"Dirty" voltage on bus.

[LMAO]

That's right. Clean power while coonnected to utility, a bit dirty power from one generator, and more dirty from another one. When both gens are in parallel, it's seems power is ok. I suggested them to take snapshots of power from each gen's ouput, with the load breakers open, so nothing would affect this test. It's important, because different people take readings using different equipment, under different conditions, at different points. Looking at this screen it's hard to tell if there is any problem. I would rather use the old good CRT scope, not this one.

I am sure others have already explained it: utility is like an infinite source; adding harmonics to it is like taking a leak in the ocean. Generators on the other hand have higher impedance so harmonic current create more harmonic voltage. One generator has even higher impedance thus dirtier power.
The only way to clean up line voltage when running on gens is by filtering.

 

[Besoeker]

I am sure others have already explained it: utility is like an infinite source; adding harmonics to it is like taking a leak in the ocean.

That's not actually correct. The UK commonly uses 11kV to 400V for distribution. I have measured harmonic distortion at the 11kV level that was outside the limits of what was deemed acceptable.
Had the utility been an infinite source no such harmonics would have existed.

 

[LMAO]

That's not actually correct. The UK commonly uses 11kV to 400V for distribution. I have measured harmonic distortion at the 11kV level that was outside the limits of what was deemed acceptable.
Had the utility been an infinite source no such harmonics would have existed.

If utility is already contaminated then you'll see the harmonics. I was referring to a relatively clean utility. I have seen both kinds. Here in NW Houston our power quality is pretty good, almost no harmonics. So if you use a very large transformer to feed non-active front end VFDs your line voltage is not going to get noticeably corrupted. But if you travel south near Mexico boarder where a lot of artificial lift pumps fed by VFDs are operating you'll see a lot of harmonics on line.

 

[Besoeker]

If utility is already contaminated then you'll see the harmonics. I was referring to a relatively clean utility. I have seen both kinds. Here in NW Houston our power quality is pretty good, almost no harmonics.

You might be surprised. And it depends on what you call clean.
I was visiting Toshiba Houston a few years ago. The supply was not all that great when I checked it.

 

[LMAO]

You might be surprised. And it depends on what you call clean.
I was visiting Toshiba Houston a few years ago. The supply was not all that great when I checked it.

Toshiba plant on the corner of Eldridge and Little York? Anyways, I consider less than 5% THDv to be clean. Higher than 8% is not acceptable. Anything in between is borderline.

 

[Besoeker]

Toshiba plant on the corner of Eldridge and Little York? Anyways, I consider less than 5% THDv to be clean. Higher than 8% is not acceptable. Anything in between is borderline.

I wouldn't. So we may have to agree to disagree.
But you might like to consider where that 5% comes from.

 

[Jraef]

This might help.
I have some graphics attached below that I created for a training class, taken from a scope where we connected a 75HP drive and motor to a load bank, then feed it from a 1500kVA transformer, then again from a 75kVA transformer (on purpose to demonstrate the effect, not to actually run it that way). So the drive and load were exactly the same in both cases. There was almost no significant VOLTAGE distortion when on the 1500kVA transformer, but very serious distortion when on the 75kVA transformer. So basically, we used this to demonstrate that with a given non-linear load, regardless of the concept of "infinite bus" on the utility side (which I agree, doesn't really exist), the more important factor is the overall available current in the system feeding the drive and the relative DEMAND that drive represents of the total available, which has the most effect on the voltage Voltage thd (at the PCC).

Here is the 75HP VFD and load when fed from a 1500kVA transformer:

In this case the Ithd was 37% from the drive, but that was such a small percentage of the Total available current from that transformer, the Itdd (Total Demand Distortion) had little effect and the Vthd was only 0.9%. You can see just a couple of little blips on the voltage (blue) sine waveform capture.

Here is the exact same drive and load when connected to a 75kVA transformer:

In this case the Ithd actually went down to 29%, but because that represented a large part of the total AVAILABLE current in the system, the Itdd went up significantly, which caused the Vthd to jump to 9.3%.

So in this case from the OP, doubling the source capacity by paralleling another generator effectively increased the total AVAILABLE current in the system, which lowered the Itdd represented by the drive, and therefore the Vthd it contributed.

 

[LMAO]

This might help.
I have some graphics attached below that I created for a training class, taken from a scope where we connected a 75HP drive and motor to a load bank, then feed it from a 1500kVA transformer, then again from a 75kVA transformer (on purpose to demonstrate the effect, not to actually run it that way). So the drive and load were exactly the same in both cases. There was almost no significant VOLTAGE distortion when on the 1500kVA transformer, but very serious distortion when on the 75kVA transformer. So basically, we used this to demonstrate that with a given non-linear load, regardless of the concept of "infinite bus" on the utility side (which I agree, doesn't really exist), the more important factor is the overall available current in the system feeding the drive and the relative DEMAND that drive represents of the total available, which has the most effect on the voltage Voltage thd (at the PCC).

Here is the 75HP VFD and load when fed from a 1500kVA transformer:
View attachment 12358
In this case the Ithd was 37% from the drive, but that was such a small percentage of the Total available current from that transformer, the Itdd (Total Demand Distortion) had little effect and the Vthd was only 0.9%. You can see just a couple of little blips on the voltage (blue) sine waveform capture.

Here is the exact same drive and load when connected to a 75kVA transformer:
View attachment 12359
In this case the Ithd actually went down to 29%, but because that represented a large part of the total AVAILABLE current in the system, the Itdd went up significantly, which caused the Vthd to jump to 9.3%.

So in this case from the OP, doubling the source capacity by paralleling another generator effectively increased the total AVAILABLE current in the system, which lowered the Itdd represented by the drive, and therefore the Vthd it contributed.

Other way to explain what we have been advising OP. Good description.

 

[kwired]

Peeing in the ocean is a good analogy of what I was trying to say earlier. Then others chimed in saying the utility may be dirty to begin with, which is true, but you still have the utility being the like the ocean and the load added from one facility isn't usually enough to cause much added contamination no matter how poor that load may be. But now switch to on site generation and that power supply is more like a local water supply tank compared to an ocean and it doesn't take as much foreign material to contaminate it.

 

[Besoeker]

Peeing in the ocean is a good analogy of what I was trying to say earlier. Then others chimed in saying the utility may be dirty to begin with, which is true, but you still have the utility being the like the ocean and the load added from one facility isn't usually enough to cause much added contamination no matter how poor that load may be. But now switch to on site generation and that power supply is more like a local water supply tank compared to an ocean and it doesn't take as much foreign material to contaminate it.

Might I politely suggest that you do some measurements and then you might be in a position to validate your points.

 

[kwired]

Might I politely suggest that you do some measurements and then you might be in a position to validate your points.

I'll start with measuring the ocean

I'll get back to you when I am done with that

 

[Besoeker]

I'll start with measuring the ocean

I'll get back to you when I am done with that

On the other hand, you could measure and analyse your supply voltage and get back to me when you have done that.

And, FWIW, the oceans account for 1,314,829,398,016,000,000 cubic metres of water.