Calculation of machine isolation transformer

[Ashu]

Good morning, everyone,
NEC doesn't seem to have any code specifying how to calculate an isolation transformer (same primary and secondary) size. In my case I need to find an optimized way to size the isolation transformer which supplies to a machine having hundreds of motors. Most of the motors in the system are Servo motors which are sized for the peak load requirement; however, the actual draw is very less. In some cases, it is only 25% of the rated nameplate data.
Historically, we have considered the sum of actual motor nameplate data plus 15% additional for the isolation transformer calculation. However, actual machine current draw is much less than that.
The question I have is should I be allowed to use the actual current draw of the machine (with some safety factor) and not based on the rated motor currents?
If yes, I see a possible indirect issue: I would assume, the side effect of it will be a reduced sized cables and breaker from the plant side which supply to this transformer. Similarly, reduced sized cable supplying to the machine. However, machine's main circuit breaker is calculated based on article 670.4 which states that the supply conductors will have ampacity not less than 125% of the highest rated motor plus the sum of other loads. This means the machine's main breaker will be larger than the plant circuit breaker supplying to machine through isolation transformer.
I am stuck in this loop while trying to optimize the transformer size to save huge cost for the machine. I will appreciate any guidance on this.
Thanks.

 

[tortuga]

if the transformer is a 1:1 your input voltage = output voltage,
Then for the purposes of 670.4 it covers the entire feeder conductors that start at 'plant circuit breaker' not just the conductors after the 1:1 isolation transformer.

 

[Deltaforce]

Good morning, everyone,
NEC doesn't seem to have any code specifying how to calculate an isolation transformer (same primary and secondary) size. In my case I need to find an optimized way to size the isolation transformer which supplies to a machine having hundreds of motors. Most of the motors in the system are Servo motors which are sized for the peak load requirement; however, the actual draw is very less. In some cases, it is only 25% of the rated nameplate data.
Historically, we have considered the sum of actual motor nameplate data plus 15% additional for the isolation transformer calculation. However, actual machine current draw is much less than that.
The question I have is should I be allowed to use the actual current draw of the machine (with some safety factor) and not based on the rated motor currents?
If yes, I see a possible indirect issue: I would assume, the side effect of it will be a reduced sized cables and breaker from the plant side which supply to this transformer. Similarly, reduced sized cable supplying to the machine. However, machine's main circuit breaker is calculated based on article 670.4 which states that the supply conductors will have ampacity not less than 125% of the highest rated motor plus the sum of other loads. This means the machine's main breaker will be larger than the plant circuit breaker supplying to machine through isolation transformer.
I am stuck in this loop while trying to optimize the transformer size to save huge cost for the machine. I will appreciate any guidance on this.
Thanks.

If the plant circuit breaker sized protect primary conductors, isolation transformer and secondary conductors, sizing machine breaker larger 670 no issue, IMHO
If the primary breaker not protect secondary, provide separare secondary breaker also

 

[Ashu]

if the transformer is a 1:1 your input voltage = output voltage,
Then for the purposes of 670.4 it covers the entire feeder conductors that start at 'plant circuit breaker' not just the conductors after the 1:1 isolation transformer.

That's correct. However, in that case according to 670.4, the conductor supplying to the machine should be calculated based on the rated motors current and not based on the actual current consumption of the machine. So, I can't reduce the size of the conductors feeding to the machine?

 

[electrofelon]

That's correct. However, in that case according to 670.4, the conductor supplying to the machine should be calculated based on the rated motors current and not based on the actual current consumption of the machine. So, I can't reduce the size of the conductors feeding to the machine?

I dont see any way to reduce your supply conductors to a more realistic level. This section appears to be yet another case of the code writers really just not knowing what they're talking about and/or not having ample experience to be making such requirements.

I recently hooked up an industrial machine that took something like a 100 amp conductors and supply breaker. The main conductors internal to the machine were10 gauge.

 

[don_resqcapt19]

Does this equipment have a nameplate? If so I am going to size everything based on the nameplate and ignore any code rules that suggest otherwise.

 

[electrofelon]

Does this equipment have a nameplate? If so I am going to size everything based on the nameplate and ignore any code rules that suggest otherwise.

In this case, the current on the nameplate is required to be the full load current of all the motors. Seems that NEC is operating out of its lane here, but thats what it says.

 

[don_resqcapt19]

In this case, the current on the nameplate is required to be the full load current of all the motors. Seems that NEC is operating out of its lane here, but thats what it says.

That is not what the NEC says.

The full-load current shown on the nameplate shall not be less than the sum of the full-load currents required for all motors and other equipment that can be in operation at the same time under normal conditions of use.

This allows the manufacturer to base the nameplate current on the real world operational characteristics of their machine..

 

[winnie]

The full-load current shown on the nameplate shall not be less than the sum of the full-load currents required for all motors and other equipment that can be in operation at the same time under normal conditions of use.

This allows the manufacturer to base the nameplate current on the real world operational characteristics of their machine..

IMHO that would make perfect practical sense, but I don't think the wording actually reads that way.

The wording says that you must take all the motors in operation at the same time and add up their full load currents. So if you have a machine with 10 motors, but only 5 operate at any one time, you only have to add up the load for 5 motors. But if you happen to have servo controlled motors that are all operating continuously at partial power. If you have 10 motors each operating on average at 50% power, you have to add up the full load for 10 motors.

Though when you talk about servo controlled motors with PWM the definition of 'operation at the same time' does get a bit fuzzy I suspect that this conversation really needs to pivot to NFPA 79 for the requirements to calculate the machine nameplate current, and I've never studied that.

-Jonathan

 

[electrofelon]

That is not what the NEC says.

This allows the manufacturer to base the nameplate current on the real world operational characteristics of their machine..

while i agree that there certainly may be cases where 2 or more motors will not operate at the same time, those cases are probably a minority, and besides i do not see the NEC's phrase as being anywhere equivalent to your phrase of "current based on real world operational characteristics".

 

[don_resqcapt19]

while i agree that there certainly may be cases where 2 or more motors will not operate at the same time, those cases are probably a minority, and besides i do not see the NEC's phrase as being anywhere equivalent to your phrase of "current based on real world operational characteristics".

Just how I want to read the rule.
In any case, I would be using the nameplate current, especially if the equipment happens to be listed.

 

[winnie]

Just how I want to read the rule.
In any case, I would be using the nameplate current, especially if the equipment happens to be listed.

This actually raises an interesting possibility. If the OP can get the equipment 'listed' I wonder if the listing agency could provide a documented lower current requirement.

 

[jim dungar]

....wonder if the listing agency could provide a documented lower current requirement.

The NEC does not apply to listed equipment. If it did there would be a lot more appliances with 20A plugs instead of 15A ones.

 

[Ashu]

Does this equipment have a nameplate? If so I am going to size everything based on the nameplate and ignore any code rules that suggest otherwise.

I am the one who design this equipment and have mentioned machine nameplate data based on the calculations considering motor nameplate information and not the actual consumption of the machine.

 

[tortuga]

IMHO that would make perfect practical sense, but I don't think the wording actually reads that way.

The wording says that you must take all the motors in operation at the same time and add up their full load currents. So if you have a machine with 10 motors, but only 5 operate at any one time, you only have to add up the load for 5 motors.

I'll respectfully disagree in 670.4 (670.3 in the 2017 version) it says in part:
:

Where unusual type loads,
duty cycles, and so forth require oversized conductors
or permit reduced-size  conductors, the required capacity shall be
 included in the marked “full-load current.”

So the designer is allowed to determine the 'full load current' and mark it.

 

[tortuga]

That's correct. However, in that case according to 670.4, the conductor supplying to the machine should be calculated based on the rated motors current

The size of the supply conductors before and after the transfromer should be the sum of

• 125 percent of the full-load current rating of the highest rated motor
• 125 percent of the full-load current rating of all resistance heating loads
• The sum of the full-load current ratings of all other motors and
  apparatus, based on their duty cycle, that may be in operation at the same time.

 

[electrofelon]

I'll respectfully disagree in 670.4 (670.3 in the 2017 version) it says in part:
:

Where unusual type loads,
duty cycles, and so forth require oversized conductors
or permit reduced-size  conductors, the required capacity shall be
 included in the marked “full-load current.”

So the designer is allowed to determine the 'full load current' and mark it.

What is "unusual type loads"? I would expect most industrial machines to have many motors that run intermittently, not a full load, etc and I would not consider that unusual.