Multisplit excessive defrosting and long heat-up
Hi All,
I have issues with a recently installed 3C30NAHZ4 with three heads.
Summer went well in cool mode.
I am in my first winter heating with the unit.
Below -15c (5F), the unit cycles very often into defrost. At -18C the other day, I was getting 15 minutes of “heating” (actually 10 minutes of cold / ambient air and 5 minutes of starting to heat) followed by 10 minutes of defrost; rinse and repeat …
When it defrosts, the outside coils are not frosted at all, compared to expected defrosts at milder temps where there is visible frost buildup
The house temp goes down drastically and raising set point does nothing for this.
Installer came over, does not suspect refrigerant charge as when it DOES heat, temps are very good. I asked if there could be an issue with thermistors falsely flagging a temperature or temp delta that calls on a defrost, they didn’t’ think so.
Instead, they mentioned that Apparently before Christmas they were made aware of a required electric control board replacement (I guess for firmware / software update). Some other clients had it replaced as they were getting 20 minutes of heads producing cold air before heating up!!
Should have it replaced within the next month…
I was wondering if anybody fell upon the same or similar problem and if the board fix is effectively the right one?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I’ve read some other complaints of very same issues. This is not normal and does indicate possible firmware issue. I’ll make some inquires but it does sound like firmware/board issue.
Interesting I have similar defrost frequency and slow heat recovery issues with my ducted MXZ-SM48NAMHZ that has behaved like this for a year since install. I have been getting the run around since for a year being told this is working correctly and no resolution. Just had a small leak repair last week and recharged to spec so very unlikely low refrigerant is the issue. At 27f with high humidity I got 10 defrosts in 6 hrs. Nearly an hour of blower off completely, 2hrs at very low output and half the time blowing heat. Added temp monitors a month ago to try and figure out what was going on. When colder, below 20 cycling is similar and recovery to heating is worse so my house starts losing temp.
Curious if your unit is cycling similarly.
Hi Dave, from what I can see in the charts yes it does look similar.
You barely hit 110 F duct temp and it goes back into defrost.
I run a non-ducted system but the behaviour seems similar.
Defrost frequency varies, your 10/6 = 1.66 defrosts per hour, in worst cases I’ve been getting three defrosts per hour.
I am waiting on my installer to receive the new control board and install, I will keep you posted. We have some very cold days coming up (getting close to the unit’s -13F capacity), so it would be good testing if they manage to change the card in time ..
Cheers
And yes … colder temps worsen heat recovery…
I am wondering if this is normal.
That is, what should be the expected time for a head (or duct) to start blowing warm/hot air.
@Mitsuman??
Wow, three defrosts and hour is dreadful. I really hope you get your unit fixed and working correctly before the very cold temps. Thankfully my defrost frequency is not that high and it can keep up at 20 even with the slow recovery.
That doesn't look like defrost to me.
Typical defrost is a short pause, run at full power, short pause back to old power level. That power trace looks more like the unit cycling. A bit strange in this cold of a weather.
Appreciate the comment and it is definitely defrosting as the blower turns off (which it doesn't do when cycling) but agree the power usage is very strange (also didn't show power above above but near perfect alignment with supply temp or btu output). I only have 5 min resolution on power but I have a 25 min period per defrost cycle with a decrease in power usage. 15 minutes at the bottom and a 5 min period in between the high and low on both sides. Getting it looked at next week but so far no insights into issue.
I am experiencing very similar behavior from a recently installed MXZ-3C30NAHZ4. Defrost cycles are occurring at intervals that range between a minimum of 25 minutes and maximum of two hours when temperatures are less than ~10F (temps have not gotten less than -10F). Two connected indoor units (MSZ-FS12NA and MSZ-FS18NA) are both capable of producing 115F degree air but their outputs are inconsistent; sometimes after a defrost cycle they produce lukewarm air or stop blowing air altogether such that the room temperature drops 20 or more degrees depending on the duration of the incident. Please do share your updates as you get them. I have a feeling that the solution to your issues will be applicable to mine.
Hey,
Your unit is the exact same as mine.
However in heating I run a 12 head and a 15 head instead of your 12+18 (a third 12 head is off since heat rises)
They are GL model instead of FH
Sure does sound like same problem… seems to be a trend!!
Can’t wait to have my firmware changed and confirm to you guys whether this is any help.
This night was terrible …
-17.5C ambient (approx 0F)
10 min defrost, 17 minutes (!!!) of cold / lukewarm air, 7 minutes of warm (not super warm air), defrost.
So cycling practically twice an hour with 25% “heating” cycle. House temp dropped 4 degrees C.
Can’t wait till they replace the stupid firmware card.
Confirming this is firmware issue indeed. Headed out tomorrow with Mitsubishi to reprogram a unit doing the same exact thing. It’s funny, it is NOT a typical defrost cycle. Almost fooled me the first time, the compressor did NOT ramp up to typical level during defrost it was spinning much slower….however, reversing valve actuated and indoor vanes pointes straight out, outdoor fan stopped. About 4-5 minutes then back to heating. I observed and documented about 5 cycles, all about 15 minutes apart. Yes the new outdoor board will fix it if they do not have access to program tool. I think there are less than 10 in the US so most likely board is fastest option to resolve. They’ll get you fixed up then you should be good to go. Very rare to have an issue like this but nothing and no company is perfect. It is an unfortunate inconvenience for sure.
Ok so this night we hit a low point of -18c (0F), temps at which the system was problematic.
(I honestly haven’t had time to check the system’s functioning now in detail, but I definitely saw differences last evening when it was warmer (Around 23f). System no longer shoots cold ish air for 10 minutes after defrost but seems to wait till coils are hot enough ..)
Anyways, normally at -18c ambient (0F) I would have to program set point to 22 or 24c (71 to 75f) and hope for a house temp at 20c (68f) in the house, but would usually wake up to 18-19c (64-66f) and a defrosting mini split.
Well, I decided to go all in and leave the setpoint at 20c (68f).
Go downstairs and …. Ambient -18c (0F) and house temp… 21c!!! (70f). TOASTY
I don’t want to celebrate too early but…. The fix seems to have worked!!!!!
ANYBODY DEALING WITH EXCESSIVE DEFROST CYCLING AND UNDERPERFORMING MITS MULTI HEAD HYPER HEAT UNITS IN HEATING : GET YOUR FIRMWARE CHECKED!!!!!!
Just wanted to thank you for this update! We just had two boards replaced on our two MXZ-2c20NAHZ4 units. I showed this thread to our installer, who first had to verify the lines weren't kinked and even replaced the charge in case it was contaminated. After four visits they contacted the distributor who confirmed the firmware issue and sent out new boards. Mitsubishi has been taking in old boards, reflashing them, and putting them back out into the field pretty regularly from what he heard.
After the first board was replaced, the defrost cycles stopped their every-15-minute airplane-toilet flush cacophony and the heads are working like a charm.
My situation was precisely as described by several. At below -8C twice per hour "defrosts" (although almost zero frost on condenser) with extended periods of cool air from the heads, result: 10-15 minutes of real heating per hour. The contractor replaced the main circuit board last March (after 4 months of head scratching, we have an electric furnace too and a Napoleon FP insert). Still same poor performance this past December and they replaced the board again with result that it will now sometimes run 2 hours hot without pause. Not burning any wood but mild winter. At -12-15C the main floor is a bit cool but the unit is working OK though not as continously.
I think the system was undersized by the contractor: 30K BTU condenser with 15K head for 1200 sq ft log house main floor + equivalent of another 800 sq ft of cathedral ceiling and loft. The 9K heads in top floor bedroom and the mostly above ground finished basement have no problem maintaining temps. Problem is, can only go to 18K head on main floor without also increasing the diameter of feeder pipe. Would that be worth a try? Have not a serious conversation with contractor due to lack of prolonged cold periods.
Any wisdom would be appreciated. Thank you Jay
What is the possibility I have this firmware problem with a 3C30NAHZ4 installed August 2024?
My HVAC contractor had two guys for 1.5 days yesterday looking for leaks, recharging, etc. But now I'm getting similar performance with the system regularly going into defrost without any frost on the coil.
Second question, how best do I present this to the Mitsu diamond contractor? Without appearing as the customer who attend too much time on the Internet?
I attached a photo of a power meter monitoring the circuit for the Mitsubishi. This shows is time is around the 8 hour mark where it starts franticky defrosting. I have another photo, but couldn't figure out how to upload more than one.
Update: last evening around 7pm it started heating. Spent much more time blowing warm air, and no longer spending 2/3 of it's cycle doing nothing(defrosting?) and then blowing cool air.
Could the problem return?
Here is a graphic of power consumption...
Final update (I hope):
My 3C30NAHZ4 has been working fine for over a week. I learned three things: the board needed to be replaced, the wall mounted units aren't nearly as good at circulating air as my previous system, and I needed to learned that the temperature sensor in the the wall mounted units, or their position, is terrible relative to my previous mini split. This 3C30NAHZ4 is a replacement. The sensor issue results in me needing to increase the set point the more the temperature drops below about 30°F. The circulation issue has resulted in at best a 2-3° difference from crown molding to baseboard, and at worst a 14° difference. The difference increase as outside temperature drops. As the difference gets large I've come to running a ceiling fan on low blowing up toward the ceiling.
In a nutshell, with my active participation (regular set point changes and ceiling fan use and disuse) I've got this system working effectively, albeit I have no idea yet how efficiently. In that regard I've attached a power usage graph. The 18k wall mounted unit in living open area generates this regular pattern, the bedroom heads 9k are irregular upping the power usage more rarely when they come on. The living area head comes on regularly. Thoughts?
Thank you for the responses to my inquiries, and for hosting these discussions. I found it all very helpful.
Mitsubishi is coming to look at mine this Tuesday. Last winter they replaced the board on the outside unit for excessive flushing noises, but yet it never really heated very well when temperatures dipped below 37 and still doesn’t.
You have the typical short cycling issue seen with these where the indoor head doesn't measure room temperature properly.
Couple of solutions. You have found the easiest one which is to run the ceiling fan.
Another is to install a remote room temperature sensor to the unit or install a wired thermostat.
https://mylinkdrive.com/USA/M_Series/R410A_Systems-3/Outdoor_Equipment/R410A_Outdoor/M21EAA307?product&categoryName=R410A_Outdoor
https://mylinkdrive.com/USA/M_Series/R410A_Systems-3/Outdoor_Equipment/R410A_Outdoor/MHK2?product&categoryName=R410A_Outdoor
https://mylinkdrive.com/USA/M_Series/R410A_Systems-3/Outdoor_Equipment/R410A_Outdoor/PAR_42MAAUB?product&categoryName=R410A_Outdoor
The thermostat can be configured to use the temperature sensor at the thermostat for controls which will fix the issue.
This is a pretty common problem with wallmounts and heat.
Show them the power meter trace. That is not normal, they can get on the phone with Mitsubishi and verify if it is the same firmware issue.
Which side, left or right, looks abnormal? Or both?
Thank you
Here's a larger timeframe...
If that cycling is going into defrost (ie you can hear refrigerant flow reversing and the 4 way valve switching), that is not normal.
Maybe 1 or 2 defrost cycles per hour around freezing weather max.
There is a firmware upgrade addressing that exact issue. Depending on the control board version it should either be flashed or physically swapped.
Are there information and instructions online for this firmware update? Thanks
No. Tech needs to call Mitsu tech and they will run him through the troubleshooting tree resulting in either a pcb replacement or firmware flash.
Do you have any comment on my "update" comment #24, Jan 10?
I had an issue last winter with excessive flushing noises and they replaced the outside board and that issue went away. Although it never seemed to produce sufficient heat below 37F. This winter has been horrible! Whenever the temps drop below 37F you can see on the screenshot my setpoint is 67F and the indoor temp never gets close to that until the outside temps start rising. I've complained to my installer and they've been told by their Mitsubishi rep that the board needs replacing. HOWEVER they are saying there are no boards available for the next 90 to 100 days!!!! By then I won't have the stupid below 37F issue.
Update:
My installer replaced a board, the last one the distributor had, but the problem persists. After they installed the board it blew warm air 120F +- 5 for 2.5 hours and warmed the condo right up. Then it started repeating a ~43 minute cycle: 15 minutes with no blowing and horizonal vanes, then vertical vanes ~70° air rising to ~120° while fan speed increased. Upon reaching peak temp and peak fan speed, the cycle would repeat. During this the condo temp would fall. It was 25°F outside. I removed power from the system for 15 minutes. Upon returning power it repeated the entire thing: 2.5 good hours, then temperature falling cycling.
Unusually it was 24-25° outside all afternoon & night. Rarely that consistent. Which I think is why what I witnessed was so consistent.
See graph. They finished installing board around 1:45, great for 2.5 hrs(ending 4:30), bad cycling until 8:45 when I powered it down. Powered up 15 minutes later and good operation for 2.5 hrs, bad cycling until 4:45 AM when on its own it snapped out of bad cycling. Until about an hour ago. It's 27 outside.
By the way...
I'm in a five story apartment style condo building. On the fifth floor. Next door is the end until 100 sq ft larger than mine. Same Mitsubishi on roof and wall units but three years old and working fine. Two doors down, done at the same time as next door same Mitsubishi system, exactly the same middle unit as mine is also working fine.
Go outside and listen to your outdoor unit. When the power dips, does it shut down completely or goes into defrost. You can tell it is defrost as there will be some refrigerant flow noises the fan will shut down but the compressor will ramp to max power for a bit. The outdoor coil will also get a bit hot.
If it is not defrosting, the issue is short cycling but need to know what is actually happening.
I can hear a sloshing, probably flow noises?, at the wall mounted units. Particularly toward the beginning of the vanes being horizontal.
The outside unit is on the roof, tough access.
On the indoor unit, you can take off the air filter so you can touch the coil while running. When it goes into defrost the coil will drop bellow room temperature, should be pretty cold.
I have an IR thermometer I use. The temperature drops. The colder it is out, the lower the temperature. Sometimes and room temperature, often lower.