly it would never be an issue.

This in turn leades to no airflow and extreme electric consumption. Add in ambient sub zero temps and the ice simply stays put. Mitsubishi initially said it's wind chill, there was no wind this time, and the ice was only on the heat exchanger end of the case, so I don't think that's the cause.

[back2space]

Another quote and reply from the op:

Another thing, these units a brand new. Mitsi did a recall this year due to some faulty compressors or somesuch that could cause the units to literally blow up!
Both my EcoDans were commisioned in early November, they ran for two weeks unattended keeping the house at 10'c with ambients above freezing, all ok. Then we had that cold snap, they were left for 12 days before our next visit.

When we got there we found them partly covered in ice, but reading the electric meter we found that they had eaten 978 kwh (21/11/10 - 4/12/10) or £9 / 24hr average. To keep the house at 10'c!

So I checked the data from my weather station, cheap thing from Maplins but still accurate enough for this. It shows that for a while everything was fine, the system was keeping the house at a steady 8'c. But then the house temps start to drop off, eventually hitting -5'c indoors.

I assume that this is the time that they iced over outside and then just started chewing electric to try to heat the place. I reckon that that £9 a day isn't right, I think it was actually low until they iced up and then it soared as the COP hit 1:1 or worse.

My weather station then records a couple of days above freezing outdoors and some strong sun on the units. I think this has partly thawed them and allowed them to start working again.

Mitsubishi have essentially told my install company it's not their problem, I've seen the emails, and the installers are scratching their heads.
Looks like it may be GSHP for me in the future..

[Bigfreeze]

First of all you should never have been sold a HP system on a house with rads with the insulation levels you stated. Your bills were always going to be astronomical. 28kw for 170sqm even at your levels of insulation is crazy tho.

Secondly, regardless of what Mitsubishi say, a buffer tank must be used in any system where you have no heat retention capacity such as a UFH screed. The units will cycle too frequently and your compressors will be shot in 5/6 years.

Thirdly, and i'm already on the record for stating this in other threads, the japanese "air to water" units are junk. Incapable of handling extemely low temps and defrosting far too often. Making ridiculous claims about their capabilities.

Your defrosting problem may be a sensor issue but apart from that your best bet is to retain one of the units to cover you're heating until the temp drops below 5C then have a oil or gas boiler back it up. Also, you must fit a buffer tank to the system.

[Bigfreeze]

Just read you're last line. If its at all possible to get these units taken back jump at it. Get a GSHP, but you should really look at your insulation levels because 1.9 is very high.

[desA]

Can the OP please measure the evaporator fin spacing? In my view, this will be a very important factor. For these extreme temperatures, defrost capability will be determined by fin spacing, as the cores already look to be fairly thin (~ 1-2 rows).

The location directly in the snow ingestion zone is absurd. Why not set up a simple canopy, or shelter above the units? Alternatively site them in a slightly more sheltered position.

[back2space]

I have asked the op to register and post on here.

[desA]

What kind of defrost mechanism do these heat-pumps use?

[mad fridgie]

I have seen this happen many times, and tends to happen when you have light floating snow, which is drawn into the coil. You end up with an ice bridge between the coil and the casing. In the early stages the coil after defrost is actually clear of ice, apart from the bridge. The bridge gets bigger, reducing the airflow, increasing ice build up, to the point where the defrost is unable to clear the coil. A cowl to reduce the snow entering is a start. I do not mitsi, so I do not know if you can adjust your termination temp or duration of time

[TiredGeek]

Hi guys, my units

As some have said, it could well be snow this year, the units were left running unattended for 12 days during one of the coldest spells for years.
There wasn't a great deal of snow for the area, only about 6 inches, but the temps dropped a lot lower than is normal at this time of year, we normally don't get down to the -20's 'till January / Feb. There was virtually no wind though which is unusual as it usually howls down the valley....

When we had a site survey carried out we discussed positioning of the units and it was decided this was the best place for them as they wouldn't sit in stagnant cold air as there is always a breeze around them here. We were also assured they wouldn't need covering or sheltering in any way, except maybe to stop the snow from the roof above falling behind them.

I'm now of the opinion that no matter what is done to them now, fundamentaly they are not suited to a life that far north in Scotland in the conditions in the valley where my house is.
I think I'll be contacting trading standards for advice to find out my next course of action so that I go about it the right way to ensure a good outcome for me

[TiredGeek]

Just read you're last line. If its at all possible to get these units taken back jump at it. Get a GSHP, but you should really look at your insulation levels because 1.9 is very high.

Yeh, I'm thinking now I'm gonna reject the units as unfit for purpose and push for the company to fit a GSHP at their expense if it costs more than I've already paid.

I'm afraid there's less than no chance to insulate the walls, it's a listed building and they have been sticklers on maintaining all of the "historic" features, even down to what paint we can use. Add that to the fact that they are solid stone about 600mm thick and using lime mortar and plaster and we're extremely limited in what we can do. They did let us replace all the doors and windows with double glazed affairs which has helped reduce drafts greatly. They actually asked us to do this as the ones that were in were very poor copies of the old ones and "the historic timeline couldn't be read clearly". New windows and doors look new and so cannot be confused with the originals, they say. It made sense at the time, honest.

We initially wanted to go for GSHP but a combination of remote location (making it hard to get the trenches dug and sand in them etc) and dry sandy soil put us off and made us look into alternatives. We were promised a system that would work just as well but very slightly more costly to run. Looks like it's failed on both fronts so far.

Can anyone advise me if dry sandy / stony / peaty soil would be any good for a GSHP as I keep getting told different things depending on who I ask....

[back2space]

I personally think you need to get back on at these builders, these units do not appear to be defrosting properly and shouldnt have so much build up of ice on them.

[Bigfreeze]

Yeh, I'm thinking now I'm gonna reject the units as unfit for purpose and push for the company to fit a GSHP at their expense if it costs more than I've already paid.

I'm afraid there's less than no chance to insulate the walls, it's a listed building and they have been sticklers on maintaining all of the "historic" features, even down to what paint we can use. Add that to the fact that they are solid stone about 600mm thick and using lime mortar and plaster and we're extremely limited in what we can do. They did let us replace all the doors and windows with double glazed affairs which has helped reduce drafts greatly. They actually asked us to do this as the ones that were in were very poor copies of the old ones and "the historic timeline couldn't be read clearly". New windows and doors look new and so cannot be confused with the originals, they say. It made sense at the time, honest.

We initially wanted to go for GSHP but a combination of remote location (making it hard to get the trenches dug and sand in them etc) and dry sandy soil put us off and made us look into alternatives. We were promised a system that would work just as well but very slightly more costly to run. Looks like it's failed on both fronts so far.

Can anyone advise me if dry sandy / stony / peaty soil would be any good for a GSHP as I keep getting told different things depending on who I ask....

How dry is the soil? Moist is good, bone dry isn't but neither is sopping wet. You're looking for a good draining soil that maintains some water but not saturated, thats the ideal.

Get rid of these things, just go through this forum alone to see the amount of problems with them. We have air to water units working at temps of -17 this year without any problems but the evaporator on the units we use is about twice the size of the whole outdoor unit in a Mits and the fin spacing about 3 times as wide. You will continually get this problem year in, year out unless you either back one of these units up with a oil boiler when the temps drop, as i suggested earlier. Or preferably get a GSHP installed, with a buffer tank.

I would imagine somewhere about 14kw will cover your heatlosses, so your electricity bill will more than half

[freedom HP]

i note with interest some of the comments above. Freezing like this is usually an issue of lack of capacity. This can be caused by applying the wrong unit or refrigerant shortage or even lack of flow rate through the water circuit. Ive seen hundreds of heat pumps by many different manufacturers that have plodded happily on through the last period of bad weather. Measuring fin spacing and questioning the defrost is a load of bull, Mitsi sell hundreds of these units and there is no inherent design problem with them, they are good units if applied well. Some freeze like this its always the same problem. The units are not big enough.

to test this theory is really easy, next time its cold switch off a 25% of the radiators connected to the system, does the unit maintain the radiator temperature at the set point in the areas where the rads are running? if so it can cope and there are no issues with refrigerant charge etc.

im not a mitsi man but check the data tables for the kit, what capacity should they give in low ambient, check this with the duty being given out, this is only possible with a flow meter and two thermometers and a calculator. As a rough guide most so called 16kW units only give 10kW at -2 ambient if you allow for defrost.

good luck and if you need more help let me know.

[Bigfreeze]

i note with interest some of the comments above. Freezing like this is usually an issue of lack of capacity. This can be caused by applying the wrong unit or refrigerant shortage or even lack of flow rate through the water circuit. Ive seen hundreds of heat pumps by many different manufacturers that have plodded happily on through the last period of bad weather. Measuring fin spacing and questioning the defrost is a load of bull, Mitsi sell hundreds of these units and there is no inherent design problem with them, they are good units if applied well. Some freeze like this its always the same problem. The units are not big enough.

to test this theory is really easy, next time its cold switch off a 25% of the radiators connected to the system, does the unit maintain the radiator temperature at the set point in the areas where the rads are running? if so it can cope and there are no issues with refrigerant charge etc.

im not a mitsi man but check the data tables for the kit, what capacity should they give in low ambient, check this with the duty being given out, this is only possible with a flow meter and two thermometers and a calculator. As a rough guide most so called 16kW units only give 10kW at -2 ambient if you allow for defrost.

good luck and if you need more help let me know.

They are inherently flawed, in that, they were never designed primarily as a Air to water HP, they were designed as AC units. Besides a few electrical changes the outdoor units are exactly the same as the AC. The fin spacing and defrosting requirements are hugely influential in the efficiency of a ASHP. If design does not matter, then we may as well bang a screw compressor and heat exchanger inside and a Searle Evap outside and Voila a perfect ASHP.

28kw on 170sqm, not covering the heatload and you think its under-speced??? Thats 165W per sqm. Even given the weather conditions, unless the guy is living in a wendy house, the units are still overspeced.
The only wrong specification, was that of coupling a HP to a radiator system in a house of that type.

[Gary]

The bottom line is: The defrost is not working properly. Why?

[desA]

Incredibly tight fin spacing on the evaporator, as seems to be the case, is hugely influential on the rate of ice build-up, & the ability for the defrost system to clear it.

I differ with the views in post #15, in this regard. Perhaps this poster could provide more detail on his theories.

I also believe that the situation can be remedied with a few minor modifications. Pulling them out & replacing with GSHP seems to be an incredibly drastic step, in my view. Give the units a decent chance.

[desA]

We have air to water units working at temps of -17 this year without any problems but the evaporator on the units we use is about twice the size of the whole outdoor unit in a Mits and the fin spacing about 3 times as wide.

Yes, indeed.

[ashp 4 life]

This model of unit are installed throughout europe. Including Sweden where my mate installs them. I think blaming the kit is the easy way out, common mistake made by a lot of people.

I'd suggest looking at the bigger picture. Not just the design, but the how the system is being used. If you set the space temp at 10 when it's been designed at 20 then the load is going to be less than half what the SAP calc suggests.

SAP calc is for 15kw and you have 28kw installed. Oversized???

[Oakgreen16v]

The 4th picture shows one unit is clearly defrosting better than the other. Blaming the design of the evaporator and fin spacing is not a valid arguement for this issue. As other posters have identified there are thousands of this model of heat pump in the UK and when the system is applied and installed correctly there is no problem with defrost and the frequency is should certainly not occur as often as has been suggested. The fact that there are percieved to be a greater volume of problems reported on this forum with Jap heat pumps (mainly Mitsi and Daikin) is explained simply by the shear number of their units in the market. (Mitsi account for 34%, Dakin account for 19% of all ASHP sold to the UK market). Inverter driven units do not require a buffer vessel as the load can be changed by altering fan and compressor frequency direct from the unit, however it is not to say a buffer would not have helped in this case but then again the Ecodan is not intended to be joined together, although it is possible.
The problem here is very likely down to lack of flow rate combined with a lack of understanding with regards to the design and application for this system.

[Gary]

Again... it is not defrosting properly. The defrost cycle terminates before the coil is fully defrosted. What terminates the defrost cycle?

[r.bartlett]

This model of unit are installed throughout europe. Including Sweden where my mate installs them. I think blaming the kit is the easy way out, common mistake made by a lot of people.

I'd suggest looking at the bigger picture. Not just the design, but the how the system is being used. If you set the space temp at 10 when it's been designed at 20 then the load is going to be less than half what the SAP calc suggests.

SAP calc is for 15kw and you have 28kw installed. Oversized???

28kw at what outdoor temp?

Oh and it's nice to see a couple of Mitsi UK guy's on here!

[desA]

Again... it is not defrosting properly. The defrost cycle terminates before the coil is fully defrosted. What terminates the defrost cycle?

Makes me wonder what the defrost temp sensor is sensing. Things like on/off settings for the defrost cycle itself, would also be of interest.

Are these hot-gas with 4-way valve defrost systems?

[monkey spanners]

If the problem was down to the fin spacing, and it seems to be accepted that these units are heavily based on their standard split units then surely we should be having the same problems with split systems when heating in the same weather conditions?

[back2space]

If the problem was down to the fin spacing, and it seems to be accepted that these units are heavily based on their standard split units then surely we should be having the same problems with split systems when heating in the same weather conditions?

I agree and my system has been running fine!

[mad fridgie]

A couple of points

The heating units will never be the right size, due the fact the heating load is always varying, it would seem to me that this house requires 500w per degree C of temperature difference (inside/outside)
The heat pump rating is just that a rating at one condition, the client may have 28kw but only at the rated condition, as the ambient changes so does the heating output, basically for every degree C you drop in ambient you loose 2.5% duty and i would also suggest you loose 20% due to defrost (to high i hear you say, the 20% is split, time not heating and heat removed from the heating circuit) This is based upon taking Kw averaged over an hour (further explanation if required)
Re fin spacing, different agruments are available, yes closer fin spacing better heat transfer, but is more suseptable to reduced airflow in frost conditions.
It is a balance (diversity) of the ambient conditions,cost and size. Even in the UK the average ambient is not that low, so on average small fin spacing is acceptable, at times performance will be poor.
Defrost, yes modern designs should cover automated defrost (not relating to performance) and I agree high flows of water should flow the heat pump at all times when the system is running. However in this case ice bridgeing is occuring, which is caused by external factors "snow", how is the software to know if extra defrost time is required to cover this eventuality. I know on older systems, you could extend the defrsot time and increase the termination temperature, which did cure the problem, but was highly in efficient at other times.
We all know when an evap is frozen solid, a standard defrost is never going to clear it! (the fridge guys will understand this more)

[r.bartlett]

If the problem was down to the fin spacing, and it seems to be accepted that these units are heavily based on their standard split units then surely we should be having the same problems with split systems when heating in the same weather conditions?

Crumbs we've had loads of callouts on frozen condensing units....

[Bigfreeze]

If the problem was down to the fin spacing, and it seems to be accepted that these units are heavily based on their standard split units then surely we should be having the same problems with split systems when heating in the same weather conditions?

Splits don't put in the running hours that heat pumps do. They are completely different applications. Making that argument is akin to saying I should use an AC to chill a cold room.

[Bigfreeze]

A couple of points

The heating units will never be the right size, due the fact the heating load is always varying, it would seem to me that this house requires 500w per degree C of temperature difference (inside/outside)
The heat pump rating is just that a rating at one condition, the client may have 28kw but only at the rated condition, as the ambient changes so does the heating output, basically for every degree C you drop in ambient you loose 2.5% duty and i would also suggest you loose 20% due to defrost (to high i hear you say, the 20% is split, time not heating and heat removed from the heating circuit) This is based upon taking Kw averaged over an hour (further explanation if required)
Re fin spacing, different agruments are available, yes closer fin spacing better heat transfer, but is more suseptable to reduced airflow in frost conditions.
It is a balance (diversity) of the ambient conditions,cost and size. Even in the UK the average ambient is not that low, so on average small fin spacing is acceptable, at times performance will be poor.
Defrost, yes modern designs should cover automated defrost (not relating to performance) and I agree high flows of water should flow the heat pump at all times when the system is running. However in this case ice bridgeing is occuring, which is caused by external factors "snow", how is the software to know if extra defrost time is required to cover this eventuality. I know on older systems, you could extend the defrsot time and increase the termination temperature, which did cure the problem, but was highly in efficient at other times.
We all know when an evap is frozen solid, a standard defrost is never going to clear it! (the fridge guys will understand this more)

A decent control system reading air off temps offset against your weather compensation will tell you HP when it needs to defrost.
These units don't work to that logic. The fact that flow is restricted alot of the time when they do defrost (or they have poor flow temps to start off with) means the unit is usually running the compressor at a lower speed, which means less defrosting power, which means less actual defrosting in the time alotted.

[wozza]

I would suggest a snow canopy, and take the guard off the back of the unit so the droplets of water dont turn to ice after defrost and bridge between the coil and guard.
I would definatley look at the defrost strategy and see why it is terminating defrost early, could be lack of water flow as the indoor heat exchanger is getting to cold and terminating the defrost.

[monkey spanners]

Crumbs we've had loads of callouts on frozen condensing units....

See thats where using logic get me....

I guess none of us can say whats wrong with the system without a site visit and a good understanding of these units and their proper aplication, which i have'nt done and don't have....

Jon

[monkey spanners]

Splits don't put in the running hours that heat pumps do. They are completely different applications. Making that argument is akin to saying I should use an AC to chill a cold room.

Would a split systems run time not be a function of its duty, the heating load and the thermostat setting?
Are short run times an inherent quality of spit systems?

Jon

[back2space]

Would a split systems run time not be a function of its duty, the heating load and the thermostat setting?
Are short run times an inherent quality of spit systems?

Jon

My multi split system during the cold weather was running pretty much 12 hours a day none stop then on and off through the night as the set points were lower.

It certainly was not cycling off during the day, ramping down yes but not cycling off.

Now we have milder temps it is cycling off.

[desA]

Reminds me of why the US closes airports during the just-below-to-just-above freezing weather conditions. Excessive ice formation on wings & run-ways - pretty uncontrollable.

In Sweden, for instance, the first snows are wet, slushy & miserable. Once the temps drop well below freezing, the stuff turns to ice & things are great. It's that intermediate time that is utterly miserable.

Perhaps system designers need to think a little more about this cold-not-cold condition & prevent ice formation in the first place?

A few reasonably simple solutions present themselves here.

[back2space]

Anyone seen the Daikin Altherma picture in one of the other posts... worse than the pics above...

http://www.refrigeration-engineer.co...ad.php?t=29513

[wozza]

At first glance it looks installed incorrectly.

Daikin specify the outdoor unit should be mounted 300mm off the floor, so the ice does build up into the unit.
It would also be advisable to fit a base plate heater

[charlie patt]

do you mind me asking what is a base plate heater a form of heater mat? does it only work at low ambient

[wozza]

Charlie,
Your right, it is a heater mat that switches on via the outdoor air sensor when it gets below three degrees.
There is also a heater tape in the kit that does the same, this is installed between the bottom of the coil and the base plate.

[charlie patt]

available via mitsi or a wholesaler?

[wozza]

Charlie,
I,m not sure what is available via Mitsi, the base plate heater that i am talking about is from Daikin.
I know it was amitsi thread but he picture posted by back to space was a Daikin unit.

[charlie patt]

ta for that your never to old to learn

[fixmyheatpump]

Dear Tired Geek,

This is my first post on here mainly fueled by frustration and annoyance after reading about the problem you are experiencing along with some of the advice being provided to you in this thread.

It's obvious you have spent a lot of time writing about your problem on here because you need help and advice having spent a lot of money on a heating system which is not doing what it is meant to do. I know some of the people here are genuinly trying to assist but disappointingly some of the posts are obviously written by non-technical sales people hell-bent on rubbishing either air source heat pumps as a technology or Mitsubishi's Ecodan as a product.

We install, service and maintain a few types of air to water heat pumps and I can assure you Mitsubishi are one, if not the best on the market. I can also say if this 14KW model didn't work down to -25C then they wouldn't put it in their literature. My reasoning behind this is the fact they have just recalled ALL of their 6,000(ish) units because of an isolated issue with ONE system, also if they were being deviant why would they quote -25C when we hardly see this temperature in the UK?

If you would be willing to message me your phone number I'm sure we can assist you with the problem you are experiencing. It would be good to get more information about the whole system and exactly how it's been designed and hydraulically configured over the phone. If it can't be fixed from our conversation we would be willing to give you a fixed (cost only) quotation on a no fix no fee basis to come up and put the system right for you, i'd expect it to be a days work + travelling.

[TiredGeek]

Well the installers have been intouch. They're going to have a good poke about the system with a Mitsubishi tech and see if they can find the problem.

I think it's bridging like other people have said. Nothing can be done about it, they're simply no good for my particular case, probably a very good system for somewhere that isn't as extreme.
Just no good for somewhere that drops below -15'c and then stays around -5'c for weeks at a time.

I honestly don't think they'll ever get it to work reliably, time to push for a GSHP system I think, at least that won't be affected by snow and ice

[Gary]

I think it's bridging like other people have said. Nothing can be done about it, they're simply no good for my particular case, probably a very good system for somewhere that isn't as extreme.
Just no good for somewhere that drops below -15'c and then stays around -5'c for weeks at a time.

But... but... but... The defrost cycle terminates before the coil is fully defrosted.

[desA]

Again, what is the defrost temp sensor sensing?

Gary, perhaps you could walk us through the exact mechanics of how such a defrost system would initiate, run, then terminate? There will also be settings/parameters in the OP's specific system that can be checked.

Tired Greek could then ensure that his system people answer these questions when they review his system.

[Gary]

I have never worked on an Ecodan, nor seen a wiring diagram or a manual. I could only guess at the control strategy, sensor placement. control settings, etc.

What I do know is that if the coil is not fully defrosted, then its heat output will be severely limited.

[yinmorrison]

For what it's worth , my experience with all Manufacturers of A/C Heat Pumps is that during really cold spells ( about -8C and lower ), no matter what the defrost sensor termination temperature is moisture is retained between the fins on completion of defrost cycles and as soon as the unit cuts back on to normal operation the water freezes up almost immediately. The sensor itself may on a section of the coil that is clear of water and by the time it detects a further defrost is necessary ,areas of the coil are pretty solid and never actually clear at all , thus the problem becomes worse. I have actually seen where it has been so bad the coil tubes are crushed and damaged beyond repair as the pipe is so thin. Why else on a Cold Room evaporator would the fins be 5 or 6 per inch.

[Gary]

The sensor itself may on a section of the coil that is clear of water and by the time it detects a further defrost is necessary ,areas of the coil are pretty solid and never actually clear at all , thus the problem becomes worse.

The sensor then should be relocated to the section that clears last. However, I'm betting the manufacturer has the sensor located in the optimum position. Possibly it has come loose or the termination temp is incorrectly adjusted.

[desA]

Some defrost temp sensors are placed on the evap elbows. When the evap temp drops below a certain value, they begin the defrost cycle.

If the ice builds up & begins blocking the evap fins, it would be expected that the evap tube temp would drop - allowing defrost cycle to begin again. There would be no direct way of 'knowing' that ice had formed in the fin section.

Perhaps some manufacturers use more sophisticated methods?

This is a very interesting technical area to explore.