Oscilloscope reading voltage with ONLY positive connected? what the....

[K5_489]

Running a cheap Amazon special scope - Hantek DSO2D10.  Still a very new newbie with using scopes...was using it to verify voltages on a ham radio that has failed, specifically checking input voltage from the Astron VS-35M linear power supply and voltage through the accessory connector on the radio. 

Using the lead that is just two alligator clips, not the typical scope probe.  Connected red clip to positive, black clip to negative on the power lead to the radio, reading 13.1 volts (ruh roh! That should be 13.8..and the voltage knob is cranked all the way up...first problem!).  Disconnected just the black clip, red clip still connected, still reading 13.1 volts.  Wait...what?  Tried this multiple times on several different lines coming out of the power supply, and found all I need is the red clip connected to see 13.1 volts.  Tried measuring a battery - 11.9V, and either clip disconnected voltage drops to zero (well, quickly fluctuating between 30 and 50mv, but that I expected).  Back to the power supply, and all I need is the red clip connected to see the 13 volts.  Switched over to a 1x probe, and same thing - just the probe needs connected without the ground clip.  Switch to 10X, and voltage drops to 1.3V. 

Is this because both the Astron and scope are powered via AC, and the scope is reading ground through that connected loop?  Or is this indicative of another problem where I'm getting a ground connection where there shouldn't be one?

[ArdWar]

Its normal and expected in this situation. VS-35M has its negative output tied to mains PE. Most desktop oscilloscope input ground are also tied to mains PE, therefore its signals are also referenced to it. There goes your signal loop.

Unfortunately this is also how people sometimes end up blowing their scope when they inadvertently short their devices through scope input ground.

[K5_489]

Unfortunately this is also how people sometimes end up blowing their scope when they inadvertently short their devices through scope input ground.

Is there something I can do to protect against doing this, or is it just a "watch where you're poking and don't be stupid...don't connect power to ground" kind of thing?

[jonpaul]

DVMs can recall a DC reading after the source is removed by an internal capacitor.

The Chinese scope may be similar.

Try a VOM and not scope to measure DCV

Jon

[BennoG]

Is there something I can do to protect against doing this, or is it just a "watch where you're poking and don't be stupid...don't connect power to ground" kind of thing?

You can use a scope with floating inputs (expensive) ore use a differential probe.

I have a scope with differential inputs but end up in 99% of the times binding the negative input to the ground. So not worth the extra cost.
Differential probe can be of use special if you are gonna measure on mains voltages or above 50V voltages, to protect yourselves and the equipment you are using.

P.S. 13.1V for a HAM radio set is no problem (it should even work down to 11.5V or even lower)

Benno

[K5_489]

Try a VOM and not scope to measure DCV

Jon

I am of the firm belief that one must use their equipment in order to become proficient at using their equipment.  Even simple tasks help to achieve that goal. 

Plus it was just easier to drop a test lead down than wander out to the truck where my Fluke meter was stashed (yes, I really need to pick up another meter...and I'm trying to justify the cost of a nice benchtop unit, lol)  figured I only needed to know if the input voltage was roughly 13.8V, and if it was dropping more than a tenth or two with a load on, so I didn't need super tight precision. 

P.S. 13.1V for a HAM radio set is no problem (it should even work down to 11.5V or even lower)

Benno

"Work" yes.  "Achieve full power output" No.  At 13.1V out of the power supply, it was sagging to 12V upon key up and only getting to roughly 30 watts output on the antenna feed.  At 14V on key up, the radio was up to 80+ watts out.  My power meter is attached to a 40-some year old Wavetek service monitor, so I don't trust that the actual numbers are 100% accurate (it's a 100 watt rated Icom 706MKIIG), but I'm still clearly getting a considerable wattage drop on the lower voltage. 

I know the difference between 30W and 80W going to the antenna is negligible in the real world, but this is a radio that seemingly worked fine one day and went deaf and dumb the next for no apparent reason.  The antenna system checks out on a different radio. 

So now I'm just going through all the checks, and verifying the radio is doing what it _should_ be doing on the bench, AND using this as an opportunity to learn how to use this equipment I bought.  It's also become clear that I need to look at the wiring system coming off the power supply, because even a 100 watt radio shouldn't be pulling this Astron down a full volt. 

[perieanuo]

eevblog have some must-watch videos on scope grounding, transformer voltage separation and so on. watch them 3 times before using scopes. otherwise at some point you can regret

[tggzzz]

Unfortunately this is also how people sometimes end up blowing their scope when they inadvertently short their devices through scope input ground.

Is there something I can do to protect against doing this, or is it just a "watch where you're poking and don't be stupid...don't connect power to ground" kind of thing?

Understand your equipment (which you clearly want to do ) before using it.

Use the correct class of probe; FFI see the references and safety pointers in https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

Never ever ever "float a scope" by disconnecting a scope's protective mains earth. And realise that isolation transformers don't magically remove high voltages, and they do defeat the operation of RCDs/ELCBs

[Laval]

Unfortunately this is also how people sometimes end up blowing their scope when they inadvertently short their devices through scope input ground.

Is there something I can do to protect against doing this, or is it just a "watch where you're poking and don't be stupid...don't connect power to ground" kind of thing?

It is possible to float the DUT using an isolation transformer but the more generally correct way to use your scope in such a situation would be to use a differential probe.

This video might be of interest to you

[K5_489]

otherwise at some point you can regret

Well, to be fair, this is part of the reason I have a cheap Chinese scope sitting on the bench here as a first rather than a multi-thousand dollar R&S or Tek - When I inadvertently do something highly stupid, it will hurt the wallet less.   

Compared to already being comfortable with using a VOM, and being fairly confident that I will avoid doing said stupid things with it, so I was comfortable buying a rather expensive Fluke 177 as opposed to a $10 Horrible Fright special. 

[Laval]

otherwise at some point you can regret

Well, to be fair, this is part of the reason I have a cheap Chinese scope sitting on the bench here as a first rather than a multi-thousand dollar R&S or Tek - When I inadvertently do something highly stupid, it will hurt the wallet less.

DMM are designed to measure the potential difference between two points of a circuit, they are basically floating. Oscilloscopes (except the battery operated ones) always have their ground lead connected to mains ground. This means that connecting the little black alligator clip to a point in a circuit is the same as connecting a jumper from mains ground to that point of the circuit which may result in anything from taking an erroneous measurement to blowing up the circuit, the probe, the oscilloscope or all of them.

Differential probes make your scope behave like a DMM. They generally have a lower bandwidth then your scope and higher voltage measurement capability.

In some cases, floating the circuit under test with an isolation transformer can do the job too. A hand-held battery operated oscilloscope is also a solution (it also behaves like a DMM).

When probing a circuit with a scope it is good to have an understanding of the circuit and the scope. This link shows and example of wrong measurement caused by grounding the circuit.

https://hackaday.com/2014/12/03/scope-noob-bridge-rectifier/

[CaptDon]

What input power is that service monitor rated for? Are you feeding a proper dummy load and sending a lower power sample to the service monitor? Most service monitors are only good for about 1 to 10 watts of input power, most won't go anywhere near 80 without internal damage.
.

[K5_489]

What input power is that service monitor rated for? Are you feeding a proper dummy load and sending a lower power sample to the service monitor? Most service monitors are only good for about 1 to 10 watts of input power, most won't go anywhere near 80 without internal damage.
.

Wavetek SSI 3000.  Internal load is rated at 100 watts, though I inadvertently sent 500-ish through it a few years ago, and it's been a little wonky ever since, and why I don't fully trust that the meter is reading power levels 100% correct. 

Annoyingly, I have a Bird 43 with 1KW elements around here....somewhere...but I can't remember where the heck I stashed it...