# Floating Scope measurements

#### quixote

Joined Jan 15, 2007
12
I found no precise answers in previous threads.

I want to measure the V-I time lag (and hence power factor) exhibited by a material that can be modelled as a capacitance. This is connected across 110V AC at 50Hz. From this my intention is to calculate (and build) the compensating inductance.

1. The capacitor is powred by an AC-AC convertor (240 V in, 110V out) showing cos ( phi) = 0.73. Does this indicate the worst case power factor that the convertor can correct?

2. Must I use RMS values or peak values of voltage across the capacitor in order to calculate the apparent power?

3. I am measuring the voltage across the capacitor using the pseudo differential (A-B) technique. In order to do this, I have connected the references of 2 scope probes together (not to ground). Is this correct?

4. I thought that probe references are connected internally to the scope earth? Is this not so?

(Previously when I tried to float the scope the circuit breaker did trip. I now realise floating a scope is not the right thing to do).

Any advice much appreciated...

#### timrobbins

Joined Aug 29, 2009
318
Why do you want to measure the phase angle difference specifically? Or are you aiming to try to compensate for the capacitance, to achieve a unity power factor correction?

Could you just measure the capacitance itself?

Why do you want to use mains level AC voltages to do the measurement?

Do you only have an oscilloscope to do the measurement?

Ciao, Tim

#### quixote

Joined Jan 15, 2007
12
Hi Tim

Why do you want to measure the phase angle difference specifically? Or are you aiming to try to compensate for the capacitance, to achieve a unity power factor correction?
-- Correct

Could you just measure the capacitance itself?
-- In this case it is a straightforward capacitance but my client may put other impedances in parallel (compact fluorescent lamps, etc) so I want a generic 'test bed' from which to calculate the corresponding compensation impedances (without having to increase budget for active PFC circuits).

Why do you want to use mains level AC voltages to do the measurement?
-- The material in question operates at 110Vac (50 Hz).

Do you only have an oscilloscope to do the measurement?
-- At the moment I have a 35MHz Hameg analogue scope with two 10x passive probes. What other instruments would you recommend?

Joined Jul 7, 2009
1,583
To get the power waveform, you need to measure the voltage across the load (easy) and the current through the load, then multiply them together. Measuring the current is the harder measurement.

The easiest tool to measure the current is to use a wideband current probe. Most folks don't have such a device and they usually aren't cheap. It depends on how nonsinusoidal the current waveform is. For a line-frequency current waveform that wasn't too far from a sinusoid, I'd use my HP 428 (a clamp-on ammeter) and just clip it to the wire carrying the current. But its bandwidth is only around 400 Hz.

Another way is to use a current transformer. These can be purchased or made. There's lots of information on them on the web or search AAC.

Probably the "easiest" way is to put a resistor in the circuit and measure the AC voltage across the resistor; if the resistor is noninductive, this is proportional to the current. But this is a non-safe thing for beginners to do in line-voltage circuits, especially if they put the resistor on the high side of the circuit. The first mistake beginners sometimes make is to connect the probe to one side and the probe's ground to the other. This usually leads to a short and some sparks. The best thing to do is to use a differential amplifier to measure the voltage across the resistor. Unfortunately, you can spend hundreds of dollars to buy an instrumentation version if you buy a new one. The old Tektronix AM502 can be used for this and they can be found used for not too much money. Some older Tek scopes came with differential amplifier plug-ins.

If you don't have a differential amplifier, use your two channel scope with the two probes to measure the voltages on either side of the current resistor (leave the ground wires unconnected) and subtract the readings. This will give you the current waveform. Then make another measurement to get the voltage waveform. If all you're interested in is the phase angle between the current and voltage and they're essentially sinusoidal, then this is all you need. Just make sure you trigger things properly to get the correct phase angle. Here, that would mean use a line trigger; do not trigger from the signal.

Years ago, I occasionally floated my 1970's scope from the power line to make such measurements, but I know what I'm doing and I was very careful when doing it (I don't do such things anymore because I finally spent a few hundred dollars and got a decent differential amplifier). I don't recommend this for a newbie. You can improve safety a bit by using an isolation transformer.

Modern digital scopes can multiply the waveforms together for you to get the instantaneous power waveform, which is quite handy. Otherwise, if you have an analog scope, you have to manually digitize the waveform and do the multiplication yourself (just write a quickie python script to do the calculation for you). This, obviously, is only needed if you want the instantaneous power waveform.

#### timrobbins

Joined Aug 29, 2009
318
Hire a 3-phase power analyser (voltage probes and current clamps) that is designed to go on mains and will clearly show you the PF and real and reactive current levels, and even neutral currents, and harmonics if the loads aren't all linear (ie. can't be compensated by just adding simple reactive power when you're wondering why you can't achieve PF=1). They are designed for this task. You can then check out various points in your distribution, and do before and after tests.

Your technique is likely to kill you before you get the results you want.

Ciao, Tim

#### russ_hensel

Joined Jan 11, 2009
825
The kill a watt ( or similar name ) is a 110 v ac single phase power meter that measures the power and power factor. About 30 bucks. The price will not kill, and direct measurements may.

#### thatoneguy

Joined Feb 19, 2009
6,359
The kill a watt ( or similar name ) is a 110 v ac single phase power meter that measures the power and power factor. About 30 bucks. The price will not kill, and direct measurements may.
These are extremely useful for their price.

If you are working with 3 phase, not so useful, and you'll need a power line analyzer to get the information.

To measure voltage drop with a 2 channel scope (non-floating), connect channel A to one side of the device (such as resistor/capacitor/inductor), Channel B to the other side of the device. Set Channel B to INVERT and set the ADD function. This will then show the drop across the resistor as a single trace.

I'd strongly suggest an electrician over trying to decipher what's going on with only a scope and a meter, though.

#### quixote

Joined Jan 15, 2007
12
Thanks to All, especially for the safety tips. First chance I get, I will invest in a power meter.

I am indeed performing the pseudo-diff method as you all indicate, but with great care. A couple of my original questions still remain though:

1. My AC-AC convertor shows cos ( phi) = 0.73. Does this indicate the worst case power factor that the convertor can correct?

2. I have connected the references of my scope probes together (not to ground). If I do not connect them together against what will the scope reference the measured voltage?

-- regards

#### thatoneguy

Joined Feb 19, 2009
6,359
Thanks to All, especially for the safety tips. First chance I get, I will invest in a power meter.

I am indeed performing the pseudo-diff method as you all indicate, but with great care. A couple of my original questions still remain though:

1. My AC-AC convertor shows cos ( phi) = 0.73. Does this indicate the worst case power factor that the convertor can correct?

2. I have connected the references of my scope probes together (not to ground). If I do not connect them together against what will the scope reference the measured voltage?

-- regards
1) Depends on your converter, make/model, etc. Have a manual for it?

2) I usually connect both probe ground leads to the ground of the circuit I'm measuring, which is usually isolated or the same ground the scope uses.

#### n1ist

Joined Mar 8, 2009
189
On many scopes, the probe grounds are connected to the case and safety ground. Floating the scope with an isolation transformer and ground lifter would work, but the entire scope chassis will be hot, and will be a safety hazard.
/mike