Lately I've been trying to wrap my mind around the idea of phase-shift between current and voltage in reactive circuits.
I figure that I can understand it better if I can get pictures of what I'm working with. So I tried the following:
Ran 120VAC through a couple of series inductors. They're the biggest transformers I have, but their magnetizing load was still pitiful, something like 0.025A.
Put a current clamp probe around the hot wire feeding the inductors. Hooked the ground and test leads of an oscilloscope to each lead of the clamp probe. This gave me a waveform on my scope that showed the current draw to the inductors.
Hooked the second probe of my 'scope between the hot and neutral wires supplying the inductors. This gave me a voltage waveform.
Triggered both waveforms from a single input. I figured that if each the current and voltage waveforms were actually 90 degrees of out phase that this would cause them to be displayed that way.
That's not what happened. I each waveform to display amplitude accurately, and they were ever-so-slightly out of step, but definitely not 90 degrees. And adding or subtracting inductance from the circuit did not change phase shift (if that's what it was).
I was thinking maybe I needed to make all my grounds common? Scope probe grounds + power neutral + current clamp grounds? But I'm really weary about grounding out my current-clamp, it's the only one I have and with an output resolution of 1mV it seems sensitive enough to be damaged if I put even a couple of volts across the output leads.
Put you hands up and back away from the current probes!
John, there are easier ways to play with Eli the Ice man. Why not just put some non-wire wound resistors in there to sample across. We know that the voltage across any purely resistive element is in phase with the current flowing through it. I already know you've got this great transformer stashed away for isolation. If you pick your ref. scope ground, and have the R last to that ground, that voltage is phase related to your series circuit current. You should see your phase shifts relative to that at other points in your R/L/C ckt.
But since I'm macho because I work 48 Volts hot, why not step the voltage down and current limit it. Do your experiments with a dorky little wall transformer from the surplus pile. The theory still applies, shock hazard drops and you don't need X10 or X100 probes. Joe
[This message has been edited by JoeTestingEngr (edited 07-04-2006).]
Re: Phase Shift With O-Scopes#130498 07/05/0604:10 AM07/05/0604:10 AM
Remember too that the current and voltage will not be 90 degrees out of phase due to the inherent resistance of the coil. If the resistance is substantial compared to the inductive reactance, then the phase shift may well be much smaller.
Another way to measure the phase shift is to switch your 'scope to X-Y mode and use Lissajous patterns.
If you use a small resistance at the "low" end of the circuit as Joe suggests, you can feed the voltage from that (which represents current) to one input. Connect the other input directly to the supply, and adjust X and Y gain for equal height and width of the display.
The shape now indicates the phase shift, in phase being represented by a straight line at 45 degrees, this gradually opening out to an ellipse as the phase angle increases and becoming a circle at 90 degrees, as on the top line here:
You can calculate the actual phase shift by centering the display and measuring the ratio of the distance between the points at which the trace crosses the zero line and overall size, like this:
Xzero / Xtotal is then equal to the sine of the phase angle (you need to adjust for which quadrant you're in -- the display above is an angle between 90 and 180 degrees).
[This message has been edited by pauluk (edited 07-05-2006).]
Re: Phase Shift With O-Scopes#130499 07/15/0607:28 PM07/15/0607:28 PM
If I follow what ya'll are saying, simply put a resistor in series with my inductors and put my scope ground at the low end of my resistor. Then I can measure my voltage at any point in the circuit.
However, I don't have a current transformer for my scope, so I would have to keep the same setup with my current-clamp and a second scope probe. This seems like this would lead to the same problems I'm having already with the current and voltage being displayed, but no phase-shift being displayed.
How does adding the resistor eliminate this? What am I missing?
Re: Phase Shift With O-Scopes#130500 07/15/0608:04 PM07/15/0608:04 PM
John, Let's say you used a 1 ohm resistor as we mentioned. A one amp current in your circuit would be viewed as a 1 volt, in phase waveform. This becomes your trigger source and your current waveform. But the 1 ohm might be too much of an influence on your total circuit so the smaller the resistance, the better, for sampling. As long as you can decrease the Volts/Division enough to trigger on and see it, you're fine. 1 ohm of resistance wouldn't make much of a difference if you have large values of inductive or capacitive reactance. You would learn even more from your experiment if you substituted an audio generator for your voltage source and watched the effects while slowly sweeping the frequency. Joe
Re: Phase Shift With O-Scopes#130501 07/16/0603:56 AM07/16/0603:56 AM
I just noticed that I put Xr in the drawing instead of X sub L. I meant that your sampling resistance should be very much less than your inductive or capacitive reactance. Also note that I showed 2 separate points to place the channel 2 probe for observing relative phase, and not that the ground for channel 2 should be connected there. Both scope grounds, if you use both, need to be connected to the same point.
I don't think you described your scope. If you're given a choice between "chop" and "alternate", use the "chop" position. Joe