Mitsubishi ducted AC's thermostat measures attic temp, not room temp. Options?

[Earl Focht]

Hello, first time. I hope I interpreted the rules properly and this is appropriate here.

A contractor installed a Mitsubishi Mr. Slim SEZ-KD12NA4.TH (http://ecomfort.com/products/mitsubishi-suzka12nath--sezkd12na4th-single-zone-mini-split-heat-pump-system-concealed-duct-wireless-remote--11500-btu/26915) ducted AC/heat-pump system to cool my home office. (I only plan to use the AC — the room has hydronic radiant-floor heat controlled separately.)

The unit's in the uninsulated attic, about 30 feet of ductwork away. The return and both output vents are all in the 18'x15' office.

They installed the (hilariously ugly and unintuitive) PAR-21MAA wired remote to control it.

It cools the room very well, and I can't believe how quiet it is, but it appears that the thermostat is measuring the temperature in the (hot) attic, not the room. This is obviously pretty bad, since it's never responding to the actual room temperature and uncomfortably freezes the room while waiting for the hot attic to (never) get cool.

I've attempted some basic research, but I'm really not qualified in this field. It looks like the temperature sensor is actually in the return-air intake on the unit, not in the PAR-21MAA "thermostat" on the wall.

When they installed it (photos attached), it looks like they cut a 1" gap into the return duct for filter access, and then they hacked together a removable filter by cutting a filter from who-knows-where and screwing two wooden handles onto it. (Feel free to tell me how horrible that is, and if I should have them do it "right", whatever that would entail.)

My guess is that the temperature sensor is pretty close to this open-air gap, and return air is mixing with hot attic air before the temperature is measured, which is why it's measuring so much hotter than the room's actual temperature.

I'm having the installer come back here this week to take a look, but I'm questioning his skill level a bit at this point, and I'd love your input on what my options are and maybe what I should ask him to do. From the product literature, it looks like this unit is compatible with a "Remote Temperature Sensor (M21-JKO-307)" —*maybe that can be installed in the room and take over from the air-intake thermometer?

Alternately, and I know this is a longshot, is there any reasonable way I can just control this unit with the regular thermostat that's two inches away? It looks like it uses custom DC control wiring and is absolutely nothing like standard AC thermostat wiring, but I'd love to get this controlled by the Nest, abandon this PAR-21MAA, and completely sidestep the duct-thermometer issue.

Thanks for your time and expertise.

[mark beiser]

In the installation instructions for the thermostat there is a chart for the "Unit function selection" where you can change the "Indoor temperature detecting" to "remote controller's internal sensor.
Most of these type systems typically default to using the units return air temperature sensor and continuous airflow.

That filter setup is really sad...

I'm curious what kind of static pressure the unit has operating with that duct system. It is a "high static pressure" unit for use with ductwork, but maximum static pressure they are designed to work with is a fraction of what a conventional system is able to work with.

[Earl Focht]

In the installation instructions for the thermostat there is a chart for the "Unit function selection" where you can change the "Indoor temperature detecting" to "remote controller's internal sensor.
Most of these type systems typically default to using the units return air temperature sensor and continuous airflow.


Thanks, Mark! That did it, I think. (That process was humorously difficult to understand, even for this C programmer.)

Now the problem is that the remote controller's internal sensor lags behind the room temperature a bit, often being inaccurate by 3-5° for over an hour before catching up. I might just have the installer try replacing the remote controller.


I'm curious what kind of static pressure the unit has operating with that duct system. It is a "high static pressure" unit for use with ductwork, but maximum static pressure they are designed to work with is a fraction of what a conventional system is able to work with.

Any way I can measure that for you?

Obviously, I'm unfamiliar with most of these terms, but if that refers to the amount of airflow from the output ducts, I can tell you that it's definitely nowhere near as much airflow as any other air conditioner I've ever seen. You can barely feel the air moving with your hand a foot below the ceiling-mounted ducts. But it does cool the 18'x15' room (with two people and two always-on computers) very quickly and evenly, without noticeably blowing on anyone, which is actually quite nice.

[mark beiser]

Thanks, Mark! That did it, I think. (That process was humorously difficult to understand, even for this C programmer.)

Now the problem is that the remote controller's internal sensor lags behind the room temperature a bit, often being inaccurate by 3-5° for over an hour before catching up. I might just have the installer try replacing the remote controller.

Yeah, pretty much all of the Asian manufacturers have non intuitive setup procedures, and badly translated instructions.

Make sure the wiring hole in the wall behind the controller is plugged.



Any way I can measure that for you?

Not really, a specialized test instrument is required.


Obviously, I'm unfamiliar with most of these terms, but if that refers to the amount of airflow from the output ducts, I can tell you that it's definitely nowhere near as much airflow as any other air conditioner I've ever seen. You can barely feel the air moving with your hand a foot below the ceiling-mounted ducts. But it does cool the 18'x15' room (with two people and two always-on computers) very quickly and evenly, without noticeably blowing on anyone, which is actually quite nice.

The units typically will move less air for the given cooling capacity than a conventional system, even when the fan is on high.
My concern when looking at the pictures is that, even though it is a "ducted" air handler, the Asian manufacturers idea of "ducted" differs from conventional equipment.
The typical application for those units is to have them concealed in a dropped ceiling area with relatively open return and very short supply ducts.
That specific unit has setting that can be made that will allow longer ducts, but still nowhere near like a conventional system.
Usually it give enough airflow capacity to either have moderately long supply ducts OR moderately long return ducts, but not both without using larger ducts than would normally be used.

At any rate, the factory default setting is for a very low resistance duct setup, and has to be changed to the higher setting by the installer if the unit is going to be ducted in the fashion that yours is.
From what you describe about the airflow, I'm guessing the setting is still on its factory default.

[jpsmith1cm]

Steve B

This is the Ask Our Pro's forum. In order to post a response here, you must have verified qualifications and have been approved by the AOP Committee. You may ask a question by starting a new thread.

You can find the rules for posting and qualifications here (http://hvac-talk.com/vbb/showthread.php?t=957002).

Additional infractions may result in loss of posting privileges.

Your post has been deleted.