Odd voltage to ground 4

[tpmc]

Hello everyone!
I have a 3 phase 208v buss with no neutral or ground (using case ground). I have a Buss plug connected with 3 phase wires (lets say A,B,C)& a ground wire going to a transformer. The transformer is 240v primary - 208v secondary. Were back feeding the transformer as we have 208v & need 240v. At the transformer I tied the ground wire from the buss plug to the ground lug on the transformer frame. The primary feed wires from the buss plug are connected phases A,B,C to X1,X2,X3, nothing is tied to X0. The secondary wires connected to H1,H2,H3 at the transformer and to the 3 phase breaker in the panel box. I ran a ground from the transformers ground lug to the panel boxes grounding bar. I installed the panel bonding screw to the grounding bar.
Reading between phases I have the correct voltages of 240v & 208v. X1,X2,X3 to ground in the transformer is 120v. H1(A)-gnd = 160v, H2(B)-gnd =98v, H3(C)-gnd =157v in the transformer or panel. I was expecting to see 120v!
Does anyone know why? Should I have XO connected to ground?

 

[waross]

It sounds as if your 240 volt system is high impedance grounded. The ground will consist of the winding and conductor resistances and capacitive reactances to ground. With a small system and an older relatively high ohms per volt meter the meter will form part of the ground impedance and change each phase voltage to ground as it is reading it.
If the transformer 240V is a wye winding then by all means ground the wye point. If the 240 volt winding is delta you don't have a good grounding point.
What I would do in such an instance depends on the size of the transformer. For a small transformer for a single machine I would probably leave it floating. For a larger system I would investigate grounding transformers or grounding impedances.
This is system grounding. This determines the magnitude of current flowing in a ground fault. It is not to be confused with safety grounding which is accomplished by connecting the exposed metal on all equipment to ground.
respectfully

 

[JensenDrive]

You did not mention an H0 bushing, so I assume one does not exist, so your neutral is floating. Sounds like a study of NEC rules on high impedance grounded systems is your next task. I would pipe up, but I am not a good resource on what the NEC says about corner grounding. There is a little info in NEC 250.36 on high impedance grouding.

 

[jghrist]

If the 240V winding is wye, and both the primary and secondary neutrals are grounded, then any neutral current (unbalanced load or third harmonic) in the 240 volt system will flow back to the 208V source through the ground wire. This could cause misoperation of ground fault protection. The 240 volt would not be a separately derived source and the neutral should be grounded at only one point, at the source.

 

[tpmc]

Guy,
I appreciate your guys quick replies. Sorry for the delay in my response. Here is more info on the transformer. It's 45 KVA, 240v delta & 208/120 wye. I'm feeding 208v 3 phase in on the wye with no neutral. Im getting 240v 3 phase out on the delta. The transformer is feeding a 3 phase panel. The panel is feeding small 3phase & single phase 240v loads. At this time there are a couple 50 amp & a couple 30 amp breakers being used.
Waross mettioned "an older relatively high ohms per volt meter the meter will form part of the ground impedance and change each phase voltage to ground as it is reading it." Not sure I understand this. Correct me if I'm wrong. I beleive your saying the voltmeters high resistance along with the resistance in the system ground together alter the voltage reading to ground. I used a Simpson 260 analog meter & a Fluke 80 series.
I appreciate all your help!
Tpmc

 

[jghrist]

The 240 volt system is ungrounded, so any measurements of line-to-ground voltage are meaningless.

 

[itsmoked]

Agreed.

Again your meter is reading a voltage,(to ground), that is present only because of stray capacitances and inductances. Probably wouldn't light a small light bulb. hence the meter even affects the reading dramatically.(for various reasons)

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[oldfieldguy]

jghrist & itsmoked are correct.

If you want to further check it, temporarily ground one pahse through a few hundred ohms of resistance and see what your voltages are then.

I have found that in many cases it there is sufficient capacitive and resistive path to ground, like a connected MCC's buswork, then these paths will form a virtual neutral and the measurements to ground with a high-impedance voltmeter like most of today's DVM's will return more believable readings.

old field guy

 

[tpmc]

Just wanted to thank all whom replied!

 

[jvjtech]

Oldfieldguy & others: I have started a new thread based, in part, on what OFG said above. Please see thread238-205765
Thanks, JVJ