Differential probe build

[Swend]

I'd say the opposite: If they don't have a common reference then there is nothing preventing them from having a common mode potential difference.

Yes there is, it's called a dielectric, in the case of the diff-probe it is the 4-6Mohm input impedance of the amplifier. But I agree it's very easy to make a sock-up (replace the 's' with a 'c') and generate smoke, therefore I will put a warning on the diff-probe "Think twice, measure once".

And if they're not galvancially isolated but you call them 'floating' that implies they do have a common mode potential difference.

Then we are discussing semantics and not electronics. For future ref. when I say "isolated" or "floating" I mean "galvancially isolated".

Your circuit can't measure a signal that has greater than ~9V between your differential scope probe ground and the oscilliscope ground.

I'm ready to believe you, but you did not present any scientific founded argument/theory/law/convention to support your claim.

In my mind that's hardly a good differential probe and you havn't made a convincing case that this is actually ok in your application. Particularly given that you want to measure high voltages.

If you are right that I can't, then you are also right about the above.

in relation to your scope ground (which is your differential probe circuit ground too).

A-ha, maybe we have something here we can clarify mutually. Can you point out on my schematic where the scope probe ground ("ground" as in a reference, not the planet's ground/earth) is connected to U1 ground? Because I can't see it?

 

[asdf44]

Your 9V supplies are +/-9V in relation to "gnd".

"gnd" is tied to scope ground via J3 pin 2.

Your "Scope Probe" input J1 pin 2 is connected to U1 IN- via a 470ohm resistor.

U1 AD8130 as an absolute maximum rating of "−VS − 0.3 V to +VS + 0.3 V " on its input pins

So....that means J1 pin 2 needs to remain within +/-9.3V of "gnd" or scope ground.

 

[Swend]

Your 9V supplies are +/-9V in relation to "gnd".

Agree

"gnd" is tied to scope ground via J3 pin 2.

Agree

Your "Scope Probe" input J1 pin 2 is connected to U1 IN- via a 470ohm resistor.

Agree

U1 AD8130 as an absolute maximum rating of "−VS − 0.3 V to +VS + 0.3 V " on its input pins

Agree

So....that means J1 pin 2 needs to remain within +/-9.3V of "gnd" or scope ground.

Agree

But where is "Scope Probe" input J1 pin 2 (or pin 1 for that matter) connected to "gnd" of the circuit? This is what I'm asking about?

 

[asdf44]

It's not but J1 pin 2 is connected to U1's input. And that input can't be more than 9.3V from "gnd".

 

[Swend]

It's not but J1 pin 2 is connected to U1's input. And that input can't be more than 9.3V from "gnd".

Yes that is correct. Now if I take J1 pin 2 and connect it to a the (-) of a 48V battery, and connect J1 pin 1 to the (+) of the same battery, what is the resulting voltage potential between U1 pin 1 and gnd? and the resulting voltage potential between U1 pin 8 and gnd? And can you please explain to me in technical terms why it is so?

 

[asdf44]

Yes that is correct. Now if I take J1 pin 2 and connect it to a the (-) of a 48V battery, and connect J1 pin 1 to the (+) of the same battery, what is the resulting voltage potential between U1 pin 1 and gnd? and the resulting voltage potential between U1 pin 8 and gnd? And can you please explain to me in technical terms why it is so?

If the circuit is galvancially isolated which it would be in the case of a battery powered circuit with no other references then you're good. The non-infinite impediance of your input circuit and the clamping diodes inside U1 will bring J1 pin 2 (battery -) to within +/-9.3V of 'gnd'.

But you'd also be just fine connecting (-) to your scope's ground in this case.

So we have an instance where your circuit is ok but isn't needed. Give an example where you think you need the differential probe functionality.

 

[Swend]

If the circuit is galvancially isolated which it would be in the case of a battery powered circuit with no other references then you're good. The non-infinite impediance of your input circuit and the clamping diodes inside U1 will bring J1 pin 2 (battery -) to within +/-9.3V of 'gnd'.

I will try to answer my own question, you just tell me if you are in disagreement. In the case of the connected 48V battery
voltage potential between U1 pin 1 and gnd = 0V
voltage potential between U1 pin 2 and gnd = 0V
voltage potential between U1 pin 1 and (-) battery = 5V
voltage potential between U1 pin 2 and (-) battery = 0V
voltage potential between U1 pin 1 and 2 = 5V

So you see, no common mode voltages. And max differential mode voltage (voltage difference between U1 pin 1 and 2) is 5V because of the zener, which just brings it within AD8130's maximum capability. The result is the same if I connect battery (-) to earth (as in planet), still no common references and therefore no common mode voltages. But if I connect scope ground to earth at the same time the battery is connect to earth, then we have the smoke scenario you are talking about.

But you'd also be just fine connecting (-) to your scope's ground in this case.

Yes indeed.

So we have an instance where your circuit is ok but isn't needed. Give an example where you think you need the differential probe functionality.

Imagine the 48V battery connected to two resistors in series R1 and R2 where R2 is connected to battery (-), now take two more resistors R3 and R4 in series where R4 is connected to battery (-), Now I wish to measure the voltage potential across R1 with one probe, and simultaneously measure the voltage potential across R4 with a second probe but on the same scope with interconnected grounds.

If we are finished with this subject now can you please help me with the filter calculation?

 

[asdf44]

I'll modify the example a bit: Your 48V battery is 4 12V batteries in series and you want to measure each with a differential probe. We'll call them battery 0/12/24/36/48V and naturally you want to tie your 4 differential probe ground inputs to battery 0/12/24/36V.

Can you do this with your circuit? Answer: No.

Because your differential scope probe inputs are clamped to within +/-9V of scope ground I know they're clamped to within 18V of each other under the luckiest circumstances. Hook one up to battery 0 and one to battery 36 and boom!

And whether things are earthed or not doesn't change that (it could only make it worse).


I'll try to be more helpful: The way to do this is to have the same selectable 1M circuit on J2 pin 2 as you have on J2 pin 1. Now the 1M impedance lets J2 pin 2 safely float far away from "gnd". It's also a properly symmetric differential circuit.


The variable you're asking about is presumably the switching frequency of your charge pump supply. No magic there. Just pick a cutoff well below its switching frequency with values that are practical.

 

[Swend]

Because your differential scope probe inputs are clamped to within +/-9V of scope ground

It's like we are going round in circles now, either you want to discuss it and consider the possibility that you could be wrong - or if you just want me to agree with you without you presenting any valid arguments, we can also do that.

So for the last time on my part: In my circuit the scope probe inputs J1 and scope ground J3 are galvanically isolated by U1. That means I can connect my probes anywhere I want in a circuit as long as there is no common reference, including your modified example.

Can I do this with my circuit? Answer: Yes.

 

[Swend]

OK, this is the new schematic and PCB layout, I still need to calculate the filter though but I made room for the components on the PCB.



I made the PCB so it will fit a HAMMOND 1591ASGY enclosure in ABS plastic, then I will line it with alu-foil.


Any objections?

 

[asdf44]

It's like we are going round in circles now, either you want to discuss it and consider the possibility that you could be wrong - or if you just want me to agree with you without you presenting any valid arguments, we can also do that.

So for the last time on my part: In my circuit the scope probe inputs J1 and scope ground J3 are galvanically isolated by U1. That means I can connect my probes anywhere I want in a circuit as long as there is no common reference, including your modified example.

Can I do this with my circuit? Answer: Yes.

The AD8130 does not provide galvanic isolation!

If you want galvanic isolation you need an 'isolation amplifier' and an isolated supply to power it.



If another forum poster jumped in it would be welcome at this point.

 

[Swend]

The AD8130 does not provide galvanic isolation!

If you want galvanic isolation you need an 'isolation amplifier' and an isolated supply to power it.

OK, you are completely right and I don't know what I'm doing, my circuit sucks and everything will blow up even before I put power on. I'm not in doubt anymore, you are absolutely right.

If you have any doubts, may i suggest you to create a new post yourself with your question, but since you also don't have any doubts - that won't be necessary since we are in total agreement, you are right and I was completely wrong, but now I have seen the light and the purest truth.

 

[asdf44]

Even galvancially isolated amplifiers (like ISO124) have peak voltage ratings (1500Vrms for the ISO124).

Can you show in the datasheet where you think AD8130 is providing isolation and reference what the limit of that isolation is?

 

[KlausST]

Hi,

when I follow your signal backwards:
* you have the scope connected to EARTH GND.
* you have a USB charger
* you have some floating signal nodes

EARTH GND and charger:
The USB charger will cause some currents form the 5V lines to EARTH GND (at least via LINE and NEUTRAL and charger internal capacitors between primary and secondary)
The impedance is not infinite. And there will be currents caused by the high switching frequency of the USB charger.
It will cause a ground loop. Not low ohmic. But rather low impedance for high frequencies.

Try this: use a DVM, choose AC voltage. Connect one line to EARTH GND (or water pipe...) and the other to the USB charger output.
USB charger connected with mains. Which ouput pin you use is not important.
I assume the DVM shows about half of the mains voltage.

Your circuit needs to withstand this. Relatively high voltage pulses when you connect the devices. ESD, machine model. You need filters and protection circuitry for a reliable operation.
And these currents must not cause noise to your measurement signals. Mainly a PCB layout problem.

*******
EARTH GND to floating signals:
Even if the nodes are floating - or better: because of floating signals - there will be (a of voltage) between floting nodes and EARTH GND.
Your AD8130 circuit can´t handle this. It can process floating signals only as long as they are within about +/-7.5V.
Your circuit has to ensure this.
Thus you (sadly) need a (high ohmic) relation to your GND.
A voltage divider with reference to GND on each input is one possible option.

But to calculate this you need to specify the voltage at the floating probes. What is the highest expectable voltage on a floating node w.r.t. EARTH GND?
Not the ESD or erroneous high peaks, but the expectable voltage during normal operation.

Klaus

 

[Swend]

* you have the scope connected to EARTH GND.

The scope (as in stand alone instrument) is NOT connected to EARTH (as in planet earth) and I'm using a isolation transformer.

EARTH GND and charger:

There will be a label on the enclosure of the diff-probe which reads "ONLY CHARGE WHEN PROBE AND SCOPE ARE DISCONNECTED!"

Even if the nodes are floating - or better: because of floating signals - there will be (a of voltage) between floting nodes and EARTH GND.

Please define the term "floating"

 

[KlausST]

Hi,

floating:
a floating node usually belongs to a floating circuit. This means the whole circuit is isolated.
But "isolated" usually means there is enough impedance that the current is limited and nobody gets hurt.
But there never is an ideal isolation. There always will be the possibility for current. Either resistive, or electrostatic or capacitive...

Imagine. You have isolating shoes, walking on a carpet.
The shoes prevent DC current to flow. (at least no reasonable current)
The carpet also is considered to be a good isolator.
But due to electrostatic effects you may get charged with high voltage. Very high voltage. Several kilovolts.
And you become a big (in size, not in capacitance) capacitor. You are one electrode, the air is the dielectric, the surronding (slightly) conductive environment (concrete, steel, water pipes..) is the other electrode.

When you come close to EARTH GND there will be a spark and noticable current. Enough energy to destroy electronic devices.

****
Thus even when the scope is not connected you need to consider ESD.
The AD8130 datasheet shows a couple of MOhms in common mode. In best case these resistance is to GND.
Then a tiny current of 1.25uA (calculated with 6MOhms) will make the inputs of the amplifier to leave the allowed area of +/-7.5V.
No need for a wanted connection. Even unwanted current flow will cause trouble.

****
Use the DVM and connect one wire to earth. AC voltage mode. Then do some tests and measure voltage at some electronic devices or other metals around that are considered to be "floating" or "isolated" with respect to EARTH GND.
What´s the voltage of your isolated scope?

There will be a label on the enclosure of the diff-probe which reads "ONLY CHARGE WHEN PROBE AND SCOPE ARE DISCONNECTED!"

Not all customer´s will read and care about it.

The scope (as in stand alone instrument) is NOT connected to EARTH (as in planet earth) and I'm using a isolation transformer.

Your scope, yes. Others not.
But even "isolated" ones will cause some current.

Klaus

 

[asdf44]

Here is a simplified circuit modeling the path between the probe grounds of two of your circuits connected to a single scope.

V5 is like the battery in the example I gave above. It has no reference to scope ground (or earth) other than its connection to your circuit. This shows your circuit clamping voltage and drawing current (bad) when more than 18V is applied between two scope probe input grounds.

 

[Swend]

Here is a simplified circuit modeling the path between the probe grounds of two of your circuits connected to a single scope.

V5 is like the battery in the example I gave above. It has no reference to scope ground (or earth) other than its connection to your circuit. This shows your circuit clamping voltage and drawing current (bad) when more than 18V is applied between two scope probe input grounds.

Well at least you made a schematic.

If your D1 and D2 are supposed to be my zeners D1 and D2

1. Then they are not connected like that to U1 pin8 in my circuit.

2. Then show me in my circuit where D1 is connected to +9V (V3 in your circuit)

3. Then show me in my circuit where D2 is connected to -9V (V4 in your circuit)

 

[asdf44]

They're not your zeners they're the ESD protection diodes inside the AD8130 that result in the "Input Voltage (Any Input) −VS − 0.3 V to +VS + 0.3 V" absolute maximum rating (with the diodes having a 0.3V drop)

Note that the datasheet doesn't explicitly diagram them but I assure you they're there. And even if they're not its even worse: the inputs would have NO protection. With the diodes the you can assume the input pins will survive 5-20mA or or so.

 

[Swend]

But there never is an ideal isolation. There always will be the possibility for current. Either resistive, or electrostatic or capacitive...

Yes I'm aware of that fact.

When you come close to EARTH GND there will be a spark and noticable current. Enough energy to destroy electronic devices.

Yes that's how ESD works.

Use the DVM and connect one wire to earth. AC voltage mode. Then do some tests and measure voltage at some electronic devices or other metals around that are considered to be "floating" or "isolated" with respect to EARTH GND.
What´s the voltage of your isolated scope?

I doubt a DVM is the correct instrument to measure isolation. But anyway I did as you said, the DVM says my scope is +80V in relation to earth, and then I took a 1.5V AA battery the DVM says it's +33V in relation to earth, then I measured between earth (in the wall socket) and a water pipe and I have 220V in relation to.... well, earth and earth. And finally I measured my cat, it was +11V in relation to earth.

Not all customer´s will read and care about it.

I am the only customer and I care extremely much, because if I blow up anything I will have to repair it myself.

But even "isolated" ones will cause some current.

Yes I'm aware of that fact. Everything is connected in the universe.