Hey there. I have a fairly straightforward question which I am not seeing an answer to when I search, but maybe I'm not phrasing it in the right way. Be that as it may...
If I connect an arduino via usb to a laptop on battery power I can reasonably assume it is not earth grounded, since everything is isolated. Cool. Laptop is an option for oscilloscope checks.
But if I connect to the PC in my workshop, that is earth-grounded, is that going to burst into flames when I connect an oscilloscope?
My assumption is that there isn't a direct tie between earth ground and reference ground on the arduino when powering over pins (via an isolated bench power supply,) even if usb is connected, but it's a fairly unsafe assumption.
Basically, is there some isolation I'm overlooking, or should I consider any connection to an earth grounded PC as potentially explosive and simply move my workflow to a battery powered laptop whenever an oscilloscope is involved?
The oscilloscope ground will be counted via its mains cable to your PC ground, and back via USB to your Arduino 0V. All the same. No problems.
The only thing to be aware of is there will be an earth loop around via mains safety earth. I've never found this to be a problem.
However, if you are working on something that, for whatever reason, cannot be connected to mains earth, then be aware that your PC and your oscilloscope both connect to mains earth.
It's my understanding that if I accidentally touch the wrong thing on the arduino The oscilloscope may potentially short to earth-ground. That's not a big deal with respect to the 3.3v/5v rail at a fraction of an amp on the arduino, but I don't know exactly what the relationship between usb and my computer's 1000w power supply is.
But my general rule of thumb with anything that is earth grounded is to not touch it with the scope, in case I mistake a pin/pad for positive when it is in fact neutral/ground, or in general just connect something that ought not to have been connected.
I guess maybe this is paranoid, but no one ever blew up an oscilloscope assuming the worst, and I bet a few have assuming the best.
Again, if this is overly paranoid I would just like to know why (is there something inherent to usb that makes a short of this kind relatively harmless? 5v 2a seems like it could fry a decent amount of small components, if not the scope, necessarily.)
The oscilloscope chassis/GND is connected to earth.
PC GND is connected to earth.
PC 5v negative is connected to its chassis which is connected to earth.
We connect oscilloscope GND to Arduino 5v negative, all is okay.
We connect oscilloscope GND to Arduino 5v positive, we can blow up components in the Arduino and/or PC, the oscilloscope GND lead might, i.e. might but probably won’t smoke.
If you must connect the oscilloscope GND lead to Arduino voltages, put a AC plug adaptor (with GND removed) on the scope’s AC plug.
Note the scope chassis is now floating, be careful !
If you have a scope with more than one channel, and need to measure voltages referenced above GND, use the scope’s differential voltage measurement technique.
As Larry pointed out, your location on the planet will have a significant influence on the response you're getting. Your profile gives no hints, so where are you?
You have to be careful with the 'scope ground, not the "input" probe. But you have to careful with almost any "wrong connection" so there's nothing unusual about a 'scope.
I think the USB spec that requires USB ports to be short-protected. If you short it, you'll usually get an "over current" message and you'll have to re-start your computer.
But 1 or 2A is enough to fry most chips.
Most laptop power supplies are ground-isolated, and of course if you are running from batteries there is no earth ground. But if you have a printer (or something) connected, it may be earth-grounded through that. Or, your Arduino may have an earth-ground so that would earth-ground your laptop when plugged-in.
Most test equipment is electrically rugged and I've NEVER heard of someone frying a 'scope.(1)
You CAN blow the current fuse on a DMM when measuring current or by accidently measuring current when you're trying to measure voltage. That's pretty common. Meters also have a maximum voltage (usually 600V or so) but you won't fry it with a couple-hundred volts when it's on a low-voltage scale, or when it's set to measure resistance, etc.
(1) I'm talking about "real" oscilloscopes... I don't completely trust those cheap little boards/gizmos that you plug into your computer. (I've never used one.)
This was the answer I was looking for. I'm just going to use the laptop if I need to measure something. Yes, it would be my fault if I connect the wrong lead to the wrong thing, but if it's a diode on half an amp the world doesn't end. I'd rather not risk it with hundreds or thousands of dollars of hardware.
On all my PC's the motherboard has mounting points that connect the 0V PSU line to the PC case. Which of course is attached to mains earth. Am I misunderstanding your comment here?
This was my fault, as an American, forgetting that the rest of the world exists.
Depending on where you are the power grid, and thus what neutral/ground means, varies. So the answer to the question, "is my PC neutral connected to earth ground" is dependent on region. In the US, you can pretty much expect a pc to tie neutral to ground, but unfortunately "region" is the best I can say for anywhere else, because Europe doesn't necessarily follow one pattern, the Americas don't follow one pattern (North America is mostly the same,) and so on.
So, everyone has to figure this out for their own specific power grid; which is very unhelpful, but no information is infinitely more helpful than bad information.
A friend confirms that also for his motherboards, here in Germany. Another one found plastic knobs holding the motherboard in the case, with no screws or other contact to the case.
I would say “highly unlikely”. Any PC I have ever worked on has had an IEC-320-C15 receptacle for the power supply. That connector has a neutral and a ground wire. The ground wire will be bonded to the metal case of the power supply. The power supply output common wires will most likely also be bonded to the metal case. The only place the neutral and earth ground are connected together is in the load center of the building where the utility provides two service conductors and a grounded service conductor. Also known as two “hots” and a neutral. Nowhere else in the building are the neutral and ground conductors allowed to be connected together. UL would never approve a design with that done.
I know this seems like a minor point but it is an important one.
You will also find that the common output of the computer power supply is only connected to earth ground at a single point to prevent ground loops.
