Connecting mains to DS1052E safely

[Apothus]

Morning.

I have a need to view the mains AC waveform and I would like to do it without killing myself or destroying my scope. Unfortunatly i am not in the position of being able to afford a differential probe input for my scope even though most examples recomend them. It is also not feasible for my to galvanically isolate the entire project as my load draws a significant ammount of current ~8amps however i think i have met an acceptable compromise.

" alt="" class="bbc_img" />

The Mains circuit is controlled by a Solid State Relay switched by a micro controller. I want to investigate if the SSR is opperating correctly and not re-firing so i know my snubber circuit is working. The load should not require a snubber as it is a simple resistive load with no inductive qualities.

My Intention is to connect a 10:1 transformer in parallel with my load. A load resistor placed over the secondary windings to ensure it is not floating (not 100% sure this is nesercery). Then connecting the probe and probe ground across the resistor with the 10X attenuation on. Is this a safe way to connect my scope to view the switched output waveform of the SSR? Have i connected the ground probe correctly? I know some examples recomend ***Forgetting*** to connect the ground clip and relying on the grounding of the scope itself however my understand is that with the galvanic isolation i need to attach the ground for reference?

Secondly I want to be able to view my DC waveform. Is it safe to connect this as normaly with the DS1052E scopes as i know they share a common ground across all inputs? should I take any more precautions?

I realise this is a bit of a long post but I want to do my testing correctly and if my diagram is correct i thought it may help other visual learners.

[IanB]

The scope ground clips are both connected to the mains earth. In your circuit diagram, are you comfortable making a short circuit between the three points "Probe B GND Clip", "Probe A GND Clip" and "Neutral-Earth connection in meter box"? If not, don't connect up the scope like that.

If you leave both scope ground clips unconnected, you can probe mains voltages using probes A & B on the x10 setting within the voltage rating of the scope. You will measure voltages relative to mains ground, which should be nearly the same as the voltages on mains neutral. (When I measure using my DMM I find only a few mV potential difference between mains ground and mains neutral.)

It would be a good idea to remove the ground clips from the probes before attaching them, and then connect the probes to the test points while the circuit is de-energized. After the probes are connected, power up the circuit without touching them. That way you protect yourself from the electric shock danger of probing mains voltages with thin plastic probes.

If you do it this way you don't need the 10:1 transformer, although you should always remember that working with mains voltages is dangerous and make sure your body is never in a current path in the event of a fault. Always keep both hands off while the circuit is powered up.

[Mechatrommer]

guide for dummies:
1) remove any ground clip on your probe tips, its the black loop clip like a pigtail. the idea is that you dont want to use the ground clip to connect to any live circuit.
2) use 1/10X attenuation setting. dont use 1X. the idea is so you dont fry your scope.
3) dont touch any metal part either on the probe or on the circuit. the idea is so you dont get electric shock.
4) be careful while approaching or departing your probe tip to/from the circuit so its not shorting any node in the circuit. the idea is so you dont get a big bang.
there's no step 5.

[notsob]

Step 5, if your hands are a little shaky, wear brown corduroy trousers

[Apothus]

are you comfortable making a short circuit between the three points "Probe B GND Clip", "Probe A GND Clip" and "Neutral-Earth connection in meter box"? If not, don't connect up the scope like that.

Not very confident at all, this is why i prefer TTL levels

From my understanding of your posts it sounds as though I am possibly being a little over cautious (not that you really can with mains). My scope should be able to measure the AC directly with 10X attenuation so long as i do not connect the ground clip/pigtail and instead rely uppon the internal earth connection of the scope. The DS1052E can handle the voltage considering the rating of 300V rms. Oh and de-energise the circuit when connecting or disconnecting the probe.

Thank you for the information.

[Psi]

Yeah, with the ground clips removed you cannot make the typical mistake of grounding something that's live.
For mains voltage/frequency the potential difference between ground at your circuit and earth ground at the scope is quite small (percentage wise). So for most mains measuring situations leaving the ground clips off is a good plan.
You can always use both probes together to take a differential measurement between two points if needed.

Just watch the 1x/10x probe switchs, sometimes they gets lose and your fingers may accidentally move them.
I'm 99% sure, with the rigol at least, that it can handle 230VAC on x1 for a short time without any damage, but it's still something you definitely want to avoid.

[Apothus]

Just watch the 1x/10x probe switchs, sometimes they gets lose and your fingers may accidentally move them.
I'm 99% sure, with the rigol at least, that it can handle 230VAC on x1 for a short time without any damage, but it's still something you definitely want to avoid.

Yes i belive it is more than capable of tolerating the voltage, the front of the DSO says max 300V rms.

The method described by everone here has worked fine, I was clearly being over cautios in my initial post.

I also got a little more confident and connected the DC component of my circuit to the other probe as well (this time connecting gnd to circuit gnd). Everything worked fine (no magic smoke released) and i was able to see both waveforms on the scope, however switching the AC on/off from an in-line switch would occasionally cause the microcontroller to stop or reset, something i found most bizzare. I tried to see if my switching of the circuit caused any strange spikes but could not see anything, with all probes disconnected this is not an issue.

Thanks again for everyones advice!

[johnnyfp]

Is that normal to have the ground of the scope connected to  earth? I thought that would give bad readings with transients and other crap coming back from earth

[tom66]

The probe however is only rated to 150Vrms on 1X, and 300Vrms on 10X.

Though I guess the worst case is the probe is damaged, which is reasonably cheap to replace, compared to the scope itself...

[Mechatrommer]

The probe however is only rated to 150Vrms on 1X, and 300Vrms on 10X.
Though I guess the worst case is the probe is damaged, which is reasonably cheap to replace, compared to the scope itself...

thats the best case scenario. at 1X which part do you think in the probe that will get damaged?

[tom66]

The probe however is only rated to 150Vrms on 1X, and 300Vrms on 10X.
Though I guess the worst case is the probe is damaged, which is reasonably cheap to replace, compared to the scope itself...

thats the best case scenario. at 1X which part do you think in the probe that will get damaged?

It has some resistors and capacitors; it's possible one of those could break down (some are used in the 1X mode only, which might be rated at 150Vrms max.) Or it could have a surge suppressor - I'm not sure - there are few schematics of oscilloscope probes available (they are just RC networks right?)

[Mechatrommer]

dismantling one of cheapo probe i bought last time revealed a dead short to dso at 1X, not sure about the original rigol probe but i suspect its the same. either way, damaging the probe is the best scenario, the worst case is damaging your $300 scope.

[tom66]

In which case it could refer to the fact that the cabling from the probe to scope isn't mains rated - only to 150Vrms. In 10X it would be 30Vrms.

[G7PSK]

Use a 100X probe or a probe converter.

[Mechatrommer]

there again 100X is good for KV range but not very good for small signal. you scope will limit to 200mV/div minimum. YMMV.

In which case it could refer to the fact that the cabling from the probe to scope isn't mains rated - only to 150Vrms. In 10X it would be 30Vrms.

how do you specify a bunch of copper wire is mains rated or not? the problem is with scope's front end, not copper cable. though it probably safe to use 1X as rigol specify the input to be 300Vrms. you may try, i havent tried.

[David_AVD]

I've used my analog CRO with 10x probes to check mains waveforms before.

You can also use two (10x) probes and do a differential measurement to get rid of any E-N offset.  One probe to circuit active point and the other to circuit neutral.  Set the scope to differential (or add & invert) mode so you read the difference between the two probes.  You need to set both input channels to the same sensitivity though.

[Apothus]

how do you specify a bunch of copper wire is mains rated or not?
[/quote]

All i could think of would be your insulation thickness for voltage and diameter for current howerver with most scopes i know of having 1M ohm impedance current is not going to be an issue and voltages would not be leaking through insulation untill you are dealing with kV levels i believe.

with the DS1052E you need to use it on 10X regardless because the larges V/Div is 10V so you could not see a mains waveform without the attenuation.

[Mechatrommer]

1Mohm impedance is to the ground, high voltage will still go direct to the dso's circuit front end. now i forgot how it looks like (Ahellene drawing), but if its a simple opamp with 12V common mode range, it sure get toasted.