Indeed , more thorough testing by the system integrator, as you folks do, would've saved us quite a bit of midnight oil. They'dve had to notice the substantial neutral current and the transformer getting too hot.
Neutral current viewed on the oscilloscope showed no discernible 60hz content. So our design people elected to disconnect the neutral as the most expedient fix.
Thanks for your interest - i do really appreciate that ..
OldJim, thanks for the clraification about why the failures appeared. For that very same reason it has always been a policy of ours to run a system or machine for 24 hours non-stop before delivering it to our customers. The time and power for the runoff cost less than even one service call, and fixing any problems prior to delivery means that the customers never see them. Shipping systems that don't have problems does a lot to enhance an organizations reputation, which is also quite valuable. And, in addition, we are able to learn from any problems that develop and avoid those design or assembly problems in the future.
Removing the neutral connection on the primary side would seem to be a reasonable solution, and it appears that nobody initially asked why that connection should be there in the initial design, since it did not carry any load current.
Oops - sorry , you're right i didnt make that clear.
What changed the first time was the duration of our test run.
Our usual run was for just an hour.
The transformers failed after seven hours at full load. They'd never before seen a run of that duration at that power. Even when we'd used the generators for longer periods it was at quite low load where third harmonic currrent wasn't excessive.
At low load the third harmonic current was modest enough, but it increased with excitation which increased with load.
Below half load they'd have lasted quite a long time.
The replacement transformers that failed more quickly were lower impedance. Engineering felt they'd be less lossy.
We reasoned that's the reason they lasted less time. Their delta secondary is a dead short for third harmonics because they're in phase as mentioned earlier. A hundred pound transformer just can't absorb all the harmonic from a two thousand pound generator. All the transformer has available to oppose the generator's ~5% third harmonic voltage content is its own impedance; and less impedance allowed more current.
Removing the neutral connection between the huge generator and the small transformer blocked third harmonic current, at the expense of mild neutral instability on the small transformer's primary side.. It's run fine since.
Thanks guys, for the questions - helps me improve my technical writing.
The generator and its excitation came as a package. The installer just hooked up the wires as shown.
The oversight originated at the supplier who integrated the package, which included the engine, generator, excitation, engine cooling and short term fuel delivery systems from several different manufacturers.
Our design reviews also failed to catch it, as did A/E's.
Lest i sound harsh on them,
it took us quite a while to find from the generator's original manufacturer just what was its harmonic content. It certainly wasn't given on the nameplate or in the instruction manuals.
I can understand that an excess amount of third hyarmonic power dissipated in the transformer would cause it to overheat, but I don't have a clue about why this problem suddenly appeared. And there is always that question about "what changed", when a working system stops working. So either that explanation was left out, or else I am missing something.
You leave the reader with the impression that it's the magnitude of the third harmonic content going into the smaller transformer that's the cause of the problem. But most power transformers can handle a somewhat higher flux level at 180 Hz than at 60 Hz. The real point here is it's the IMBALANCE at 180 Hz that's the real culprit (because this showed up in the neutral lead), and that could as well have been an imbalance in the load at that frequency as imbalanced generation across the phases of 180 Hz coming out of the generator. If you're going to publish a resolution of the problem please try at least to be accurate!
Digital healthcare devices and wearable electronic products need to be thoroughly tested, lest they live short, ignominious lives, an expert will tell attendees at UBM’s upcoming Designers of Things conference in San Jose, Calif.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.