Protection for a 480 volt machine

[Greg Wrenn]

I have a forge machine that has heating probes that will heat up dies. Once in awhile the probes will go bad and because we have a 480 delta system with no neutral, the bad probe will create fluctuations in the voltages on the lines throughout the plant. This last round I had 400 plus volts on each of two lines to ground and another line with about 60 volts to ground. I was suggesting that we get an isolation transformer for that machine so that I could give it its own neutral. Is their another solution that might be cheaper but still at least keep the issue to that machine.

 

[tom baker]

Is an ungrounded delta system, or corner grounded system?

 

[jim dungar]

I could give it its own neutral.

You are asking for isolation. You need a transformer.

Do you have a ground detection scheme?

 

[Greg Wrenn]

I am still learning some of the terminology, I can tell you that one of the first 480 panels that is inside the building that the outside transformer feeds first has a ground to it so I would say we have a grounded system.

 

[GoldDigger]

I am still learning some of the terminology, I can tell you that one of the first 480 panels that is inside the building that the outside transformer feeds first has a ground to it so I would say we have a grounded system.

A grounded system is one in which the current carrying circuit conductors have a solid reference to ground voltage. Even an ungrounded system will have a safety ground connection for the enclosure and non current carrying parts. The main justification for an ungrounded system is that a short to ground, as with your heating probes, will not trip a breaker and shut down the process catastrophically (like in a system where molten product will harden in the machines.) You need to look at the risks of an unplanned power loss.

 

[jim dungar]

I am still learning some of the terminology,

Terminology is very important. Incorrect usage can end up with incorrect answers and overlooked opportunities.

What does the nameplate on your service entrance equipment say?
Other than messed up voltages, how do you know a probe went bad?

 

[drcampbell]

How critical is the forge machine to the overall operation? You could install current relays, a phase monitor and/or a ground-fault monitor which will shut it off when a fault occurs.

How much of an actual problem are the different voltages to ground? Motors and heaters won't be affected, and if you had variable-frequency drives or anything else sensitive to it, you'd have an entirely different (and bigger) problem.

By "probes", do I assume correctly that you're talking about cartridge heaters or insertion heaters?

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

 

[GoldDigger]

One other comment: If your incoming POCO service is truly designed as ungrounded, there is almost certainly not a fourth neutral wire as a current carrying wire from the center tap of their transformer. That means that none of your loads, even single phase loads, should be connected to anything other than the three "hot" wires. And as long as that is the case, the odd voltage readings you get should not cause any operational problems with any equipment. The ground fault in the heater should have no effect on the line to line voltages.

 

[Greg Wrenn]

Terminology is very important. Incorrect usage can end up with incorrect answers and overlooked opportunities.

What does the nameplate on your service entrance equipment say?
Other than messed up voltages, how do you know a probe went bad?

When I take out the heating probe and test it. It has a short in it. I can test continuity between the outer casing and the wires that go through it. I should not get any reading but the tester goes straight to zero. Also I can tell if a probe is getting weak because the resistance on a probe has a certain reading and sometime I will get weaker resistance than I am supposed to. The problem with our electrical system is that, that one bad probe feeds all the way back through our 480 system. Usually the probes do not continue to heat up the dies on the machine to the correct temp and is easier to find. This time the dies were still reaching temp with the three probes.

 

[Greg Wrenn]

How critical is the forge machine to the overall operation? You could install current relays, a phase monitor and/or a ground-fault monitor which will shut it off when a fault occurs.

How much of an actual problem are the different voltages to ground? Motors and heaters won't be affected, and if you had variable-frequency drives or anything else sensitive to it, you'd have an entirely different (and bigger) problem.

By "probes", do I assume correctly that you're talking about cartridge heaters or insertion heaters?

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

The forge is the machine that starts the process for a lot of the other processes. It creates the yokes that the other machines drill out and/or thread before being assembled with other machines. That was why I was thinking an isolation transformer to keep it separate from everything else. I have had four machines blow fuses during this time and lost one transformer on a single phase 480 machine.

The probes are long cylinder probes that are inserted into the plates that hold on to the dies.

 

[Greg Wrenn]

One other comment: If your incoming POCO service is truly designed as ungrounded, there is almost certainly not a fourth neutral wire as a current carrying wire from the center tap of their transformer. That means that none of your loads, even single phase loads, should be connected to anything other than the three "hot" wires. And as long as that is the case, the odd voltage readings you get should not cause any operational problems with any equipment. The ground fault in the heater should have no effect on the line to line voltages.

The way it was explained to me by a previous maintenance supervisor is that even on our 240 volt bus bar system that we use for some machines here we use the ground as the neutral. This is an old building and it has been suggested before that they need to upgrade the system but it is felt the cost is not justified for doing so.

 

[Greg Wrenn]

I have replaced the defective probe and the voltage returned to "Normal" for this plant building. L1 - 280v, L2 -299v, L3 - 267v. I am going to go ahead and install an isolation transformer to isolate this machine. It seems like it is the best option from everything I am seeing and learning. Thanks all for your help.

 

[GoldDigger]

I have replaced the defective probe and the voltage returned to "Normal" for this plant building. L1 - 280v, L2 -299v, L3 - 267v. I am going to go ahead and install an isolation transformer to isolate this machine. It seems like it is the best option from everything I am seeing and learning. Thanks all for your help.

It is good that you have identified a defective part that was causing the unintended ground connection, and an isolation transformer will prevent this from interacting with the rest of the plant next time. But you then need to install a ground detector on the secondary circuit so that you can identify a single ground fault and repair it before a second ground fault occurs and takes the machine down without notice.
As for the rest of the plant, you appear to have an ungrounded or high resistance grounded system feeding line to ground loads (since there is no neutral). This is simply unacceptable design and needs to be investigated and corrected.