GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Minisplit troubleshooting

DarkNova | Posted in Mechanicals on


I built a well-insulated house this fall and after learning about minisplits on this site, we decided to heat our house with 3 minisplits. I can write more about this at a future date, but for now I’m having trouble with one of them and am trying to debug it to get as much information as I can before approaching the HVAC tech that installed it. We are in climate zone 7, so we have a lot of cold nights here (we’ve been below -22F several times this winter).

2 of our minisplits are Fujitsu 15RLS3H, which have been working just fine, no problem heating our house.

The minisplit we are having trouble with is a Mitsubishi MSZ-FH09NA, which is heating a smaller zone of our house which has a 5,300 BTU/hr design heat loss at -12F (our 99% design temp). The base pan heater was installed. The reason we went with the Mitsubishi here, rather than the Fujitsu, is that the Mitsubishi is rated for a capacity of about 6,800 BTU/hr even at these cold temperatures, which is sufficient for this zone, and as Dana has pointed out in this forum, the Mitsubishi can throttle down to a low-end of 1,600 BTU/hr, whereas the Fujitsu can only go down to 3,100 BTU/hr before cycling, so the Mitsubishi should be a lot better fit for this zone when the temperatures are more mild.

So, that’s the background, now here’s the problem…

I’ve been measuring things to confirm that the Mitsubishi is not functioning as it should. Today it is 34F outside here. I set the Mitsubishi to the highest setpoint, 88F. I set the fan to speed 3 and let it go for a while, then measured the temperature of the airflow coming out. It was blowing 85F air, and the incoming air was 70F. According to the engineering sheet for this minisplit, fan speed 3 is 225 CFM, and at my elevation BTU/hr is close to 1.0 * CFM * dT, so that would put the output at 3,375 BTU/hr. I changed to fan speed 2 which is rated for 167 CFM and got a temperature output of 92F, so that would put the output at 3,674 BTU/hr.

Someone please let me know if this is an inappropriate way to calculate the BTU/hr output, but as a point of reference I did this on my Fujitsu a couple weeks ago (it was 5F outside) and got:

– Fan speed 4 (547 CFM): 120F output (78F input) – dT: 42F — 22,974 BTU/hr output
– Fan speed 3 (459 CFM): 131F output (78F input) – dT: 53F — 24,327 BTU/hr output

At 5F the design manual for the Fujitsu’s show 19,900 BTU/hr at 5F outside, 75F inside, so it would seem that this experiment achieved a close result, not factoring in defrost cycles.

So, back to the Mitsubishi, the engineering sheet shows a maximum heat output of 12,000 BTU/hr, so we are far short of that. I measured with a power meter that the unit was drawing about 1150W during this time, which would be equivalent to 3,900 BTU/hr of resistive heat, so it would appear that the COP of the unit is less than 1 right now.

So it is clear that something is wrong, but what? The lineset is a little longer than the 25 feet precharged, so they were supposed to add about 2 oz of refrigerant. If they forgot to do that, would it affect things this drastically?

I measured the hot line of the lineset at about 177F while it was running.

If anyone has any insight into what needs to be investigated more, or any more tests I should run, before I approach the HVAC tech with my findings, I’d really appreciate it! Thanks.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Dana Dorsett | | #1

    Setting the blower speed to a higher speed rather than letting it modulate will result in a lower than normal exit air temperature. The cfm numbers are only approximate, and easily skewed by how much lint or other crap has been picked up in the filter &/or coil. But it sounds like the thing isn't really delivering.

    With a new installation you should to be chasing the contractor if it doesn't seem to be operating to spec. You have the data to back it up, even if the approximation is crude. The most likely cause is an improper refrigerant charge, possibly due to a refrigerant leak.

  2. DarkNova | | #2

    Thanks Dana. I will follow up with the contractor to have them investigate the refrigerant level. I'm somewhat surprised in all this that the minisplit hasn't detected any error condition. I know there's quite a few sensors and I would think it would be able to detect abnormal operating conditions by measuring the temperature drop across the coils and comparing it with "proper" values based on the interior/exterior temperatures, but perhaps they can't do that.

  3. simpsoap993 | | #3

    Nick - I would definitely have them check for leaks. I posted a thread on here not too long ago and was having similar issues with my multi-head Mitsubishi setup. The contractor ended up finding a leak after he pumped down the system to get an accurate measurement of how much refridgerant was in the system. I had been measuring the head output temp, and I was seeing around 85-88F, since the leak has been fixed, no more issues (crossing my fingers).


  4. DarkNova | | #4

    Thanks Andy. Was the leak on yours on one of the flares or somewhere else?

  5. DavidLeff | | #5

    I too had symptoms like you describe (although with a Fujitsu 15RLS2). I noticed the problem about 6 months after installation when we switched from cooling to heating. Eventually the Fujitsu began giving an error code. In my case there were two refrigerant leaks. One at the flare of the outdoor unit and one at the flare of the indoor unit.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |