Over-voltage and Mitsubishi Heat Pumps

[kwired]

I can see them doing that just to honor a manufacturer warranty, maybe? But that is exactly what I was told by the HVAC tech that’s what his recommended fix was. The homeowner is getting in touch with the POCO, maybe their pots need to be retapped in the town before I look into a buck/boost for the heat pump or their whole house. We’ll see what’s budgeted and who’s paying for what fix before I go back.

I’ll keep checking in for the continued and useful help from everyone, but I’m on hold until I get more answers from the homeowner.


Sent from my iPhone using Tapatalk

contacting POCO is the best thing. if entire area has too high of voltage could be malfunction of regulating equipment, though most the time it tends to go even higher than what OP has which is still within acceptable range for delivery.

 

[JoesBagelandDeli]

Any update on this? I'm interested to know how this was resolved. We're having a similar issue and i'm curious if the transformer internal to the unit can be swapped, or if it perhaps has multiple taps that can be adjusted.
Thanks!

 

[WattsGood]

Any update on this? I'm interested to know how this was resolved. We're having a similar issue and i'm curious if the transformer internal to the unit can be swapped, or if it perhaps has multiple taps that can be adjusted.
Thanks!

Nothing heard back yet other than POCO was scheduled to check their transformer supplying them and the neighbors. I'll send a text out and see what's up.


Sent from my iPhone using Tapatalk

 

[Gasquet]

Hey guys, joining this conversation a bit late, but having some similar issues with my Mitsubishi system. All the low voltage boards have been fried twice now. I'm on my third set but am keeping the system off until I know I have solved the power quality issues.

Measuring 249V from the power company, but additionally have a 9V phase imbalance (129V on A and 120V on B). Mitsubishi states that a phase voltage imbalance should be <3%. You guys have not been mentioning above if you see a phase imbalance.

My system works fine until there is a power outage, then all the low voltage boards are fried. No bad fuses, no breakers trip, and no surge protectors are tripped, which is odd. I have seen voltages as high as 252V transiently. I have a "Ting" circuit monitor on the higher of the two legs that consistently reads 128V average with spikes above 132V.

I recently jackhammered three deep holes in solid rock to add three grounding plates associated with two structures. The neutrals had multiple places where they were tied to ground. My electrician reduced that to only one place on the site where that is supposed to happen. No improvement to the phase imbalance. It does however go away when there is no load (home off, guest house off, PV array off). It returns as I add loads from individual circuits, with no obviously failing circuit. It just scales with load. Electrician identified at least four splices in the drop between my transformer and the house.

Austin Energy claims there is no issue at their end and that 250V is within power specifications. They are likely testing without load. I am pushing back and trying to get them to change taps at the transformer to reduce my voltage down to 240V. I think I am the only house on this transformer.

Did the buck/boost transformer help solve your problem? I'm thinking of installing that or an isolation break transformer and adding a grounding rod dedicated to the heat pump alone to get it separated from the power quality problems that otherwise are not disturbing anything.

The sensitivity to phase imbalance seems to be a design weakness in addition to the 253V maximum voltage specification. I'm also considering converting the unit to 120V operation to work around the weaknesses. Thoughts?

 

[LarryFine]

It sounds like there is a problem with the neutral conductor between the utility and the house.

 

[ptonsparky]

You don't have a phase voltage imbalance. You have 249 volts, single phase. No other phase to reference. You do appear to have a neutral issue as Larry suggested. Get someone in to fix that.

 

[winnie]

Hi @Gasquet

Please keep in mind that the rules of this site explicitly prohibit providing 'DIY help'. With that said, you sound like me: a technical person who has useful knowledge to share. If you add to the discussion of physics here, I'm sure you will be welcomed.

The information about the line voltage imbalance is interesting, but I think it is a red herring. The unit is likely rated for 208 or 240V, and is supplied by two 'hot' wires of a single phase service. It doesn't have a neutral voltage reference, and so should not care about leg imbalance on residential split phase system.

One might imagine some sort of communications wire which is supposed to be near ground potential somehow being sensitive to the leg-leg balance, but such a system supplied with 208V (2 legs of a 3 phase system treated as single phase) would have its 'neutral' at 60V relative to ground.

Adding more ground rods will not fix a bad neutral issue.

-Jonathan

 

[winnie]

Also: L-N imbalance is a symptom of high neutral impedance combined with unbalanced loading.

The high neutral impedance could be indication of a failure such as a poor connection, or it might simply be a long service wire.

 

[ptonsparky]

Also: L-N imbalance is a symptom of high neutral impedance combined with unbalanced loading.

The high neutral impedance could be indication of a failure such as a poor connection, or it might simply be a long service wire.

I've found that to an issue. Once in 40 plus years.

 

[Crash117]

Hey guys, joining this conversation a bit late, but having some similar issues with my Mitsubishi system. All the low voltage boards have been fried twice now. I'm on my third set but am keeping the system off until I know I have solved the power quality issues.

Measuring 249V from the power company, but additionally have a 9V phase imbalance (129V on A and 120V on B). Mitsubishi states that a phase voltage imbalance should be <3%. You guys have not been mentioning above if you see a phase imbalance.

My system works fine until there is a power outage, then all the low voltage boards are fried. No bad fuses, no breakers trip, and no surge protectors are tripped, which is odd. I have seen voltages as high as 252V transiently. I have a "Ting" circuit monitor on the higher of the two legs that consistently reads 128V average with spikes above 132V.

I recently jackhammered three deep holes in solid rock to add three grounding plates associated with two structures. The neutrals had multiple places where they were tied to ground. My electrician reduced that to only one place on the site where that is supposed to happen. No improvement to the phase imbalance. It does however go away when there is no load (home off, guest house off, PV array off). It returns as I add loads from individual circuits, with no obviously failing circuit. It just scales with load. Electrician identified at least four splices in the drop between my transformer and the house.

Austin Energy claims there is no issue at their end and that 250V is within power specifications. They are likely testing without load. I am pushing back and trying to get them to change taps at the transformer to reduce my voltage down to 240V. I think I am the only house on this transformer.

Did the buck/boost transformer help solve your problem? I'm thinking of installing that or an isolation break transformer and adding a grounding rod dedicated to the heat pump alone to get it separated from the power quality problems that otherwise are not disturbing anything.

The sensitivity to phase imbalance seems to be a design weakness in addition to the 253V maximum voltage specification. I'm also considering converting the unit to 120V operation to work around the weaknesses. Thoughts?

Get the power company to test with Beast of burden. Puts extreme single phase loads on to test the neutral. If it’s higher on one phase than the other the test will show it.

 

[Gasquet]

Get the power company to test with Beast of burden. Puts extreme single phase loads on to test the neutral. If it’s higher on one phase than the other the test will show it.

Austin Energy is sending a tech out next week. I'm not dumb enough to DIY this. I just want to be informed to talk intelligently with the tech and not get steam rolled as in the past. I have been complaining about a problem for years and they do nothing about it. Now that it is blowing up my new heat pump (twice), and the third one is currently offline, I am more than a little motivated to get them to take corrective action.

I'm assuming that the "Beast of Burden" mentioned is some kind of high power load standard that gets plugged in instead of the meter? Is "Beast of Burden" an industry term the tech will recognize? What current do you consider sufficient current load and what voltage imbalance tolerance should they be working to be compliant to? 100A would be 1/3 of cumulative service adding the two main breakers on site. 10A would seem way too small. I am going to ask the tech to show me the load he uses so I can call BS if it is too wimpy.

I can't visualize the Physics for how a single neutral with a resistive open circuit with grounding could result in a voltage difference. All transformers have a tap for grounding to reference the AC to ground and indeed without load output voltages match. Why would a resistance on a shared neutral cause an offset? Seems like imbalance would only happen if one of the hots has a resistive splice if there is only one neutral. Are there two neutrals somewhere such that the resistances could be mismatched if one neutral were higher resistance than the other? Even if there is, both are tied to ground both at the transformer and at the service panel, effectively making them behave as one. Something has to break the symmetry for a voltage offset to show up. Help me understand this failure mode please.

Sorry if I am abusing the terminology guys. I understand it is single phase service. Referring to each hot conductor leg as a "phase" for lack of a better term in my brain.

 

[GoldDigger]

Austin Energy is sending a tech out next week. I'm not dumb enough to DIY this. I just want to be informed to talk intelligently with the tech and not get steam rolled as in the past. I have been complaining about a problem for years and they do nothing about it. Now that it is blowing up my new heat pump (twice), and the third one is currently offline, I am more than a little motivated to get them to take corrective action.

I'm assuming that the "Beast of Burden" mentioned is some kind of high power load standard that gets plugged in instead of the meter? Is "Beast of Burden" an industry term the tech will recognize? What current do you consider sufficient current load and what voltage imbalance tolerance should they be working to be compliant to? 100A would be 1/3 of cumulative service adding the two main breakers on site. 10A would seem way too small. I am going to ask the tech to show me the load he uses so I can call BS if it is too wimpy.

I can't visualize the Physics for how a single neutral with a resistive open circuit with grounding could result in a voltage difference. All transformers have a tap for grounding to reference the AC to ground and indeed without load output voltages match. Why would a resistance on a shared neutral cause an offset? Seems like imbalance would only happen if one of the hots has a resistive splice if there is only one neutral. Are there two neutrals somewhere such that the resistances could be mismatched if one neutral were higher resistance than the other? Even if there is, both are tied to ground both at the transformer and at the service panel, effectively making them behave as one. Something has to break the symmetry for a voltage offset to show up. Help me understand this failure mode please.

Sorry if I am abusing the terminology guys. I understand it is single phase service. Referring to each hot conductor leg as a "phase" for lack of a better term in my brain.

If you have only 240V, line to line loads, the neutral is not important. But if there is a small extra resistance in one of the legs, the voltage drop on that line conductor will be greater than on the other, which will result in a difference in Line to Neutral voltage. Which is not the result of neutral offset.

If instead you have one large single phase load, and an extra (small) resistance in the neutral, there will be a voltage drop in the neutral which will move its voltage closer to that of the loaded line. But at the same time, this will result in an increase in the Line to Neutral voltage on the unloaded leg. That increase in voltage on the unloaded side is the definitive symptom of a compromised neutral, anywhere between the loads and the secondary of the utility transformer. The asymmetry is in the unbalanced load.

Tying the neutral to ground at one or more points will not greatly affect the observed results, since the resistance of each earth electrode is high enough that little current will flow through it, unless the metallic neutral is open or nearly open.

 

[ptonsparky]

Austin Energy is sending a tech out next week. I'm not dumb enough to DIY this. I just want to be informed to talk intelligently with the tech and not get steam rolled as in the past. I have been complaining about a problem for years and they do nothing about it. Now that it is blowing up my new heat pump (twice), and the third one is currently offline, I am more than a little motivated to get them to take corrective action.

I'm assuming that the "Beast of Burden" mentioned is some kind of high power load standard that gets plugged in instead of the meter? Is "Beast of Burden" an industry term the tech will recognize? What current do you consider sufficient current load and what voltage imbalance tolerance should they be working to be compliant to? 100A would be 1/3 of cumulative service adding the two main breakers on site. 10A would seem way too small. I am going to ask the tech to show me the load he uses so I can call BS if it is too wimpy.

I can't visualize the Physics for how a single neutral with a resistive open circuit with grounding could result in a voltage difference. All transformers have a tap for grounding to reference the AC to ground and indeed without load output voltages match. Why would a resistance on a shared neutral cause an offset? Seems like imbalance would only happen if one of the hots has a resistive splice if there is only one neutral. Are there two neutrals somewhere such that the resistances could be mismatched if one neutral were higher resistance than the other? Even if there is, both are tied to ground both at the transformer and at the service panel, effectively making them behave as one. Something has to break the symmetry for a voltage offset to show up. Help me understand this failure mode please.

Sorry if I am abusing the terminology guys. I understand it is single phase service. Referring to each hot conductor leg as a "phase" for lack of a better term in my brain.

You have a lot of good questions and some of the guys here can explain it all, but it's time to get your electrician involved. Get a new one if you've already had one there that didn't help.

There is a good video explaining it. Someone will post it before long.

 

[winnie]

I can't visualize the Physics for how a single neutral with a resistive open circuit with grounding could result in a voltage difference. All transformers have a tap for grounding to reference the AC to ground and indeed without load output voltages match. Why would a resistance on a shared neutral cause an offset? Seems like imbalance would only happen if one of the hots has a resistive splice if there is only one neutral. Are there two neutrals somewhere such that the resistances could be mismatched if one neutral were higher resistance than the other? Even if there is, both are tied to ground both at the transformer and at the service panel, effectively making them behave as one. Something has to break the symmetry for a voltage offset to show up. Help me understand this failure mode please.

You are using the approximation that 'ground' is a perfect zero impedance reference. This may be a good approximation in some fields of work, but it is not even close when discussing low voltage residential wiring with soil as the reference.

A good residential ground rod might have a 'resistance to remote earth' of 25 ohm's, if you are lucky. The service conductors to your home should have a resistance below 0.1 ohm (ish). The parallel path through the soil doesn't significantly change the total neutral circuit impedance.

As to why neutral impedance causes unbalanced voltage, the neutral resistance alone is not sufficient. You must also have neutral current. You must have some amount of unbalanced loading on the split phase service so that some current flows on the neutral.

Draw out the three wire circuit from transformer to home, including the resistance of the service conductors. Use say 0.1 ohms for each service condictor. Ignore grounding. Add a couple of resistors to represent your home loads, say 2 ohms for one leg and 3 for the other. Assume the transformer is a perfect 120/240V source, and consider the transformer neutral as your 0V reference. Calculate the voltage at every node. Now change the neutral resistance and see how the voltage changes.

Finally, add in grounding, but model the ground as a 0 impedance current source/sink, but every connection to ground is a 25 ohm resistance.

Jonathan

 

[Gasquet]

If you have only 240V, line to line loads, the neutral is not important. But if there is a small extra resistance in one of the legs, the voltage drop on that line conductor will be greater than on the other, which will result in a difference in Line to Neutral voltage. Which is not the result of neutral offset.

If instead you have one large single phase load, and an extra (small) resistance in the neutral, there will be a voltage drop in the neutral which will move its voltage closer to that of the loaded line. But at the same time, this will result in an increase in the Line to Neutral voltage on the unloaded leg. That increase in voltage on the unloaded side is the definitive symptom of a compromised neutral, anywhere between the loads and the secondary of the utility transformer. The asymmetry is in the unbalanced load.

Tying the neutral to ground at one or more points will not greatly affect the observed results, since the resistance of each earth electrode is high enough that little current will flow through it, unless the metallic neutral is open or nearly open.

Thanks for that explanation. That makes a lot of sense. Behavior of the voltages with load suggests the second case. If I remove load, both legs move to 125V approximately.

 

[Gasquet]

Draw out the three wire circuit from transformer to home, including the resistance of the service conductors. Use say 0.1 ohms for each service condictor. Ignore grounding. Add a couple of resistors to represent your home loads, say 2 ohms for one leg and 3 for the other. Assume the transformer is a perfect 120/240V source, and consider the transformer neutral as your 0V reference. Calculate the voltage at every node. Now change the neutral resistance and see how the voltage changes.

Ah, yes thanks. It becomes a voltage divider modulated by the impedance to ground. That also explains another comment above about imbalanced loads. I don't think I have enough 120V loads to cause a meaningful imbalance, but I can go look at that today just to make sure that all the refrigerators and the wine fridge are on different legs at least.

 

[WattsGood]

Hey guys, joining this conversation a bit late, but having some similar issues with my Mitsubishi system. All the low voltage boards have been fried twice now. I'm on my third set but am keeping the system off until I know I have solved the power quality issues.

Measuring 249V from the power company, but additionally have a 9V phase imbalance (129V on A and 120V on B). Mitsubishi states that a phase voltage imbalance should be
My system works fine until there is a power outage, then all the low voltage boards are fried. No bad fuses, no breakers trip, and no surge protectors are tripped, which is odd. I have seen voltages as high as 252V transiently. I have a "Ting" circuit monitor on the higher of the two legs that consistently reads 128V average with spikes above 132V.

I recently jackhammered three deep holes in solid rock to add three grounding plates associated with two structures. The neutrals had multiple places where they were tied to ground. My electrician reduced that to only one place on the site where that is supposed to happen. No improvement to the phase imbalance. It does however go away when there is no load (home off, guest house off, PV array off). It returns as I add loads from individual circuits, with no obviously failing circuit. It just scales with load. Electrician identified at least four splices in the drop between my transformer and the house.

Austin Energy claims there is no issue at their end and that 250V is within power specifications. They are likely testing without load. I am pushing back and trying to get them to change taps at the transformer to reduce my voltage down to 240V. I think I am the only house on this transformer.

Did the buck/boost transformer help solve your problem? I'm thinking of installing that or an isolation break transformer and adding a grounding rod dedicated to the heat pump alone to get it separated from the power quality problems that otherwise are not disturbing anything.

The sensitivity to phase imbalance seems to be a design weakness in addition to the 253V maximum voltage specification. I'm also considering converting the unit to 120V operation to work around the weaknesses. Thoughts?

Sorry to hear you’ve got the same issue. And this is an ongoing thing around the country? I’m assuming you are in Austin, TX?

Last I hear from my friend is that POCO came out and did in fact lower the taps on the transformer. I have not been back yet to see what the readings currently are, but no boards have been fried since and the unit is currently running fine.


Sent from my iPhone using Tapatalk

 

[GoldDigger]

Sorry to hear you’ve got the same issue. And this is an ongoing thing around the country? I’m assuming you are in Austin, TX?

Last I hear from my friend is that POCO came out and did in fact lower the taps on the transformer. I have not been back yet to see what the readings currently are, but no boards have been fried since and the unit is currently running fine.


Sent from my iPhone using Tapatalk

Glad to hear that, but it does seem to suggest really marginal design by Mitsubishi! With a zero or even negative safety margin.

 

[Gasquet]

Sorry to hear you’ve got the same issue. And this is an ongoing thing around the country? I’m assuming you are in Austin, TX?

Last I hear from my friend is that POCO came out and did in fact lower the taps on the transformer. I have not been back yet to see what the readings currently are, but no boards have been fried since and the unit is currently running fine.


Sent from my iPhone using Tapatalk

Yes in Austin. "Beast of Burden" test performed today and Austin Energy says no issue found. Sigh... They aren't going to change taps and reduce my voltage, even though I am the only house on the transformer, and I specifically asked them to lower the voltage. I think I will need to install a buck boost transformer, but I still have an imbalance in the voltages to find and correct. On the higher voltage leg, continuous monitoring shows voltage up to 138V at my kitchen.

Glad you got your issues fixed. That is good to know.

 

[retirede]

Yes in Austin. "Beast of Burden" test performed today and Austin Energy says no issue found. Sigh... They aren't going to change taps and reduce my voltage, even though I am the only house on the transformer, and I specifically asked them to lower the voltage. I think I will need to install a buck boost transformer, but I still have an imbalance in the voltages to find and correct. On the higher voltage leg, continuous monitoring shows voltage up to 138V at my kitchen.

Glad you got your issues fixed. That is good to know.

You have an issue with your neutral somewhere downstream of where the utility conducted their test.