Helping diagnosing serial communication error with Fujitsu mini-split

Sethamin

I'm a homeowner and I have a Fujitsu Halcyon mini split which suddenly stopped working.

I checked the error code on the condenser (model AOU24RLXFZ) and I see three sets of leds blinking once, which means "Serial communication error". One oddity here is that I only have 2 air handlers connected to it (though it supports three), but I'm seeing three lights.

I followed the troubleshooting steps in the service manual, so I next checked the voltage on the power to system, and I got 209VAC across terminals L1-L2 (which is normal).

Next I checked the voltage across terminals L1-L3 for the cable to the indoor units, and I got 59VAC, which is too low. It's supposed to be between 90VAC and 270VAC. So clearly there's an issue there.

The next step in the manual says:

Check the parts as follows.
- Outdoor unit fan motor (PARTS INFORMATION 5)
- Active filter module (PARTS INFORMATION 6)
- Filter PCB (Check the wire of CN34)

There's not much more information on exactly how to check those, though. Any suggestions on further troubleshooting steps? Fujitsu won't talk to me because I'm not a certified tech. I'm trying to figure out if it's worth fixing this system (e.g. maybe just a bad cable?) vs something that I'm not going to get fixed (e.g. a bad main board). I've dumped so much money into this system over the last few years that I've resolved not to spend much more on it, I'd rather replace it if it's a major repair.

109A_5

@Sethamin

First unless you know for sure your home's nominal Voltage should be in the 209 VAC ballpark which seems very low to me. Most modern USA residential and similar electrical services are 120/240. Around here 208 VAC is commonly found across two hot legs of a 3 phase service often found in small commercial situations.

The reason I focus on this 'FIRST' is if the Voltage is low to the appliance, the defect causing the low Voltage situation may also cause the appliance to malfunction as it tries to draw more current when trying to operate normally. I know the manual states 209 VAC is within the specifications, but how do you know it stays at 209 VAC during a fast instantaneous increase in the system's loads ? If you don't verify proper power you may wish you did hours later. In other words poor power quality can cause erratic or erroneous operation.

If you are quite convinced the Voltage to the unit is normal for your building's Electrical Service.

As you stated, Manual page 02-04;
Check the parts as follows.
- Outdoor unit fan motor (PARTS INFORMATION 5)
- Active filter module (PARTS INFORMATION 6)
- Filter PCB (Check the wire of CN34)

With "PARTS INFORMATION 5" and "PARTS INFORMATION 6"
These instructions are further along in the manual on how to briefly test these two parts.

SERVICE PARTS INFORMATION 5 Manual page 02-43
SERVICE PARTS INFORMATION 6 Manual page 02-44

In the manual I found here.
https://hvacdirect.com/media/hvac/pdf/AOU24RLXFZ-Service-Instruction.pdf

Also the third error LED, LED3 may always blink in your situation since that hardware is missing. It is odd that the "Outdoor unit to Indoor unit D" is not flagged also, however that may be normal in this situation.

HVACNUT

Hypothetically, you do resistance tests on the fan motor and active filter module. Turns out it's the motor. How do you plan on obtaining one?

mattmia2

Do other electrical things on your house seem normal? Are lights getting dim and bright or are they seeming normal? Are other things with motors sounding odd? Check the voltage at a few 120v receptacles on different circuits or on both legs of a 240v dryer or range receptacle. Sounds a lot like you have a bad neutral if you have a 120/240v single phase residential service.(I am assuming this is in the US. Is it?)

109A_5

Lets be clear here folks, a compromised neutral ONLY will not drop a 120/240 down to 209 VAC. A compromised neutral ONLY will cause the neutral in the house to swing with the load(s) so you will (for example) see L1 = 130 VAC to the neutral and L2 = 110 VAC to the neutral still adding up to 240 VAC. The utility's transformer typically has a 240 VAC secondary that has a center tap. The center tap becomes the neutral and is usually also grounded. So the house has 240 VAC but normally nothing is above 120 VAC to ground.

If L1 = 120 VAC to the neutral and L2 = 89 VAC to the neutral than L2 has a problem somewhere.
If L1 = 104.5 VAC to the neutral and L2 = 104.5 VAC to the neutral than your 240 VAC is too low.
If L1 = 120 VAC to the neutral and L2 = 120 VAC to the neutral and your total is 209 VAC you have two hot legs of a 3 phase service to your outdoor unit which for a typical residence may be very odd unless it is a larger building of some sort (condominiums) and not a typical single family type home.

Looks like the outdoor fan is available here https://www.amazon.com/Fujitsu-K9602843015-Outdoor-Condenser-Motor/dp/B08WLL1BYH

HOWEVER the manual recommends the main PCB be replaced also if you are going to replace the fan motor. The main PCB may be harder to find. (Kind of the shotgun troubleshooting method going on here)

Also I am not clear as to why a failed outdoor fan motor would cause 3 communications errors unless the indoor units need to know (and / or command) the outdoor fan.

mattmia2

it can't swing line to line down to 209, but it can swing one leg from hot-neutral up to 209 v. if that leg has much less load than the other leg to neutral which will be around 30v.

Once again there is a question I have asked that is ignored. This time if it is a 120/240v single phase service and if it is in the US.

109A_5

Sorry, I looked at the manual. I guess that is cheating, My bad.

Check Point 3 : Check the voltage of power supply
Check the voltage of power supply
>> Check if AC187V(AC208V-10%) - 253V(AC230V+10%) appears at outdoor unit terminal L1 - L2.

Sethamin

These are good points. I am in the US and I have single phase power. It's an old house but wiring is modern. Each hot leg measures 122V to ground, so I don't understand why I'm not getting 244V across both of them (there's no neutral on this circuit since it's a 240V circuit with two hot wires and a ground). I measured again today and now I get 211V across both, which still seems lower than it should be. Why would that be?

mattmia2

Is the 211v open circuit or under load?

Sethamin

Only thing I can figure is that I don't actually have single phase service.

Anyway, it's hard for me to see how that's the issue. I have a Mitsubishi condenser running off the same panel (though a different circuit) with the same voltage and it's fine. That's not definitive proof, but it's a data point.

Sethamin

mattmia2 said:

Is the 211v open circuit or under load?

Well the condenser is on but just giving me an error - do you consider that under load? I'd guess it's drawing a tiny bit of current, but not much.

mattmia2

If you have mostly industrial or farm customers they could have given you 2 legs of a 3 phase and a neutral but someone else who knows power systems better than I could tell you more about that.

Sethamin

mattmia2 said:

If you have mostly industrial or farm customers they could have given you 2 legs of a 3 phase and a neutral but someone else who knows power systems better than I could tell you more about that.

I'm in the middle of Brooklyn so probably that doesn't apply.

pecmsg

Run new wires overhead from indoor to outdoor. See if the codes still come up?

Sethamin

pecmsg said:

Run new wires overhead from indoor to outdoor. See if the codes still come up?

Yeah that is an idea. Before I do that, I was thinking I should first disconnect the wires from each indoor unit to isolate which set of wires are bad (if that's the case).

The counterargument is: why am I getting an error on all units if it's a bad wire to one of them?

109A_5

Good troubleshooting demands obvious oddities be understood before summarily dismissed, and corrected if needed. Ignoring them often makes troubleshooting impossible (if your not testing, your guessing), guessing gets expensive. Also different equipment can have different sensitivities to the same power quality issues. Multi-meters are very slow, there may be load changes going on in the outdoor unit that you are totally unaware of.

I would check the power (Voltage) at the main panel, main breaker (be extremely careful!!!). If it is good at the main breaker the circuit breaker and the wiring to the Fujitsu Halcyon outdoor unit would be highly suspect.

Also how many transformers on the utility pole? Only one for your home and maybe a few other homes or three in a cluster ?

Sethamin

Voltage at the main panel (200A service) is 208V across both hot wires. Voltage on the 100A breaker feeding the subpanel (which has breakers for the condensers) is 208V across both hot wires. Voltage from each hot wire to ground/neutral is 119V. There is no utility pole, I'm in the city where all power lines are buried; service enters my cellar from underground.

I think the logical conclusion is that I just happen to receive two hot wires in different phases for my main service. I don't have any other appliances which run on 240V (gas fired boiler, gas fired water heater, gas dryer) so it's just the condensers that would care. From what I've read they generally seem fine with 208V.

mattmia2

208 is fine for a motor as long as it is rated for it. It will just draw more current. For a resistive heating load it reduces the output significantly.

109A_5

OK, I think we have beaten the AC power thing to death. It appears Brooklyn does use 3 phase power (or a direct subset) directly to residential loads (Google search). And your measurements make sense.

Does the indoor unit(s) also give the error code for Serial Communication Error ?
2-1 ERROR DISPLAY
2-1-1 INDOOR UNIT
Manual page 02-01

Operation Timer Economy
(Green) (Orange) (Green)

1 times 1 times Continuous


I was hoping it was a power issue since that may be the least expensive repair alternative, and it truly needed to be verified first.

109A_5

Did you have the L1, L2, wires disconnected ? Could they have been swapped ?

https://hvac-talk.com/vbb/threads/1717261-Mini-Split-Heat-Pmmp-Serial-Communication-Error

Did you do the tests on pages 02-04 and 02-05 of the service manual in an early post?

Fujitsu High Wall Air Conditioner 12.000 BTUs Serial Communication over 220VAC line.
https://youtu.be/HDqMhQzB0HU
In this video I think it uses the line Voltage positive waveform alternations for communications in one direction and the negative waveform alternations for communications in the other direction.

Sethamin

I did a little bit more troubleshooting this morning. I disconnected the indoor unit wiring from the outdoor unit and fired it up. Still got serial communication error (probably not unexpected), but now I see fluctuating voltage from 110V-120V across L1-L3 (which is in proper range). If I connect either indoor unit, then the voltage on these terminals drops to 59V (though disconnected terminals still show 110V-120V).

Next I checked the outdoor unit fan motor control. No short there.

I checked the "Filter PCB" by finding the connector (CN34). Looks fine to me, but I pulled it out and reseated it just in case. Made no difference.

I tried to check the active filter PCB, but that board is kind of buried. It seems that you have to remove the main PCB to get access to it. Maybe when I have more time I'll try it. I think next I'm going to try running temporary replacement wiring between the outdoor and indoor units just to rule out bad cabling as the root cause.

109A_5 said:

Does the indoor unit(s) also give the error code for Serial Communication Error ?

There's nothing on the indoor units - it's as if they're not powered. Their respective wired thermostat also has no power (no voltage being supplied to them from the indoor unit). I do see the same voltage across L1 and L3 as I do on the outdoor unit, but the units just look off. No lights.

109A_5 said:

Did you have the L1, L2, wires disconnected ? Could they have been swapped ?

Yeah I saw that thread, but no, my wires are not swapped. I have 14/3 running to each head unit so they're all different colors and easy to match up.

pecmsg

Disconnecting one head WILL show the trouble code. Minimum is 3 heads!

pecmsg

109A_5 said:

Did you have the L1, L2, wires disconnected ? Could they have been swapped ?

Not going to make a difference with single phase.

Sethamin

pecmsg said:

Disconnecting one head WILL show the trouble code.

I have two heads connected to this condenser, and disconnecting one or the other didn't change the error codes. I still got "serial communication error" for all three possible heads. That probably suggests it's something wrong in the condenser itself.

pecmsg said:

Minimum is 3 heads!

I'm not sure what this means. The condenser supports a maximum of 3 heads, and I have it hooked up to 2.

pecmsg

Sethamin said:

pecmsg said:

Disconnecting one head WILL show the trouble code.

I have two heads connected to this condenser, and disconnecting one or the other didn't change the error codes. I still got "serial communication error" for all three possible heads. That probably suggests it's something wrong in the condenser itself.

pecmsg said:

Minimum is 3 heads!

I'm not sure what this means. The condenser supports a maximum of 3 heads, and I have it hooked up to 2.

and it's reading that and sending loss of comm! It needs all 3 head connected.

Sethamin

pecmsg said:

and it's reading that and sending loss of comm! It needs all 3 head connected.

That doesn't make any sense. It was working just fine with 2 heads. There's definitely no requirement that you need to run it with 3 heads.

pecmsg

What 2 heads are connected?

mattmia2

So let's back up, what is this l1 l2 l3? Is it the ac mains coming in to the unit or is it a supply out to the heads?

109A_5

OK, both indoor units appear dead ? So lets prove they are not. (dead indoor units may cause a communication error on the outdoor unit) Do any of the control functions work? With someone watching the indoor unit cycle its power (which may be the circuit breaker to the whole system) See if there is any sign of life. Also you could disconnect the L3 (with power off) and see it it creates a communication error at the indoor unit(s). Maybe the indoor unit(s) can hear the outdoor unit so they are happy but dormant, and the outdoor unit can't hear the indoor units so it is upset. This may provide direction.

109A_5

pecmsg said:

109A_5 said:

Did you have the L1, L2, wires disconnected ? Could they have been swapped ?

Not going to make a difference with single phase.

Apparently in some models communications are achieved by using say L1 (or L2, not sure) and L3. Reversing L1 and L2 in relationship with the other units may break the communications path, however the AC power will still be present and intact.

Another way of looking at it is the combination of L1 and L3 communicate in one direction and the combination of L2 and L3 communicate in the other direction. Hence the troubleshooting technique on pages 02-04 and 02-05 of the service manual in an early post. If L1 and L2 become swapped between indoor and outdoor units, no communications.

Anyway @Sethamin verified the wires were not swapped.

Using a technique like this may make a simple robust communications method and it only needs an additional external wire and will tolerate wiring errors without smoke and only minimal circuit board components to implement.

Sethamin

So it's been a while but I have a short update.

I went through all the troubleshooting again in the service manual (measuring both voltage and resistance) and everything looks fine on the fan motor and active filter module. I also ran a brand new set of wires (temporarily) from the condenser to one of the head units (and disconnected the other one) and I still got the same error. At this point I am 90% convinced it is the main PCB that needs to be replaced.

I haven't found a replacement one, but I did find one for an AOU18RLXFZ (same condenser model but 18k BTU instead of 24k BTU) here:
https://surpluscityliquidators.com/products/heater-control-board-aou18rlxfz.html?208346
Does anyone know if they use the same main PCB?

In the service manual for the respective condensers it shows a different part number for each main PCBs. BUT if I zoom in to the top left of that board listed above it has the exact same set of numbers as on my unit (picture is from my unit):

Might they be the same?

109A_5

Hello @Sethamin, It is certainly possible they are exactly the same. Although the one in the linked vendor's picture has an additional sticker on the board. The different part number in the service manual may just be for accounting purposes.

However they may be different. The board for the larger BTU unit may have more robust components for the motor control. So the higher BTU board may replace the smaller BTU board but not the other way around. The software and firmware may also be different. The DIP switch settings may be different. Only the design engineers may know for sure. And if they did not provide a cross reference it is a guess. The numbers silk screened on the circuit board may be for the board only and not what is populated on it.

Sometimes programmable micro-controllers have a revision sticker on the chip.

From what I can see of your board if the numbers are in the same corner the board foil is different (mounting screw hole too) and yours has some corrosion damage. The foil is mildly contaminated under the solder mask.

JUGHNE

Had a Fuji double unit throw the comm code.

The mother board had a blown fuse.....book says to replace PCB and fan motor. (no matter that it tests seemly OK.

Why the fan motor? The replacement fan motor is a new part number. Perhaps necessary for the new design? of PCB.....only about another $300 for the customer.

You cannot see the fuses without removing the board. This means having to redo the heat sink.
Plus the 15 or so plug connections have to accessed from the bottom of the board with minimal lead length.

The LED indicators are also on the bottom, it is a joke to try to see them in sunlight...actually worthless for TS.

If you have to change the main PCB, be sure to take pictures and count the connections on the bottom.
There may be some ports that are unused.

The Fuji change out procedures of routing cables seem to be written by a fourth world person who also writes up paddle fan instructions. They are comical in nature.
I would have thought, that by now, the Japanese could have hired a English speaking person write these procedures for the western user.

It is always a relief to find a smoked board....then you know it is toast.
It would be convenient if PCB's had a self destruct device...think Mission Impossible tapes....that would tell you that "I am toast".
If the compressor ohms out OK then I might proceed.
In this case I just changed all three boards, the reason is that to just have another service call justifies the expense....in my mind.

This unit was installed in 2006, if a double head unit was available it would have been changed out.
Mitsi double heads are not available to me until Nov.

This is a nursing home and the employees need the AC.
The residents have sweaters and lap blankets on them....most never notice that the AC has failed for their room.

This repair cost the home almost $1500......it should have been more but I am married to these things.

I can see the future of heat pump hype now happening. They are great until they aren't.

These are not the sole heat source for these nursing home rooms, there is a well sized electric heater in each room. I would not have installed these in a 40 bed home without any back up heat.

GGross

"They are great until they aren't"

Couldn't you say the same thing about every piece of equipment? It certainly would not be the first time someone had a $1500 repair bill

ratio

When some things aren't great, they're a lot less not great than other things.

Sethamin

So another thought: maybe the condenser is fine and BOTH air handlers are bad? That might explain why the voltage on the leads is fine when disconnected, but then drops out of range when connected. For that to be the case, though, we'd need some event that fried both units simultaneously, which would have to be a power surge. So maybe I should go look at the fuses on the indoor units?

HVACNUT

Is this system still under warranty?
Regardless, hooking up with Fujitsu tech support or using the Mobile Technician will get you the answer.

Sethamin

HVACNUT said:

Is this system still under warranty?
Regardless, hooking up with Fujitsu tech support or using the Mobile Technician will get you the answer.

Sadly no, it is not under warranty.

I've contacted Fujitsu, and while they've been helpful in the past, they won't help me this time because I'm not an HVAC tech (I have my EPA 608, but I don't think that's gonna cut it). I could call a local tech, of course, but I'm in a very high cost of living area and it'll probably cost me about $1000 just to diagnose the issue.

The mobile technician app looked promising, but I installed it and it doesn't support either my indoor or outdoor units.

pecmsg

Sethamin said:

Sadly no, it is not under warranty.

I could call a local tech, of course, but I'm in a very high cost of living area and it'll probably cost me about $1000 just to diagnose the issue.

The mobile technician app looked promising, but I installed it and it doesn't support either my indoor or outdoor units.

W O W

JUGHNE

Do you see the fuse(s) location on the PCB?
Can you post a picture of the entire board?

The 3 boards and fan motor cost less than that $1000 diagnostic svc call would.

Sethamin

Aha! I think I found the issue!

So there are no fuses on the indoor units. I went to check the fuses on the condenser instead. The fuse on the first hot leg looked fine:

But the fuse for the second hot leg is clearly blown:

Multimeter confirms this: the first fuse has no resistance and the second fuse has infinite resistance.

So this is a 10A 250V fuse:

Will any T4 10A 250V fuse do the trick here, or do I need something special? As in, can I just use something like this from HD:
https://www.homedepot.com/p/Cooper-Bussmann-AGC-Series-10-Amp-Silver-Fast-Act-Electronic-Fuses-5-Pack-BP-AGC-10/100200559
Or do I need to special order a Fujitsu specific fuse?