[llafro]

I have been a lurker on this site for some time and have found a lot of useful information here. I have a question that I haven't been able to answer by searching.

Background:
I am working on installing a 480 V 3-phase 70 kW 3-71 Detroit diesel generator under a railroad passenger car. The generator and control panel we are using are second-hand, and of course I have no manuals. I managed to work out how the control panel works and have wired it and the various gauges on it to work. The system is set to start and stop the generator automatically when the "hotel" power provided by the locomotive or ground power is turned off. The only part of the system that I can't get working is the voltmeter.

When running off the generator, the output is right on at 479 volts, and 60.01 Hz. Thanks to the electric governor on the system, the output is rock solid. The car is wired to run large loads at 480, and uses 208 wye for the smaller loads.

The voltmeter appears to be calibrated for 240 volts. I installed a small 480x240 240/110 general purpose transformer between two legs of the 480 VAC from the generator. The problem is that I am getting something like 265 volts from the transformer, which makes my voltmeter read 530 volts. Nominally, I should be getting 240 volts out of this setup. Any suggestions as to why I'm not getting the correct voltage?

 

[Greg]

Your meter is out of cal. I would suggest installing a Newport digital meter. If you want send the spec on the equipment and I can lay out the metering for you just tell me what you want, i.e. VAC, AMP, KW.

 

[micromind]

Small general-purpose xfmrs are usually wound with a slightly higher turns ratio. As any xfmr is loaded, output voltage will fall (impedance, core losses, etc.). This is more pronounced on smaller units, so they're intentionally wound for higher than nameplate voltage at no load. At full load, the output is pretty close to nameplate.

For metering, potential xfmrs are used (sometimes called voltage xfmrs). These have an exact turns ratio, resulting in exact voltage output. They're also expensive.

If the gen is 12 lead, and is wye connected for 277/480, you'll have 240 across 7 & 8. That's where I'd connect a 240 voltmeter.

 

[brian john]

1. Use a multimeter to verify output voltage.
2. Use a on site voltage supply to check the meter calibration
3. 0ther than that you have the wrong taps from the transformer.

 

[llafro]

Your meter is out of cal. I would suggest installing a Newport digital meter. If you want send the spec on the equipment and I can lay out the metering for you just tell me what you want, i.e. VAC, AMP, KW.

I think the panel meter calibration is correct. Using my handheld meter the other day, I got 268 volts at the terminals for the meter, and 479 volts across any two of the three outputs from the generator. The panel meter reads approximately 540 or so (it's a bit hard to read, as the scale goes from 0-600 VAC on a 2.5" meter).

Small general-purpose xfmrs are usually wound with a slightly higher turns ratio. As any xfmr is loaded, output voltage will fall (impedance, core losses, etc.). This is more pronounced on smaller units, so they're intentionally wound for higher than nameplate voltage at no load. At full load, the output is pretty close to nameplate.

For metering, potential xfmrs are used (sometimes called voltage xfmrs). These have an exact turns ratio, resulting in exact voltage output. They're also expensive.

If the gen is 12 lead, and is wye connected for 277/480, you'll have 240 across 7 & 8. That's where I'd connect a 240 voltmeter.

So, it could be that I would only get 240 volts with more of a load across the transformer? The transformer is only 100 VA. I know the meter imposes almost no load on the circuit.

1. Use a multimeter to verify output voltage.
2. Use a on site voltage supply to check the meter calibration
3. 0ther than that you have the wrong taps from the transformer.

1. The reading I get from my handheld on the output is 479 volts, and 60.01 Hz. on the output from the generator.
2. Using a socket in the car that measures 119 VAC on the handheld gives approximately 240 on the panel meter, so I think the calibration is correct at 2:1.
3. The transformer is connected per the diagram on the front with H1 and H4 connected to line, H2 and H3 connected together, X1 and X4 connected to load, and X2 and X3 connected together. The transformer is a Square D 100SV1A. I bought it new in the box on eBay for $5.99. So, if it turns out not to work, I won't be heartbroken

I have been wondering if connecting the transformer as an autotransformer might not give a better reading. I may just try it to see.

It's just my own irritation over not being able to make this thing work right that has me going to this extent. I plan on buying one of these in any case to meter things:

http://www.powermeterstore.com/p3781/em_6459_(1_0).php