Reactive power - Reactive Generator research from GotoLuc - discussion thread

[gotoluc]

I tried it and it stopped and displays "self calibration failed at step 3"

I'll try it without my 100x probe

Luc

[gotoluc]

Here are the shots with the factory 10x probe on channel 1. Looks to me about the same offset as using the 100x probe!  so I guess it's not the probe causing the offset.

The load is my 10 Ohm 1% 50w Resistor with 10 volts RMS across it = 10 watts
I also took the opportunity to make my own precision CSR (shunt resistor) out of Resistive wire which makes 0.1 Ohm with about 2 1/8" inch of wire. Now I have a precise CSR which also should more reliable then a wire wound CSR.

I think this should be precise enough to get back to testing. What do you think?

Luc

[poynt99]

Here are the shots with the factory 10x probe on channel 1. Looks to me about the same offset as using the 100x probe!  so I guess it's not the probe causing the offset.

I suspected the current probe might be the problem, not the 100x voltage probe.

The load is my 10 Ohm 1% 50w Resistor with 10 volts RMS across it = 10 watts
I also took the opportunity to make my own precision CSR (shunt resistor) out of Resistive wire which makes 0.1 Ohm with about 2 1/8" inch of wire. Now I have a precise CSR which also should more reliable then a wire wound CSR.

I think this should be precise enough to get back to testing. What do you think?

Luc

The only potential issue with the nichrome wire idea, is that the value may change quite a bit if it warms up. I guess you will have to try it and see.

Your MEAN values seem a little closer to one another now...within 2%, so perhaps that is acceptable for what we are doing here.

Unless I find something else, I see no reason that you can't go back and start remeasuring the tests you wish to remeasure for comparison. I for one would like to see you remeasure the simple cap and resistor load circuit.

I would caution however, that it would be a good idea to check the MEAN power with CH2 both inverted and non-inverted. If there is a big discrepancy between the two, then we should try to figure out why. And remember to get about 10 MATH cycles (5 Voltage cycles) on the display for the measurement (you have 12 in your scope shot above, and that is perfect).

[gotoluc]

I suspected the current probe might be the problem, not the 100x voltage probe.

I have 3 other factory probes, I can try them but quite sure I keep rotating them, so I don't think the probe is the problem. I also do probe comp check quite often.

The only potential issue with the nichrome wire idea, is that the value may change quite a bit if it warms up. I guess you will have to try it and see.

Very good point!... I'll make sure to use it only on small loads that don't heat the wire. Anyways, I'm only interested in circuits that don't use much power.
 

I for one would like to see you remeasure the simple cap and resistor load circuit.

Well, I quite sure you know the results!... It should be under unity with probes set to DC coupling. One very valuable lesson learned there. Too bad it took so many pages to figure it out. Also found out many others were doing the same thing.

I would caution however, that it would be a good idea to check the MEAN power with CH2 both inverted and non-inverted. If there is a big discrepancy between the two, then we should try to figure out why. And remember to get about 10 MATH cycles (5 Voltage cycles) on the display for the measurement (you have 12 in your scope shot above, and that is perfect).

Yes, I've got that also and now understand your reasoning  to use invert as a way to check for potential problems.

Thanks for your help and I do hope you will put up a YouTube video Tutorial to help other researcher with this needed information as I've checked a while back on YouTube and there is nothing on how to use a scope for power calculations.

Luc

[poynt99]

Luc.

Thank you.

I sincerely hope there was more than one lesson learned here by all. I certainly learned a few things.

And yes, I still hope to get your buy-in to the CH2 inversion requirement when making power measurements where polarity is an issue (like when claims of OU are involved). My video will hopefully explain clearly why the inversion is required.

So unless there is something you or anyone else would like to see with my setup, I guess this will be the only video I need to make (perhaps a short series of vids is appropriate) . Do let me know.

.99

[TinselKoala]

In that last set of scopeshots the "Ch2" baseline is not set exactly on the center graticule marker. It has a "slight" offset in the negative direction.

I know this seems very weird but I have encountered scopes that put the trace position offset into the math. If this scope behaves that way, then it might make a difference if you get both channel baselines exactly on the center graticule.

This would be a great time to check that, since that last set of traces is so close to being equalized but still has that little displacement of the Ch 2 baseline. Set up just exactly like that but use the vertical position control to get that baseline marker set right on the centerline, and then see if there is any difference between this set and that last set in terms of symmetry and absolute values for the Math.

[poynt99]

TK,

I just tried this, and shifting one trace up or down does make a difference in the computed MEAN power, as I noted back when I was doing the Ainslie tests, but there is still no change in the absolute value of the MEAN when I go from inverted to non-inverted on CH2.

Maybe the inverter amp (if they are using one) is just not matched that close on the TDS2004 as it is in other models.

[gotoluc]

In that last set of scopeshots the "Ch2" baseline is not set exactly on the center graticule marker. It has a "slight" offset in the negative direction.

I know this seems very weird but I have encountered scopes that put the trace position offset into the math. If this scope behaves that way, then it might make a difference if you get both channel baselines exactly on the center graticule.

This would be a great time to check that, since that last set of traces is so close to being equalized but still has that little displacement of the Ch 2 baseline. Set up just exactly like that but use the vertical position control to get that baseline marker set right on the centerline, and then see if there is any difference between this set and that last set in terms of symmetry and absolute values for the Math.

Good eye TK! ... my finger must of touch the adjustment at some point and didn't notice.

However, the math must detect that change as it make no difference to the Mean on this scope. I guess the scope needs to be re-calibrated somehow.

I could also use the average between Inverted and non Inverted if I need an accurate amount.

Luc

[poynt99]

Actually Luc, the more accurate is probably the non-inverted.

But I would hope the worst difference we would ever see would be 2% or so.

I don't know if there would be a calibration for this, but one could always contact Tektronix if necessary. My gut feeling is that the electronics which inverts the signal is what is causing the difference, and if that is the case, there should be a consistent percentage difference between the two settings.

One test you could perform would be to try CH3 and CH4 for the current probe (keep the voltage on CH1) and see what results you get with them.

[gotoluc]

Good idea!... I'll try channel 3 and 4 first.

I'll let you know the results

Luc

[MileHigh]

The big moral of the story here is to stop jumping to conclusions and to be honest about yourself and your limitations.  The other big moral of this story is that real debate is what is needed on the forums and the tyranny of the majority of the "believers" is not a good thing at all.  There are many winners around here.

[gotoluc]

That works better!... so it must be out of calibration on ch 1 & 2

Thanks for the good idea

Luc

[gotoluc]

The big moral of the story here is to stop jumping to conclusions and to be honest about yourself and your limitations.  The other big moral of this story is that real debate is what is needed on the forums and the tyranny of the majority of the "believers" is not a good thing at all.  There are many winners around here.

Hi there MileHigh,

I had to edit out something you said in your post. Lets not do any finger pointing please.

As for being honest about limitations, I feel quite confident that I did that.

By working together we can accomplish much and I think this is the message this forum needs.

Luc

[ramset]

Luc
Quote
By working together we can accomplish much and I think this is the message this forum needs.
--------------------------------------
You are right about working together as well as pointing fingers
and I am Glad you took the poynt at his word.[and the offer of help]
 
 
Good to see and a pleasure to read.
 
we are all the better for it.
 
Chet
 

[DilJalaay]

Hello all,


I want to test the Luc reactive power circuit by using inverter 220vac (600w) instead using grid supply.


By doing this i will be able to calculate the battery current(input) to out put on the reactive load.


As inverter has no Neutral or ground, can i connect my scope 1x/10x probe directly across the out put of the inverter, for seeing voltage phase etc?


I am going to use Phillips 3070/ 100Mhz.


Help from the experts are highly appreciated.


Regards,
D.J