Reactive Generator Research for everyone to share

[tim123]

Hi Luc,
  I've done some tests, but I'm not sure my results match yours... Please see attached schematic...

- Can you tell me if the layout I'm using is right? Do I have the scope probes in the same places?
- I'm using A-B for the math function...

Basic Findings:

- At low capacitance, the 2 channels are out of phase
- With enough capacitance, the 2 channels go into phase
- Then adding extra caps - makes no difference.

- Power factor, and wattage goes up as phases converge
- Power thru the load (bulb) goes up as phases converge

It all looks pretty linear to me: When out of phase, less power usage is shown, but the bulb is less bright too (or off)... As the capacitance increases, the phases converge, power factor goes up, bulb gets brighter.

Some Details:

- My variac uses about 10w on minimum.
- In tests below, my standard for a 'lit' bulb - is where just one particular part of the filament turns red.
- I used a separate DMM to measure AC voltage across the bulb.
- Power factor shown by the meter varied between 0.5 and 0.9
- Power factor generally increases as the voltage goes up.

Without Reactive Circuit:
- Bulb connected directly to variac - lights with 19.5v AC
- Wattmeter says 12.9W

With Reactive Circuit:
- Bulb lights with 20v
- Wattmeter says 12.6W
- Note - I can increase the voltage from the variac to any amount, reduce the capacitance accordingly, and end up with the same voltage across the bulb, and (more or less) the same wattage shown on the meter. It varies between 12.3 and 13 W.

Do you have any comments or advice?

Regards
Tim

[gotoluc]

Hi Tim,

so it looks like you are on 220vac grid?

What are you doing with the MOT Secondary?

Luc


EDITED

I modified your diagram and added instructions

[tim123]

Hi Luc,
  the secondary was already shorted (said so in diagram).

I don't understand what the difference would be to use another resistor in series with the bulb...? It's a resistor itself... (In which case the probes you drew are in the same place as mine - so I did get that right then...)

Regards
Tim

[gotoluc]

Hi Luc,
  the secondary was already shorted (said so in diagram).

I don't understand what the difference would be to use another resistor in series with the bulb...? It's a resistor itself... (In which case the probes you drew are in the same place as mine - so I did get that right then...)

Regards
Tim

Hi Tim,

yes, I now see the Secondary says shorted!

A current measuring Resistor needs to be separate from the load and placed after it on the return Neutral side. It can be of a small value like 0.1 Ohm or even smaller when large loads are used, but the smaller the value the more precise its value needs to be to get good power calculations. You cannot use a light bulb as a Resistor to do any power calculation since its Resistance changes as it is being lit.

Make sure your probe Grounds are always together and on the Neutral side of the grid or you can fry your scope as they are usually common ground and grounded though the Grid ground. So you can imagine what would happen if you introduce them on the hot side... POOF!

Have fun

Luc

[tim123]

I've attached an image of my test bench - just for info...

Luc, I'm not sure what I should be looking for...

The capacitors don't seem to 'tune' anything - as I said - you just need 'enough' (for me, about 30uF) to correct the power factor, and any extra makes no difference.

Using less than enough (less than 30uF) - and the phases diverge. With about 5uF or less they're 90 degrees out. At this point there's very little power being delivered to the load, but my variac still draws some power (10-12w).

For my 13w in (to 'light' the bulb - one part of filament goes dull red), I get about 20v across the 50ohm bulb. (I know a bulb's resistance changes when it gets hot - but I'm not letting it get very hot.) I'm not sure if my DMM reads peak, or RMS, but either way the output power is 9watts or less. (8.8 if RMS, 6.3 if peak, I think)

So, with any amount of caps, it looks clear that I'm getting less out than I'm putting in.

So, questions:

1) Do you have a cap switch box - so you can change values while the circuit is running? If not - how do you do it?

2) Have you found that the phase is 90 deg out with very low capacitance, and converges as capacitance increases, to a plateau?
  i.e. Do your results concur with mine?

3) It seems to me that a resistive load *requires* the phases to be together... I think the very definition of resistance means that the power has to be 'used up' (converted to heat). It's possible that an inductive load would be different. What do you think?

Regards
Tim