240 volt , 3 phase , need neutral

[Smart $]

But since the existing design uses one of the three phase conductors to power their controls, installing the CPT would mean modifying the design. Personally, that's likely what I would've done. But I can also understand the risk of voiding the warranty.

Not if the CPT is external to the warranteed equipment.

BTW, if this is truly equipment design for 240/120V 3? 4W, will either have to use the 1? autotransformer setup, or a 1:1 3? isolation transformer... assuming the equipment has only 4 power supply terminals.

 

[big john]

Not if the CPT is external to the warranteed equipment...

How do you propose connecting the secondary of the CPT to the control circuit without disconnecting that control circuit from the manufacturer's source within the cabinet?

 

[kwired]

The neutral point of a zig-zag would make it a 240V wye.

Correct, zig-zag kind of blew over my head at first - and you are correct that it would give about 139 volts instead of 120. Would be kind of expensive way to do it compared to autotransformer as well. Auto transformer or a isolation transformer only sized to the control circuit load is going to cost less.

One could use a full delta with center tap on one side as an autotransformer as well but kind of puts you in similar amount of copper and steel needed to build the transformer and uses more then needed to just supply a limited power control circuit.

The OEM does, apparently: If they're that particular, I don't see any way of installing a CPT that would make them happy. Mind you, if they're that particular I can also imagine them complaining about any derived neutral. :roll:

Agree, also why I earlier mentioned asking the manufacturer why they were too cheap to put in a control transformer in the first place.

Is the machine even listed? Most machines I have seen that need a neutral in the supply instead of installing a control transformer usually are not listed.

How do you propose connecting the secondary of the CPT to the control circuit without disconnecting that control circuit from the manufacturer's source within the cabinet?

We can open another can of worms with modifying the equipment by adding an aux contact.:roll:

 

[Smart $]

How do you propose connecting the secondary of the CPT to the control circuit without disconnecting that control circuit from the manufacturer's source within the cabinet?

As I previously posted...

...
BTW, if this is truly equipment design for 240/120V 3? 4W, will either have to use the 1? autotransformer setup, or a 1:1 3? isolation transformer... assuming the equipment has only 4 power supply terminals.

 

[ggunn]

A quite possibly stup... I mean, ignorant question which very well may have already been addressed: Why couldn't the OP place a 1:1 three phase transformer with a center tap between A and C phases out at a 3P4W panel, and tie the neutral thus generated through a GEC to a ground rod at the panel? Wouldn't it qualify as a separately derived system?

 

[GoldDigger]

A quite possibly stup... I mean, ignorant question which very well may have already been addressed: Why couldn't the OP place a 1:1 three phase transformer with a center tap between A and C phases out at a 3P4W panel, and tie the neutral thus generated through a GEC to a ground rod at the panel? Wouldn't it qualify as a separately derived system?

Yes, it would, but it would have to carry the full power of the machine just to provide the control power in a form the machine expects.

 

[ggunn]

Yes, it would, but it would have to carry the full power of the machine just to provide the control power in a form the machine expects.

If the machine requires a 4W service and internally runs its control circuitry off one of the non-high legs and ground, it seems to me that his choices are pretty much down to either installing a full load isolation transformer at the machine or pulling a neutral from the service. Bummer.

 

[Smart $]

If the machine requires a 4W service and internally runs its control circuitry off one of the non-high legs and ground, it seems to me that his choices are pretty much down to either installing a full load isolation transformer at the machine or pulling a neutral from the service. Bummer.

One other option, the 1? CPT 240:120 autotransformer method discussed.

 

[ggunn]

One other option, the 1? CPT 240:120 autotransformer method discussed.

If he did that would the neutral thus generated be at equal potential with earth ground at the machine? Would he (could he) bond the neutral to ground at the panel? It seems potentially (no pun intended) problematic to me.

But I am an engineer, not an electrician. I am here to learn.

 

[winnie]

If you use a non-isolated transformer to derive a neutral (zig-zag transformer to get a 3 phase neutral, or 240-120 transformer connected to one of the hot legs to get a 'single phase' neutral, then that derived neutral will be at _approximately_ ground potential.

However there is no guarantee that it will be at exactly ground potential; that would depend on the system voltage and impedance balance. Additionally you could not bond this derived neutral to ground because any difference in the two system 'neutral' voltages would cause current to flow through ground/ on bonded metal/ through the egc.

I don't know the current equivalent, but in the 2002 code 210.9 and 215.11 require bringing the grounded conductor to circuits supplied with autotransformers, by implication prohibiting the use of autotransformers to derive a neutral.

-Jon

 

[GoldDigger]

If he did that would the neutral thus generated be at equal potential with earth ground at the machine? Would he (could he) bond the neutral to ground at the panel? It seems potentially (no pun intended) problematic to me.

But I am an engineer, not an electrician. I am here to learn.

Nothing would blow up if the derived neutral were connected to ground, but the load current drawn by the controls in the machine would be able to return to the source neutral over the EGC if you did that.
And that would be objectionable current. It would trip a GFCI breaker at the source if there were one.
Without the extra bond the "neutral" current would get back to the source over the other hot lead and all would stay in balance.

 

[Smart $]

If he did that would the neutral thus generated be at equal potential with earth ground at the machine? Would he (could he) bond the neutral to ground at the panel? It seems potentially (no pun intended) problematic to me.

In theory, yes. However, even with an actual neutral run from the source, there exists a potential for a voltage gradient between it and ground at the panel under nominal conditions.

The pseudo-neutral cannot be bonded to ground. It would amount to a second MBJ/SBJ.

But I am an engineer, not an electrician. I am here to learn.

We all are here to learn or disperse knowledge. Engineer or electrician has no bearing.

 

[kwired]

If you connect a winding as an autotransformer between the two 120 to volt ground legs then the midpoint of that winding will be near ground potential, only because midpoint of the source end is grounded.

If you connect same winding between one 120 volt to ground lead and the high leg - you still have 120 volts to the mid point, from each ungrounded conductor, but one of them is 208 volts to ground at the supply end. Not fully sure but just doing a little drawing of some triangles to represent the supply and where that tap would end up at tells me it would be 120 volts to ground from our so called "derived neutral".

 

[GoldDigger]

If you connect a winding as an autotransformer between the two 120 to volt ground legs then the midpoint of that winding will be near ground potential, only because midpoint of the source end is grounded.

If you connect same winding between one 120 volt to ground lead and the high leg - you still have 120 volts to the mid point, from each ungrounded conductor, but one of them is 208 volts to ground at the supply end. Not fully sure but just doing a little drawing of some triangles to represent the supply and where that tap would end up at tells me it would be 120 volts to ground from our so called "derived neutral".

If A and C are the two ends of the source center tapped 240, then if your autotransformer runs from A to C you will have the midpoint, M, near the grounded neutral of the source. True.
If you run your autotransformer from either A or C to B, you will indeed have 120V from M to source neutral. An interesting result all right.
But that is not particularly relevant since we have presupposed that the machine derives its 120V either from A to neutral or from C to neutral.
If you try to derive your neutral from other than AC, then you will have to make sure that the machine also expects the neutral to be the midpoint of the same leg that you connected to.
And in that case for sure you should not ground your derived neutral!!!!

 

[kwired]

You shouldn't ground this "derived neutral" anyway if it is not part of a separately derived system. It is kind of no different then having more then one tap in the 240 volt supply winding. If you do this between A-C it just so happens the voltages of each winding align with one another when there is no load. Start putting load on them and they will start to vary because of voltage drop and you may even have some current circulating current between the two if they are bonded together, lose one of the connections and the other one will try to carry the load of the other winding, not so good for the smaller control transformer if it is trying to carry a heavy neutral load from other power or lighting circuits.

 

[ggunn]

You shouldn't ground this "derived neutral" anyway if it is not part of a separately derived system. It is kind of no different then having more then one tap in the 240 volt supply winding. If you do this between A-C it just so happens the voltages of each winding align with one another when there is no load. Start putting load on them and they will start to vary because of voltage drop and you may even have some current circulating current between the two if they are bonded together, lose one of the connections and the other one will try to carry the load of the other winding, not so good for the smaller control transformer if it is trying to carry a heavy neutral load from other power or lighting circuits.

OK, assuming that he can derive a neutral with an autotransformer between the A and C phases and connect it to the machine in question, what sort of current handling capacity will it need to have to be code compliant? If he were to pull a neutral from the service, wouldn't it have to be the same size as the phase conductors since it isn't just for current sensing? Would his derived neutral be any different in that respect? Would his autotransformer have to be capable of handling the same current as the phase conductors?

 

[winnie]

OK, assuming that he can derive a neutral with an autotransformer between the A and C phases and connect it to the machine in question, what sort of current handling capacity will it need to have to be code compliant? If he were to pull a neutral from the service, wouldn't it have to be the same size as the phase conductors since it isn't just for current sensing? Would his derived neutral be any different in that respect? Would his autotransformer have to be capable of handling the same current as the phase conductors?

Using an autotransformer to derive a local 'grounded conductor' in an already grounded system can _never_ be code compliant under the NEC.

If you don't connect this derived neutral to ground, then you have a conductor that is at _approximately_ ground potential, but which is not held to ground potential. Its voltage will fluctuate depending on the voltage balance of the supply conductors.

If you actually do connect this derived neutral to ground, then you have essentially created a ground fault. At best you have the equivalent of a ground to neutral fault; but read kwired's description: 'neutral' current will flow if the 'hots' connected to the autotransformer are not balanced. This transformer would need to be sized for the entire connected system, not just the local load. And the current would be flowing on all bonded metal, which are not supposed to carry current.

-Jon

 

[Smart $]

Using an autotransformer to derive a local 'grounded conductor' in an already grounded system can _never_ be code compliant under the NEC.

If you don't connect this derived neutral to ground, then you have a conductor that is at _approximately_ ground potential, but which is not held to ground potential. Its voltage will fluctuate depending on the voltage balance of the supply conductors.

If you actually do connect this derived neutral to ground, then you have essentially created a ground fault. At best you have the equivalent of a ground to neutral fault; but read kwired's description: 'neutral' current will flow if the 'hots' connected to the autotransformer are not balanced. This transformer would need to be sized for the entire connected system, not just the local load. And the current would be flowing on all bonded metal, which are not supposed to carry current.

-Jon

Are you saying this cannot be done compliantly by grounding the derived neutral, not grounding the derived neutral, or both?

I agree grounding the derived neutral would be noncompliant, but I don't see how leaving it "ungrounded" is a technical violation. First, it is not an SDS, and second, the system is already grounded. There would still technically be no conductor at greater than 150 volts to ground.

 

[winnie]

Take a look at 210.9 and 215.11 Circuits Derived from Autotransformers (in branch circuits and feeders, respectively). NEC requires that a grounded conductor be brought to and used by the autotransformer(s), with exceptions for certain line-line loads.

You might be able to use the derived neutral for something...but you would have to derive the neutral from a circuit which includes the neutral.....

-Jon

 

[Smart $]

Take a look at 210.9 and 215.11 Circuits Derived from Autotransformers (in branch circuits and feeders, respectively). NEC requires that a grounded conductor be brought to and used by the autotransformer(s), with exceptions for certain line-line loads.

You might be able to use the derived neutral for something...but you would have to derive the neutral from a circuit which includes the neutral.....

-Jon

Okay, I agree... autotransformer method is noncompliant.