Modernizing FCU: 120 V to 24 V and PSC to EC motors

[aisbell]

Few smart thermostats are available for our 35 year old 120 V fan coil unit (FCU) with no control board, 3 fan speed power relays, and two 3-speed PSC motors. Its blower wheels are vibrating and its motors are nearing their end-of-life, so it’s time to modernize it to 24 V with EC motors. The simpler approach would be to add a 24 V transformer and replace its 120 V DPDT power relays with 24 V equivalents. However, this would not reduce the rat’s nest of wiring, wire nuts, and ¼" spade connectors of the current system.

Alternatively, I could add a Temco Controls RB-XF-PB-110 FCU control board that includes a 24 V transformer, three 24 V fan speed relays, and two 24 V cool and heat valve actuator relays. This would simplify the wiring. However, it would introduce a component that could fail. For an investment of less than $70 including the cost of the board, international shipping, Trump’s tariff, and a tariff processing fee, I’m moving forward with this board.

I have attached a wiring diagram of my current FCU configuration:

120 V PSG Motors FCU Wiring.pdf (129.2 KB)

Please post any experience you have with this board.

[aisbell]

I need to keep our FCU operating every night, so I am slowly modernizing it in stages. The first stage involves replacing the 3 120 V fan speed power relays with the RB-XF-PB-110 FCU control board that supports either 120 V or 24 V valve actuators. I have completed this stage so that our thermostat is now running at 24 V while the PSC motors and cool valve actuator still run at 120 V.

One concern is that the SPST fan speed relays on the RB-XF-PB-110 FCU control board are designed for isolation rather than power. They are rated at 12 FLA @ 125 VAC with each fan motor rated at 6 FLA. Unlike the OEM DPDT power relays in which each motor gets its own set of fan speed relay contacts rated at 12 FLA, each of the RB-XF-PB-110 FCU control board fan speed relays must power both motors which together have a 12 FLA. I would not expect the High-speed relay to last anywhere near the 30 years that the heavy-duty OEM relays have lasted.

Fortunately, we always run the fans on their Low setting. I measured 4.6 A for the total current of the FCU set to Low speed, so the RB-XF-PB-110 control board Low-speed relay is experiencing less than half of its rated FLA while the Medium- and High-speed relays are inactive.

The photo below shows the RB-XF-PB-110 FCU control board mounted on the outside of the end of our FCU below a junction box containing a rat’s nest of wires tenuously connected by wire nuts. This junction box was originally mounted at the same location but on the opposite side of this wall where its access required removing the unwieldy sheet metal bottom of the FCU. It originally contained the 3 fan speed power relays and the wiring rat’s nest. It’s much more convenient to access on the outside of the FCU air plenum.

Flexible conduit containing the 120 V power wires is connected to the right side of the junction box while the flexible conduit containing the thermostat wiring is connected to the left side. They were originally connected to the side wall of the FCU behind where the junction box is mounted. Flexible conduit containing motor and valve actuator wiring that originally connected to the side of the junction box now connects to the side wall of the FCU behind the junction box. This allows motor and valve actuator wiring to pass through the FCU side wall into the junction box.

24 V thermostat wiring passes between the control board box and junction box at the top left of the control board box while 120 V power, motor, and valve actuator wiring passes between them at the top right.

I have attached a wiring diagram of this intermediary stage of my FCU modernization:

24 V PSC Motors FCU Test Wiring.pdf (184.3 KB)

I’m pretty happy with the way this first stage has turned out. I hope that moving to the next stage will be just as successful.

[aisbell]

I am awaiting the delivery of 2 Genteq Evergreen IM 6005 electronically commutated (EC) motors designed to replace PSC motors. The image below shows the connections between an Evergreen motor and a FCU control board. There are important differences from PSC motor wiring:

• The motor is powered by always-on 120 V power wires.
• 120 V signal wiring is powered only when the thermostat signals the motor to run at the speed selected by the thermostat. All motor speed relay outputs must be connected together so that when the thermostat signals the fan to run at any speed, the 120 V signal wiring will be powered. Only the HV2 signal wire should be connected so that the motor will run at ¼ hp. Because only signal wiring is connected to the motor speed relay outputs, the relays would not experience any significant power and would not need to be power relays.
• Thermostat fan speed wiring connects directly to the motors to signal the speed at which they should run. Because our thermostat supports only 3 fan speeds, a maximum of 3 of the 4 fan speed wires will be connected to the thermostat fan speed wiring.

[aisbell]

RB-XF-PB-110 FCU Control Board Concerns

• Even when a fan speed relay’s 24 V coil is providing a 24 V load, the voltage across the 24 V transformer’s output wires is 29.6 V which seems way too high. That’s almost 25% above spec. The 24 V transformer’s input voltage was 118 V, so a high input voltage does not explain the high output voltage.

The Evergreen motor documentation contains a warning (bolding is mine):

DO NOT connect the Low Voltage SIGNAL (speed) connections to the existing PSC motor connections on the HVAC system. Any voltage above 30VAC connected to these terminals will cause permanent damage to the motor.

29.6 V is very close to the 30 V limit above which the motor’s electronics would be damaged. Our residential voltage can vary above and below 120 V, so periods when our residential voltage exceeds 120 V could increase the 24 V transformer’s output to greater than 30 V. I do not want to damage the expensive Evergreen motors. Because of this, I am considering replacing my RB-XF-PB-110 FCU control board with a high-quality 24 V transformer plus 3 24 V isolation relays.

Is my FCU control board defective?

Is there an adjustment that could reduce the output voltage of its 24 V transformer?

• When powered, the Evergreen motor’s 120 V signal wire turns on the motor. This wire would be connected to the outputs of the fan speed relays. The voltage of the outputs of all 3 fan speed relays when these relays are open should be 0 V, but it’s 36 V! I am concerned that the Evergreen motors would interpret 36 V on its 120 V signal wire as an “on” signal so that the motors would always be running when the FCU control board is powered. This would be unacceptable.

Is my FCU control board defective?

Is there anything that I could do to eliminate their output voltage when the fan speed relays are open?

Do these problems make my RB-XF-PB-110 FCU control board unsuitable for use with my Evergreen motors?

[maurice]

From my quick read through all this, the main concern is the transformer sending out 28V and higher, this would be attributed to the fact there’s no load on it. Try adding a thermostat, a Tstat10E for example, it will pull around 1VA and should bring the transformer voltage down.

Fan Control:
I’ll assume you are considering a general purpose Tstat10 which can be programmed with your own logic. This way any relay and analog output can be programmed to operate based on your custom sequence of operations. Search around on this forum for examples, but its basically IF THIS THEN THAT type logic that you can learn and tweak as you like.

10V SIGNAL WIRING:
You mentioned analog output, 0-10V signals on your fan. The Tstat10 has two analog outputs that can work fine with adjustable speed drives. If your fan does in fact have this feature be extra carful that the GND of this control signal is in fact isolated from the 120VAC. The blue line in your drawings shows as the ‘COM’ line and is connected to the GND of the transformer board, so it may be isolated, be extra sure by fully probing it out with a multimeter with the devices powered while checking voltages and current, then non powered to measure resistances.

[aisbell]

I measured the output voltage of the 24 V transformer when the FCU control board was installed with the thermostat requesting cooling at the low fan speed, so the thermostat and the coils of the low fan speed and the cool valve relays were 24 V loads that did not decrease the 24 V transformer output voltage significantly. It remains too high.

Unlike our Peco wireless thermostat, the Tstat10E thermostat cannot be carried from room to room so that the temperature of a particular room can be controlled. When we sleep, we don’t care about the temperature in any room other than our bedroom, so we place our thermostat in our bedroom overnight. The fixed location of our thermostat, the location where a non-wireless thermostat would be installed and where the wireless receiver of our thermostat is installed, is in our kitchen where the oven, range, and clothes dryer in the adjoining laundry room warms it and where the exhaust of our heat pump water heater in the laundry room cools it. This is a terrible location for controlling the temperature in the remainder of our apartment, so we won’t be replacing our Peco thermostat with a Tstat10E thermostat whose features we don’t need.

Our Evergreen EC motors are 5-speed, not variable speed. Each speed is signaled by a 120 V signal in the HV harness combined plus a 24 V signal on the appropriate speed signal wire. Our cool valve is fully opened when 24 V powers the actuator and fully closed by a spring with no partial opening. Nothing in our FCU produces or responds to a 0-10 V signal.

Maurice, is a 24 V transformer output voltage of 29.6 V under load within spec for the RB-XF-PB-110 FCU control board? Seems dangerously high to me.

[maurice]

The voltage sounds high, in my experience we should be +/- 10% once a ‘load’ is applied. How much is a ‘load’ you ask, I would say 1VA will be enough. You could check the current going through the secondary in your test and see what you find. If the transformer output is too high for your setup then we can produce a custom transformer. Its a lot of effort for a single unit though, perhaps you can round up a transformer locally for now.

[aisbell]

I might have discovered another reason why IEC used DPDT fan speed relays in our FCU: to isolate one motor from the other. The Temco’s RB-XF-PB-110 FCU control board has SP relays such that the fan speed wires to both motors must share each fan speed relay output. I wanted to be able to keep our FCU operational each night during my modernization process, so I disconnected motor 2’s low fan speed wire from the low fan speed relay expecting that to result in motor 2 not running when the thermostat was requesting low fan speed. I was surprised when both motors continued running.

With a SP low speed fan relay being shared by the low fan speed wires of both motors, disconnecting the low fan speed wire of motor 2 does not interrupt power paths to ground through some of motor 2’s windings. Power entering through motor 1’s windings through its low fan speed wire is divided between continuing to ground through its low fan speed windings and through its medium and high fan speed windings, then through its medium and high fan speed wires to the outputs of the medium and high fan speed SP relays, then through motor 2’s medium and high speed fan wires, and through motor 2’s windings to ground. This would not be possible with DP fan speed relays. I don’t know the effect of not having isolated fan speed wires when all fan speed wires of both motors are connected to their SP relays. However, it’s probably not ideal. Power to each motor through one fan speed wire would have several paths to ground. This explains the unexpected voltage measurements on the output of the medium fan speed relay, 90 V, and the high fan speed relay, 62 V, when both motors are running at low fan speed. Because of this, I would not want to use Temco’s RB-XF-PB-110 FCU control board with more than 1 PSC motor.

Because I don’t understand how Evergreen IM EC motors operate, I do not want to risk damaging them or having them operate sub-optimally by not isolating each motor’s HV and LV control wires from those of the other motor. It’s not possible to isolated these motors using one RB-XF-PB-110 FCU control board. If you know why not isolating the control wires of 2 Evergreen IM EC motors would be fine, please explain.

[maurice]

I stared at this writeup for a while and ran it through my trusty AI tool but am still a bit scratching my head on this project. I think the root of it is you have figured out you cannot mix GNDs and neutrals between the various bits which makes sense. These multi speed fan may or may not have isolated low voltage sections.

The Tstat10 has a SINGLE common at terminal 17 on the relay outputs which may not tie in well with your sytem.

If you need isolated outputs you can use one of these relay boards which have fuly independent dry contacs for each output. The Tstat10 can energize the low side and your various multi speed fan controls can be wired independently.

If you are tired of messing with the multi speed fans you could consider a small variable speed drive. I see there’s many good options on ebay for under $100.

[aisbell]

I apologize for not writing more clearly. As I wrote at the beginning, I am modernizing our FCU in stages. The first stage is to replace the 120 V fan speed DP relays with Temco’s RB-XF-PB-110 FCU control board. That has worked except that the relays on this control board are SP so that the fan speed wires of both PSC motors must share the same relay terminals. This electrically connects the windings of both motors such that there are multiple pathways to ground through various motor windings. I don’t know whether this might cause problems. Isolating the motor windings might be one reason why the FCU originally used DP instead of SP relays. I don’t want to risk problems that might be caused by the motor windings not being isolated which would make Temco’s RB-XF-PB-110 FCU control board unsuitable for FCU’s with more than 1 blower motor like ours.

My next step would be to replace the PSC motors with EC motors. However, the HV and LV control wiring of one motor would not be isolated from that of the other motor because of the SP relays of Temco’s RB-XF-PB-110 FCU control board. This might make this control board unsuitable to control 2 EC motors.

I am awaiting support from the EC motor manufacturer to determine whether I can use Temco’s RB-XF-PB-110 FCU control board as is or whether I might need to install additional relays to isolate the motors from each other.

[maurice]

You are doing great at explaining everything,

The multi speed fans typically dont isolate the low and high voltage sides so this may be what has stalled your project.

I am standing by to help more.

[aisbell]

Our 3-speed PSC motors have only HV wires, so there are no LV wires to isolate from the HV wires. The isolation I was referring to was the HV fan speed wires of one motor from those of the other. When these HV fan speed wires share the same SP relay terminals on the RB-XF-PB-110 FCU control board, the windings of both motors are electrically connected such that there are multiple pathways to ground which doesn’t seem ideal. The original DP relays isolated the windings of each motor from those of the other, so I would not used this FCU control board with more than 1 blower motor.

Moving to the next step of replacing the PSC motors with EC motors is stalled because I don’t know whether the lack of electrical isolation between the 2 EC motors when connecting them to the SP relays on the RB-XF-PB-110 FCU control board would cause problems. I hope to learn the answer soon.

[aisbell]

I couldn’t wait any longer for answers to my questions about installing 2 Evergreen IM motors in a FCU, so I installed them but did not connect their LV fan speed signal wires which would result in the motors running at their lowest, most efficient speed. This is sufficient for this cooler season but might be insufficient on the warmest, most humid days, so I wanted to connect 2 of these fan speed signal wires to provide the 3 speeds supported by our thermostat:

Thermostat G1 (L) not connected to a fan speed wire;

and to implement the greatest fan speed range:

Thermostat G2 (M) connected to the ML fan speed wire;
Thermostat G3 (H) connected to the H fan speed wire;

or to implement the lowest fan speeds that we would normally need:

Thermostat G2 (M) connected to the L fan speed wire;
Thermostat G3 (H) connected to the ML fan speed wire;

When these motors are first powered, they perform an automatic rotation sensing procedure by rotating briefly and potentially repeatedly in each direction to determine the correct rotation direction based on the torque required to spin the blower wheels in each direction. With 2 motors sharing the same air plenum, I was concerned that the airflow caused by the blower wheel on one motor might affect the required torque of the other motor potentially resulting in automatic rotation sensing failing or the incorrect rotation detected. Rotation direction could probably have been set manually with a jumper, so this made me wonder whether something other than rotation sensing might be occurring. However, these motors are pre-programmed with no on-site programming possible.

Because of this, I decided to apply power to each motor individually to allow each to sense the correct rotation direction. This seems to have worked correctly.

I was concerned about connecting the LV fan speed signal wires because of the high output voltage of the RB-XF-PB-110 FCU control board’s 24 V transformer, ~29 V even with one 24 V fan speed relay and the 24 V cool valve actuator relay being powered. Maybe that load isn’t high enough to reduce the voltage, so I replaced the 120 V cool valve actuator motor with a 24 V version to put an additional 24 V load on the 24 V transformer. As Maurice suggested, this is sufficient to reduce the 24 V transformer’s output voltage to ~25 V.

I then felt confident enough to connect the motors’ LV fan speed signal wires to complete the installation. 3 120 V power wires, 6 thermostat wires, 2 cool valve actuator wires, 18 motor wires, and 11 FCU control board wires must be connected correctly in the wiring junction box. I’m not a fan of conventional wire nuts that could loosen, fall off, expose wiring to short circuits, allow connected wiring to disconnect, etc., so I used 3 2-wire and 17 3-wire Wago 221 Lever-Nuts instead. This resulted in a neater, more secure installation whose wiring connections are less likely to fail. I applied custom labels to each Lever-Nut to document its connection for future reference in addition to using as many color-code wires as possible as shown in the wiring diagram below:

24 V ECM Motors FCU Wiring.pdf (235.0 KB)

I will be testing this installation over the next couple of weeks to determine whether it is reliable enough to leave running for just over 3 months while we’re traveling internationally. Due to the introduction of a FCU control board and motors with integrated electronic controllers that are sensitive to power surges, this modernization will likely be less reliable than the 35-year-old initial system that had never failed. However, early indications are that the Evergreen motors use significantly less electrical energy than the PSC motors that they replaced when running at their lowest low speeds. However, I do not have a way to compare their air flow per watt, so some of the Evergreen motors’ energy savings might be due to their lower fan speed and the replacement of blower wheels whose rusty vanes might have had greater aerodynamic drag.

I’m happy to be finished with this modernization and with the results at this time. The modernized FCU is quieter with no large clacking mechanical relays, less air noise with the new blower wheels, and quieter motors that slowly ramp speed changes including when they’re powered on and off. I’m confident that the cost of running our FCU will be reduced.

[maurice]

That’s a lot to digest, I would not have made it that far.

I’d give that VFD I mentioned up above a good look for the next variable speed fan or pump retrofit project.

Have a look at the remote access setup when you’re on your long trips.

Accessing a remote site

First you will need to create an account at one of the dynamic dns services such as no-ip.com, pick a name for the remote site such as “BuildingSite101.dyndns.com”. The dynamic IP service will let you refer to “BuilsingSite101.dyndns.com” even though the ISP assigns dynamic IP addresses to your remote site. If you are lucky enough to have a fixed IP address you can use that as well. Next connect to the T3000 controller and fill in the details of your dynamic DNS account. The T3 controller will…