EEVblog #279 – How NOT To Blow Up Your Oscilloscope!

And how NOT to blow up your Arduino.
Your oscilloscope ground clip lead can be a potential accident waiting to happen.
Why? And how do you avoid it?
The basics about measuring mains earth referenced equipment with your oscilloscope.

Be Sociable, Share!

    About EEVblog

    Check Also

    EEVblog #800 – Siglent 1000X Oscilloscope Teardown

    EEVblog #800 – Siglent 1000X Oscilloscope Teardown

    Teardown of the new Siglent 1000X Series Oscilloscope How to remove warranty void security stickers ...

    • Andy

      Thanks for the video!
      I’ve encountered this problem before when testing the output of a bridge rectifier.
      It would short the input ground and the output ground which would make it look like a half wave rectifier and cause a fair few issues!

    • Maalobs

      This reminds me of a thing back at school; every labdesk would have a row of earthed power outlets protected by a fuse at the individual desk, and there were also a couple of unearthed outlets there.
      The unearthed outlets were only to be used with oscilloscopes.
      This was fine, but if you were wearing the antistatic wrist strap while working and touched the metal chassis of your scope, there was a very nasty sensation around your wrist as you created a short circuit as described in the video. If the MegaOhm resistor hadn’t been present in the wrist strap, the sensation would probably have been more dramatic.

    • Graeme Pattie

      The other way to just use tektronics ths720 scope it,s battery operated but has 1000volt isolation between grounds on BNC input and the meter input or latter scope with the same type of input.

    • Paul Jones

      You didn’t mention using two ungrounded probes with the A-B Math function – Perhaps not the best way but good enough on a budget.

    • Graham

      The fact that there are only two pins on a power connector does not mean that one of them is not connected to earth. In Australia at least all neutral lines are strapped to the earth in the power board via the neutral link. So any potential that is on the neutral lead is automatically connected to earth at the power board via the neutral link.

    • james derp

      This reminds me of an incident during an electrical class at high school.
      One day he was demonstrating how to use an oscilloscope to test one students’ circuit.
      He attached the ground clip on the probe and started to test… about 6-8 seconds later some smoke began to rise from mr. weak plastic-insulated alligator-probe wire.
      Every student laughed at his incompetence but I never knew -exactly- why until today.

      Now that I think about it, each bench had a common power supply, non-isolated. That’s probably what caused that smoke and embarrassing-curcuit-disconnect. lol

    • f4eru

      Removing the earth from an earthed oscilloscope is against security rules of class I isolation, but sometimes it’s very handy. Remember, in this case, the Y1 or Y2 capacitors in the line filter will put the GND of your scope at line voltage/2, floating at hundred volts or so (for 220VAC)

      Also a thing to note :
      If you use a laptop with no earth connection (class II PSU, the GND of the laptop will float around at 100V… as long as no peripherial grounds it ! As soon as you plug a peripherial (earthed or not), a transient will flow into it from discharging Y capacitors. This can destroy circuitry, especially if a signal wire is connected before the power wire. This is the reason USB connectors are designed so that power wires make contact first !

      I usually use an isolation transformer for the DUT (if i want to sense mains connected circuits like switching power supplies) plus eventually an unearthed scope, and i eventually also earth GND of my circuit.

      But remember ! you are violating security rules! Be very careful…

      Also, another thing with putting the GND of the scope at a random point of your circuit is that you will inject currents, and may wreak havoc into your circuit : Even with isolation, there are capacitive couplings which transmit 50Hz AC, and PSU switching frequencies. Beware of wrong measurements….

    • Salas

      Two channel floating solution:

    • perplexed

      It’s always been bad practice to place your scope ground clip to anything other than the ground on your test board. Even if your scope is isolated there may be enough coupling going on to at least provide you incorrect readings.

      You dont need to buy differential probes or any other gear, just use two channels and the subtraction function as someone mentions.

      • EEVblog

        Differential mode using A-B on a scope has pretty bad CMRR.

    • Njay

      I don’t understand why (at least the cheap) Rigol doesn’t have fuses on their probe input ground. Wouldn’t that avoid some disappointments?

      • f4erumain Pluie Max1ch will blow the

        A fuse in the ground will be worse :
        the GND fuse would blow and the full voltage would be on the tip (which blows the input amp if in millivolt mode and you inject 220V)

        And also it would induce an impedance in series with GND, wich would make a big measurement error, esp. in high freq.

        Also, it’s forbidden, due to security rules.

        In fact, you will not blow your scope. The GND connection in almost all scopes goes through massive ground planes, most times also the housing.
        Thus, the thing blowing is mnost likely your circuit, and/or the differential breaker on your mains.

        A good tip for your lab : put a 30mA differential breaker !!!! potential life saver for a few Euros/dollars

        • Njay

          You shouldn’t be poking mains with your oscilloscope :D. Actually I was remembering a situation with my Rigol where I had some current flowing through the probe’s ground (measuring a small power supply I think) and the current was high enough to make some damage but low enough for I to discover it only by smell a few minutes later. Some wire plastic cover must have melted inside (never checked, I’m saying it just by the smell).

          • f4eru

            Yeah, when you make or repair mains circuitry, you have to probe it…

            The iso transformer for the DUT is IHMO the best solution.

    • Tom

      In case anyone’s interested, olimex do cheap usb 2 isolator boards for around £30.

    • Salas

      There are battery scopes about too.

      • rick

        Right despite Dave saying “do not buy those”, I think the only use of this type of equipment is to provide the type of floating measurements he is talking in his blog. But the bandwidth is limited.

      • rick

        Sorry, I was referring to handheld oscilloscopes.

    • Markus Gritsch

      Great video!

      I have another grounding related question: On some switch mode AC-adapters, I feel a small “vibration” when touching the GND connector. Measuring the voltage on this GND pin reads over 100 V AV relative to the mains earth pin, but very high impedance. The voltage already breaks down when connecting 1 MOhm.

      I read somewhere that this is related to the capacitors used in the switch mode power adapter, but I would like to understand it better. Maybe you could explain this phenomenon in another video.

      Thanks, Markus

    • seb

      At end of video (20:00) when arduino is powered by switchmode psu, I can hear some noise on audio when you touch positive pin with alligator clip. Something is happening !!

    • Paul

      “Didn’t have time to build it to scale or to paint it”. LOL. Back to the future!!!

    • John

      How about handheld oscilloscopes? Sure they cost quite a bit more than desktop counterparts but the differential probes aren’t too cheap either
      and handheld oscilloscopes are floating by design…

    • Rubi


      I prefer the diy usb isolators from circuitsathome.

      I bought some pcbs at their shop and built them myself.

    • Aram

      Thanks Dave… I can’t watch this enough — Thumbs up…. and it is a big issue when servicing professional audio power amplifiers. I have gone round and round with factories and even Tektronix on the best way to test. I have resorted to an Iso, but I think it is time to look into diff probes… but they are very expensive. Something feels wrong about placing a probes ground clip when the DUT is isolated in this sort of circuitry. Also many amps have bridge mode which creates another set of issues. I try to use the diff mode on the scope.. but it is kind of a pain, and a bit limited. Good luck to you…Glad the language got toned down.. you’re getting more professional everyday!

    • Francirius

      a man told me once that u can measure these circuits by putting a 1k (or less) in series with the ground pin…ive never probe that, have u heard something like that??

    • Frederick


      on school we use pc scopes (don’t ask me why) but my question is or these have the same grounding?

      • EEVblog

        Through the PC, the same grounding on the USB cable. Same as the Arduino example.

    • Paul

      I thought of this particular weblog while tinkering with my scope. I just thought myself of putting a 1k resitor in the groundlead just as Franciscus thought.

      It probably wouldn’t interfere with measurements because that 1k is negligible in comparison with the 1Meg or 10Meg input impecance of the scope.

      So I think that the worst that could happen is that that 1k resistor burtst into flames. Much better than burning your scope.

      Maybe a 100mA (or the smallest you can find) is even better.

    • Rick

      That happened to me too. Longtime ago I had an osciloscope and a function generator connected to a circuit and suddenly I heard a bang. But the osciloscope was fine, the generator however died, when I opened it I saw some traces of smoke in it. I shall buy a new scope now and I am really glad to realize now what happened. Excellent video! “Thumbs up”;)

    • Rafael

      Dave, I just wanted to let you know I shared your video to a customer at:

    • rick

      And what if we just isolate an equipment from the earth connection by using an “adapter plug” without earth connection on it? Something like that,PR-203.html

    • Use Oscilloscope

      Nice post, I will forward it to my readers so they don’t damage their scopes :)

    • Pingback: DIY Isolation Transformer « Oakkar7, another Blog()

    • Brandy

      Great tips! :) Have you seen this? Cute infographic for newbies.

    • Pingback: Como NO romper tu osciloscopio! | Automatismos Mar del Plata()

    • Lawyers

      I’m usually in order to blogging and that i really thank you for content. The content offers really peaks my personal interest. I’m going to save your site as well as maintain looking at for brand spanking new information.

    • Manoel Lemos

      Hi David, thanks for this awesome video.

      My question is: how to safely measure phase error on a transformer (primary will be connected to mains) with an oscilloscope?

      Will it work if I forget about the alligator clips and use the two probes (one in the primary and other in the secondary)?

      Thanks in advance!!!

    • Pingback: reverse engineering help()

    • tecmax

      The Isolation Transformers are known to be the best options for conferring protection to the electrnonic devices. Thus these devices are considered to be vital for enhancing the endurance of the devices. Servo Stabilizer Manufacturers, Isolation Transformer Manufacturers

    • tecmax

      The Isolation Transformers are known to be the best options for conferring protection to the electrnonic devices. Thus these devices are considered to be vital for enhancing the endurance of the devices.

    • Natacha

      Ha ha great video! Mine litteraly just exploded… but I’ve found some great oscilloscopes here:

    • Pingback: Processor Low-power Optimizations in PIP-Watch | Jara's Spare Time Electronics()

    • Allan Hogue

      I know this is an old blog but I hope you see it and read this before scoping power circuits.
      The old trick was an isolation transformer and a plastic garbage can as a desk. Sum channel 1and 2. A good way to get killed. the frame of your scope could easily reach 500vac to ground. If you or your people work on these circuits a scope with true differential inputs can help. We worked on 600 hp VFDs and 750vac heater controllers and something like the Tecktronic PL2020 is a good solution.

      • Dave Jones

        Yeah, nasty stuff!

    • Pingback: Oscilloscope probe Ground clip connect to negative supply causes a short()