PSU heatsinks electrified?

[ashiekh]

First an advisory before you read this post: NEVER EVER plug an oscilloscope up to a switch mode power supply unless you REALLY know what you are doing.
You WILL short out the power supply because the scope is almost always* (ref 0:48 in the video) referenced to ground, but the power supply is not, see this video for more in-depth details: EEVblog #279.

Now I learn that if one has a dual channel scope one can use the probe from each channel (add the channels and invert one) to effectively have things floating and avoid this issue (saves having to buy a differential probe).

[asf14tomcat]

for job use differential probe. or some scope have built in isolate input like tektronix TPS series or some hioki, picoscope scope. really expensive than normal scope ( with same features)
just a hobby. cut off ground cable in your scope power cord. always use 10x and up passive probe. cheapest way but dangerous for young player

[ashiekh]

Or just use the dual channel method and stay safe.

NEVER cut the ground wire.

[Per Hansson]

Now I learn that if one has a dual channel scope one can use the probe from each channel (add the channels and invert one) to effectively have things floating and avoid this issue (saves having to buy a differential probe).

It's not safe.
Show me a general scope with anything but a CAT-I rating.
You are measuring a circuit connected to mains here.
The scope needs at the very minimum a CAT-II rating for such measurements.
Either use the proper equipment (a differential probe) or just don't do it.
There is no point for the average person to connect a scope up to mains and kill himself in the process!

[ashiekh]

I thought most scopes were good to 300V; the one I have access to is good to 400V

[Per Hansson]

I stand by what I have written in my previous post and see no reason to change any recommendations within.
A standard oscilloscope is designed for CAT-I environments, the primary side of a PSU is a CAT-II environment.

[ashiekh]

So you are saying that most 'scopes shouldn't be used for mains voltages?

Now it is true that 240V AC is 340V peak, but in the States we are half this.

[Per Hansson]

So you are saying that most 'scopes shouldn't be used for mains voltages?

Yes.

https://en.wikipedia.org/wiki/Measurement_category

[ashiekh]

Many thanks, this has been an education; the reason why I visit this forum.

I would imagine most oscilloscopes are CAT I rated exactly because of the grounding issue, which would make them unsuited for direct mains measurements; the dual channel trick removes this restriction.

[gdjacobs]

Very much appreciated; perhaps reason enough to stick with a battery powered (floating) digital multi-meter.

Isolation transformers are a very useful tool.

I like those Fluke oscilloscopes,

Scopemeters are very nice if pricey.

not just because they can be run on their own internal batteries (and so have no ground) but also because one can download the screen picture into a computer.

Capture and data connectivity is common now, but it's been available for many years.

But I agree that these are lethal voltages and the point of this discussion is not to have people to take risks.

1) It never hurts to probe for voltages before working bare hand
2) De-energize whenever possible
3) If in doubt, one hand in the pocket, wear isolation tested footwear
4) RCD protected circuits can save lives
5) Always be aware of arc flash risks