COP 20.00 (2000%) Times, Reactive Power Energy Source Generator,

[TinselKoala]

Prasad: remember, current does not "contain" MW.

Watts are produced when current flows thru a resistance and all generators, bus work, transformers etc have some unless they are superconducting. So when current flows thru a resistance, watts are produced in the form of heat in an amount equal to I-squared-R. Those watts dissipate energy which must have come from somewhere. In this case, "somewhere" is your generator which converted mechanical energy into electrical energy. Your generator received the mechanical energy from the conversion of thermal energy in your turbine, therefore the turbine must have burned more fuel. So increased reactive power generation increases fuel consumption.

The law of conservation of energy always applies

That's right in concept but a pretty silly way of putting it, and the manner of expression reveals to me that the poster still has misconceptions about power and energy.

The Watt is a _rate_, the rate of energy units passing your measurement point per time unit. The Joule is a quantity, the unit of energy, and the second is the time unit used in the Systeme Internationale. So a Watt is One Joule Per Second, zipping past your Joule Detector (tm TKLabs). This Power can be dissipated in a resistance as heat, but that's not the only way Joules can pass a measuring point! Those "Watts" can be measured at any point in a circuit, not just in the dissipating elements, as the Joules flow past on their way to being put to use somewhere else.
Does your car "produce Miles Per Hour" when it is rolling down the road? Well, I suppose you could put it like that but it's pretty silly to look at it that way, and looking at it that way will "steer you wrong" eventually.

But the main point is clear: Losses have to be made up or the system collapses, and if you have more losses in your transmission line, you need to replace those losses with increased fuel consumption at the prime mover.

[Kator01]

Tinselkoala,

x_name41 believes that the difference between load on-modus minus load-off modus is:
Pdiff = 1.9504 - 1.863 = 0.087 Watt

he then compares this to the 24 V- Bulb in operation P = 24 x 0.02 Watt = 0.48 Watt


P bulb 0.48 W div by 0.087 W input-difference equals cop = 5.5

...but ....

...first ...the bulb is not full lit and

..second.. to simply substract power-output of a source  under load with  power-output of the same source without load is one of the
biggest fallacies, it´s a wrong method because the  transformer of a power-source might use 10 watt in idle-mode
and still operate below 1 Watt hysteresis-losses while delivering another 9 Watt  to a load, indicating 10 Watt on its meter.

So it will deliver effectively 9 Watt to the load while the uninformed experimenter believes that he has got his load operating at 9 Watt using almost nothing ( i.e 10 W minus 10  Watt)

So the only way to measure correctly consist of using a 0.1 Ohm shunt in the minus-line and measure the voltage-drop across the shut and multiply this with the voltage at the load.

Kator01

[TinselKoala]

Tinselkoala,

x_name41 believes that the difference between load on-modus minus load-off modus is:
Pdiff = 1.9504 - 1.863 = 0.087 Watt

he then compares this to the 24 V- Bulb in operation P = 24 x 0.02 Watt = 0.48 Watt


P bulb 0.48 W div by 0.087 W input-difference equals cop = 5.5

...but ....

...first ...the bulb is not full lit and

..second.. to simply substract power-output of a source  under load with  power-output of the same source without load is one of the
biggest fallacies, it´s a wrong method because the  transformer of a power-source might use 10 watt in idle-mode
and still operate below 1 Watt hysteresis-losses while delivering another 9 Watt  to a load, indicating 10 Watt on its meter.

So it will deliver effectively 9 Watt to the load while the uninformed experimenter believes that he has got his load operating at 9 Watt using almost nothing ( i.e 10 W minus 10  Watt)

So the only way to measure correctly consist of using a 0.1 Ohm shunt in the minus-line and measure the voltage-drop across the shut and multiply this with the voltage at the load.

Kator01

Exactly (more or less.)
It's like the garden hose analogy. Turn on your garden hose valve and direct the hose out into your back yard. Measure the flow rate at the valve. This is the input power from the power supply. Now bring a bucket over and stick it into the stream from the hose and fill up the bucket. This is your output power in the light bulb. Is there a difference in the flow rate at the garden hose valve, whether or not the bucket is collecting water from the other end? No, of course not. So the water in the bucket is free! Right.

[listener191]

Hence the Babcock statement "and therefore burn less coal" doesnt ring true for distributed power.

He seems to forget about the reality of transmission systems and the fact that most of these large generators are designed to run at maximum efficiency with a PF not less that 0.8

I would still like to see a practical test with SERPS running on a small generator, which would eliminate distribution losses.


Barry

[listener191]

All,

Attached is a SERPS timing circuit that is digitally controlled by a clock internal to the LTC6993's.

Trigger is taken to two LTC6993-1 from the postive going zero cross. These set the delays for the start of switching pulses for the 1st and 4th Quadrants. The outputs are OR'd and trigger an LTC6993-3 on the falling edges. The LTC6993-3 controls the pulse width.  1st quadrant Charge 2nd quadrant Discharge.

A similar arrangement is triggered from the negative going zero cross for 2nd and 3rd quadrants.  3rd quadrant charge 2nd quadrant discharge.

This is the timing scheme I have decided to try first however, by adjusting the DIV input voltage and the SET current, a wide range of timing can be achieved

By using a potential divider, the SET current can be voltage controlled either to gang the pulse width control together or allow control of all devices from the D/A on a microcontroller.

Although a microcontroller could handle the timing using interrupts, I chose not to spend time on software development for the first tests and use a manually controlled setup. Also I dont like relying on CPU's for low level timing functions and prefer those to be controllable but autonomous once setup.

The LTC6993 is a cheap device, offers a lower component count than a 555 based circuit and has easy control interfacing.

The values shown have been calculated to work within range of the respective quadrants for 50/60Hz, as a starting point.

Barry