Measuring floating voltages with oscilloscope

[AndrewT]

Tektonix make it very clear that disconnecting the PE/earth connection of the scope is potentially very dangerous.

floating measurements have been made by knowingly defeating the built-in safety ground features of oscilloscopes or measurement instruments in various manners. THIS IS AN UNSAFE AND DANGEROUS PRACTICE AND SHOULD NEVER BE DONE!

What about floating the DUT with the scope properly earthed?

[cliffforrest]

Originally Posted by DF96

It is threads like this which are good arguments for a newbie-tag: I can't measure 500kHz with a 20MHz scope (but by the way I haven't yet read the manual so I don't know how 10:1 probes work, but I can't tell you that because I don't know that I don't know it).

^n

[AndrewT]

Originally Posted by jitter

................100 Vp ....... (= 200 Vpp or 141 Vrms).

70Vac ~= 100Vpk = 200Vpp

[jitter]

You're right, I scr*wed up, Vrms = 1/2*sqrt2*Vp (not ..*Vpp).

[akis]

Originally Posted by AndrewT

Tektonix make it very clear that disconnecting the PE/earth connection of the scope is potentially very dangerous.[/COLOR]

What about floating the DUT with the scope properly earthed?

It is not that simple (for me).

Here is a most basic experiment. The oscillator is battery powered. On the attached schematic you'd think that the voltage between a-b and b-c would be equal and half of the total voltage, that across a-c.

The scope does not measure half because it has a capacitance and a resistance, so we expect a bit less than half.

However how can we explain that the scope measures Vab to be wildly different to Vba (ie turn the terminals around) ? Similarly Vbc is wildly different to Vcb. The correct reading (something a bit less than half voltage) is taken in a particular direction only.

I cannot explain this, and at the risk of being called a complete newbie, I am asking here.

[cliffforrest]

Probably because it really isn't floating? (at 500Khz)

Much lower impedance to ground at one end (ground as in what the scope is connected to.)

Why are you making life so difficult for yourself, AKIS?

Trying to measure voltage with source impedances about 1MR is FRAUGHT with difficulties, even if you really understand your test equipment and measurement
techniques.

Try it with 1K resistors. If you get the right answers your problem is as above.

[DF96]

Nothing working at AC much above audio frequencies is 'floating', because of stray capacitance. Welcome to the world of RF! Nothing is exactly as it appears on the circuit diagram.

Please don't be offended by me implying that you are a newbie. This is simply a statement of fact, and may help people give you more appropriate advice.

[akis]

I am examining this because the differential circuit was going to have such a resistive network at its front end.

There may be something wrong with my scope / probes. I will buy a new scope anyway. In the meantime if anyone would like to try that little test above, I would be grateful to know if you are also getting weird readings when you turn the terminals over. By "weird" I mean reading just 5% of the voltage when you reverse the terminals.

[DF96]

My money is on 'operator error', which includes misinterpreting results, ignoring parasitic components, and using unofficial 'grounds' (including the operator's body). We have tried to explain what is happening, but you seem to desire a more esoteric explanation.

[Loudthud]

Originally Posted by Loudthud

Checkout the OPA117 by Burr Brown. A differential amp with a gain of one, 300V common mode range on the input when used with +/- 15V rails. Not totally isolated but fairly high impedance.

Never float the ground on your scope!

OOPs! That should have been INA117P. TI now owns Burr Brown. I was just going from memory. Common mode range on the input is just +/- 200V. If the input impedance is too low, use a differential opamp buffer operating off of floating power supplys (or batteries. Then use the INA117 to sum the outputs and reference the output to ground.

Here's the marketing blurb: Amplifiers and Linear - Difference Amplifier - INA117 - TI.com

Here's the data sheet: http://www.ti.com/lit/ds/symlink/ina117.pdf