PT voltage reading problem

[jraef]

This one has me scratching my head, but I figure the group consciousness will hold the necessary experience for a forensic evaluation from afar. Don't let me down...

We have a 3 phase 480/277V Y-Y PT in an MCC, 4:1 ratio feeding a Multilin PQM. It's been sitting for 4 years not connected to anything (in Las Vegas, so it's been warm and dry). Now they are finally commissioning the system and they are getting a strange reading from the PT. Line to Line it is reading 120V as it should, but Line to Neutral it is reading 81V, not 69V as it should. I'm thinking they have a problem with the N-G connection and it is floating a bit. Any other ideas on what might be happening or reasons why mine would be "inconsistent with reality"?

Drawing attached, it's a scan of a fax, so it's a bit fuzzy but you should be able to see the pertinent information.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[johnadowsett]

If your HV neutral is floating, then you do not have a path for the 3rd harmonic (and other harmonics) magnetising HV currents. The result is the ph to ph voltages will read ok, however the phase to neutral voltages will not read correctly as they have a 3rd harmonic & other harmonic components.

 

[waross]

I have seen strange readings a couple of times on unloaded wye:wye banks. Once with distribution transformers and once with small dry types, about 10 KVA.
We got a scope on the dry types. There was a distorted waveform, probably related to transformer hysteresis. Picture a positive half cycle followed by about 5 degrees of zero volts, then the negative half cycle followed by the same 5 degrees of zero volts. As soon as even a small load was connected the voltages snapped to the correct ratio. It didn't help that this was back in the day of d'Arsonval meter movements which did not like distorted wave forms. Try some 100 Watt lamps across the PT secondaries and see what happens.
The impedance of your relay may be too high to load the transformers.
And it may be something else entirely. grin

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[prc]

All the phases in secondary showing 81 V or only one phase?

 

[jraef]

Yes, all 3 phases are showing 81V to the neutral.

I passed on the neutral bonding possibility, they are checking today. Hadn't thought about the loading possibility, worth a shot at testing it. Thanks for the suggestions so far. I'll report back.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[DTR2011]

I noticed the neutral CT is not grounded in the enclosed pdf. You will want them to check that as well.

 

[jraef]

Yes you're right. I'll mention it. Thanks.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[DTR2011]

You can use the PQM itself to capture a waveform and analyze (and share) it.

 

[scottf]

It shouldn't be a loading issue. The VT error at 0 VA burden is not going to be 17%. Most VT manufacturers (at least in IEEE standard) design the VT to be within the stated accuracy class of 0VA up to the stated burden rating.

Almost has to be a floating neutral issue from what you described.

One other thing that stuck out to me was that you listed the VTs as being rated 480Y/277V. Assuming a 120V rated secondary, that wouldn't be a 4:1 ratio, but would be 277:120V = 2.3:1 ratio.

If the ratio really is 4:1 with a 120V rated secondary, then you technically are using VTs intended for line-to-line connection. (probably rated 480Y/480:120V)This means you're operating the VTs at approx. 58% rated voltage. IEEE states that the VTs only have to meet the stated accuracy class from 90%-110% rated voltage. Normally, a typical VT wouldn't have that degree of accuracy change at 58%, but since you mention it's a 3-phase VT, it could be using some non-traditional construction, e.g. an auto-transformer, etc...