Delta-Wye Transformer - Neutral Grounding

[LR1688]

Hello

I'm an engineering student at a South-African university. I'm working on a high temperature oven project. The oven has three graphite heating elements. Each with a resistance of about 1 ohm. To power these elements I have a 18 kVA delta-wye transformer. Primary side = 380V, secondary side phase to neutral voltage is 100V.

I connect each of the three heating elements between phase and neutral on the secondary side. To control the power to each heating element I use a SCR with phase angle firing to limit the power. I can control the power input to the three heating elements independently of the other two elements. I will therefore be able to create scenarios where the load is extremely unbalanced. For instance two phases on full power and one phase turned off.

I understand it is good practice to ground the neutral of the transformer on the secondary side. What is the reason(s) for this?

How should I do this grounding?

I hope the questions if formulated clearly, please let me know if I can clarify anything.

 

[Dennis Alwon]

bump

 

[ron]

There are lots of articles comapring the pros and cons of solidly grounded, vs resistance grounded versus ungrounded.

Some basic
http://ecmweb.com/content/basics-grounding-systems

Some written by resistance ground equipment manufacturers
http://www.eaton.com/ecm/idcplg?IdcService=GET_FILE&dID=167373

Some more technical IEEE documents

Essentially it is a trade-off of operational issues versus safety.

 

[Smart $]

Reason is to facilitate the operation of overcurrent device during a line to ground fault. When the system is not grounded, the first line to ground fault will not trip the overcurrent device (no path for current). The second fault will assist overcurrent protection tripping, but may expose personnel to perhaps fatal voltage gradients on exposed metal parts.

Ground by running a GEC to the X0 terminal of the secondary. Also install a system bonding jumper between X0 and grounding bus/case/supply-side bonding jumper.

 

[Besoeker]

Hello

I'm an engineering student at a South-African university. I'm working on a high temperature oven project. The oven has three graphite heating elements. Each with a resistance of about 1 ohm. To power these elements I have a 18 kVA delta-wye transformer. Primary side = 380V, secondary side phase to neutral voltage is 100V.

I connect each of the three heating elements between phase and neutral on the secondary side. To control the power to each heating element I use a SCR with phase angle firing to limit the power. I can control the power input to the three heating elements independently of the other two elements. I will therefore be able to create scenarios where the load is extremely unbalanced. For instance two phases on full power and one phase turned off.

I understand it is good practice to ground the neutral of the transformer on the secondary side. What is the reason(s) for this?

How should I do this grounding?

I hope the questions if formulated clearly, please let me know if I can clarify anything.

I think it was presented quite lucidly if I may say so.
It is usual for distribution transformers in UK to be delta star with the secondary star point to be earthed (grounded) at the transformer. I think your 380V system is most likely that way.

It fixes voltages WRT to earth and fault current paths.

Your 100V isolated system?
Don't know.....
Earthing the star point would also do the same thing.
Necessary? Mandatory? Probably not for a lab experiment.
And probably wouldn't add to what you get out of them.

A little off topic but, if you don't mind saying, which part of SA are you from?
I've been there a bit, Joburg, Durban, the Cape......and friends from Stellenbosch University.

 

[augie47]

Not to muddy the waters as I am ill equipped to do more than mention the fact that 250.21(A) notes that circuit AC circuits for "industrial electric furnaces" if applicable.

 

[Bugman1400]

Another important fact that is missing is, if your single phase 100V heating elements are connected when you have a neutral fault on one of the phases, the other two phases will now be applying 173V to the 100V devices. A grounded neutral prevents this from happening.

 

[Tony S]

How? The elements are connected phase to star point. You will get 173V to ground on two elements but the voltage across them will still be 100V.

 

[mbrooke]

Reason is to facilitate the operation of overcurrent device during a line to ground fault. When the system is not grounded, the first line to ground fault will not trip the overcurrent device (no path for current). The second fault will assist overcurrent protection tripping, but may expose personnel to perhaps fatal voltage gradients on exposed metal parts.

Ground by running a GEC to the X0 terminal of the secondary. Also install a system bonding jumper between X0 and grounding bus/case/supply-side bonding jumper.

Not if the system is bonded correctly.

 

[mbrooke]

Another important fact that is missing is, if your single phase 100V heating elements are connected when you have a neutral fault on one of the phases, the other two phases will now be applying 173V to the 100V devices. A grounded neutral prevents this from happening.

Not if the center of the elements isn't grounded down. I know of distribution systems running ungrounded where the pole pigs were banked ungrounded wye delta. Even when a single phase grounds down no neutral shift occurs on the pigs themselves since its left floating. Of course you still have the 1.73 voltage rise on the remaining 2 phases to ground.

 

[Bugman1400]

Another important fact that is missing is, if your single phase 100V heating elements are connected when you have a neutral fault on one of the phases, the other two phases will now be applying 173V to the 100V devices. A grounded neutral prevents this from happening.

This is misstated.....The potential to ground is what is elevated during a fault, not the potential across the phase devices.

Thanks Tony S and mbrooke.

 

[mbrooke]

This is misstated.....The potential to ground is what is elevated during a fault, not the potential across the phase devices.

Thanks Tony S and mbrooke.

Just to add, if the elements were connected to ground at the center, then the heater on the faulted phase would shut off, while the other two get 1.73 times voltage... but that's only if its grounded. Ungrounded elements means no voltage shift.

 

[Smart $]

Just to add, if the elements were connected to ground at the center, then the heater on the faulted phase would shut off, while the other two get 1.73 times voltage... but that's only if its grounded. Ungrounded elements means no voltage shift.

You guys are losing me here. There will be no [major] change in voltage on the other two lines to neutral while the third is faulted to neutral, or ground when grounded. I have no idea where you are getting this 1.73 time the voltage. Someone please provide a diagram or clearly explain the conditions.

 

[mbrooke]

You guys are losing me here. There will be no [major] change in voltage on the other two lines to neutral while the third is faulted to neutral, or ground when grounded. I have no idea where you are getting this 1.73 time the voltage. Someone please provide a diagram or clearly explain the conditions.

IF the heater neutral is grounded down, a change will take place when a phase grounds down. If the heater neutral is ungrounded, then no change will take place on the heating elements when a phase grounds down.

 

[Smart $]

IF the heater neutral is grounded down, a change will take place when a phase grounds down. If the heater neutral is ungrounded, then no change will take place on the heating elements when a phase grounds down.

Only if a neutral conductor is not run or is open between secondary and load. Similar to losing a neutral on a wye service, and having a line to ground fault. Essentially two faults have to occur for that to happen. That's getting pretty far off the wall for a discussion on whether or not to ground a wye secondary... IMO :happyyes:

 

[mbrooke]

Only if a neutral conductor is not run or is open between secondary and load. Similar to losing a neutral on a wye service, and having a line to ground fault. Essentially two faults have to occur for that to happen. That's getting pretty far off the wall for a discussion on whether or not to ground a wye secondary... IMO :happyyes:

I slightly, but it helps understanding for anyone who reads this in the future.