Afraid to use my oscilloscope?

[DennisCA]

Let me start by saying I've seen the "how not to blow your oscilloscope up" video as well as the "scopes for dopes video" and while I feel I learned from those videos it wasn't enough to assail my worries.

I still don't feel like I understand well enough to know where to safely probe, wheter to use the ground clip or not on the probe. For instance I have this alarm clock radio, the alarm function isn't working, I think the oscilloscope could help me test the IC and other paths on the scope, but I feel hesitant to try.

The plug is ungrounded, but if I understand correctly it's still earth referenced via the neutral and because my scope is grounded, I should be able to just use the probe (fixed 10x)  on the inputs and outputs of the chip to see what's going in and out? The transformer is 230V > ~8.5V, though it has a center tap for ~4.3 volts so it should be safe from high voltages, but I feel very uncertain about everything I do at the moment.



Are there any procedures, or videos or texts pertaining that teach me how to help one determine where it's safe to probe? What to look for, general principles, clues, tips? In addition to the two videos mentioned earlier.

I don't think an isolation transformer for the DUT is the answer to every question. Also I think I used an IT that I would need to put the earth clamp on the probe and provide an earth reference that way. But I am uncertain where it ought to go, don't even have a schematic for this. I looked at the chip layout but didn't see an earth pin, that would've been my goto otherwise.

I know the proper way is to learn about circuits and electrical theory to the point that it just makes sense, but I am not there yet... Got a book about circuits I started on recently... I feel mucking about with real circuits also teaches me things as well as keeping me motivated to read on.

[capt bullshot]

At the left hand side, you should notice the mains transformer. The "dangerous" mains voltage ends here. Apparently, there are a few components on the PCB that are connected to mains (like these blue resistors). Stay away from these with the oscilloscope. If you can't guess which components from experience, remove the PCB and follow the traces on the other side. This radio looks like a quite old design, they've made some connection through high value resistors to mains - maybe to make "sensor" keys work, but might be something else anyway.

The transformer (and the high value resistors) isolates the rest of the radio from mains, so there's no earth connection to the circuitry. As it is isolated, you have to connect the ground clip somewhere, preferably to circuit GND, to get results.

So to use your scope, you should first of all find the common GND of the isolated side (this might be the "-" minus) terminal of one of the larger electrolytics and connect the ground clip there. The shields of the tuneable inductors might work as GND as well, usually I'd expect them connected to circuit GND.

[Nusa]

Do sanity checks with your multimeter first...that'll save you a lot of grief. In fact, do all your initial sanity checks with the multimeter. There's little reason to use the scope before you have identified a signal worth watching.

[Ian.M]

Without a schematic (and preferably a full service manual), its difficult for  beginners to assess such a board.

Start by getting the board out and taking better^* photos of both side of it and posting them here so we can help you identify if there is an isolation barrier, and if so, which areas of the board are mains 'live' and which are reasonably safe low voltage.

The four resistors near where the mains lead is connected to the board are suspicious - they may indicate the existence of a non-isolated zero crossing detection circuit for timekeeping.

* PCB photos need to be well lit (avoiding shadowing), in focus, high enough resolution to read component markings, and in most cases taken with the camera lens centered over the board to avoid perspective distortion.

[Doctorandus_P]

1. First unplug your gadget (it probably is unplugged, but can't see the mains plug).
2. Unscrew the PCB from it's housing, so you can also examine the bottom.
3. Follow the mains wiring. (It probably goes directly to the transformer)
4. Make notes (metal or with a felt tipped marker) of where the mains voltage goes.
5. Isolate that part of the circuit, so you can't touch it.

This can for example be done by first cutting out some corrugated cardboard to even out the underside of the PCB combined with some plastic or a few layers of duct tape over the cardboard. Then clamp your custom made isolation to the PCB (screws if mounting holes are available, or use duct tape / isolation tape or a few small dabs of hot glue)

Usually anything after the transformer is safe to poke at with your oscilloscope, especially when you use a 1:10 probe.

The biggest risk is the crocodile gnd clamp. There often is not a good attachment point, and it gets pulled in all kinds of directions while you are measuring. I often identify the "GND" connection of the circuit, and then solder something to it which I can securely grab with the crocodile clamp.
The danger here is that if the crocodile clamp slips of the GND connection then it has the potential to short two parts of the circuit together on the PCB.

If you want to be extra safe, then you can modify the GND lead of the scope probe and put a small fuse in it.

For general safety you have to be aware that some big high voltage capacitors (primary circuit of SMPS power supplies) can hold hair raising charges for some time after a circuit is unplugged.

There are also some circuits that inherently work with high voltages.
* Nixie tubes -> Few hundred volts.
* CCFL backlight upto a few kV (Keep your scope away from this).
* Fluorescent displays 60V to 150V (approx, I think so somewhere in that region).
* more ...

Edit:
The big resistors are indeed suspicious, and they are likely used for detecting the zero crossings, but for that you need both the "phase" and the "neutral". It's a bit weird they did not just connect this to the secondary side of the transformer.

Idea:
It's quite easy to make a small isolation transformer if you have two identical transformers. For example use one for 230Vac -> 12Vac, and then use the other backwards for 12Vac ->230Vac.

Yet another idea:
You can analyze the power supply section, and then solder some wires just after the rectifier and then inject a DC voltage from an external power supply. This way you do not have to plug your gadget into mains voltage at all. If your gadget has multiple secondary windings this is more difficult.

[DennisCA]

Thanks for all the suggestions so far, however I noticed nobody remarked upon what I mentioned of not using the ground clamp and instead relying on the common earth reference between the DUT and oscilloscope, I have seen youtube videos where they do that and also seen posts on forums where people say that's how they mostly use their scope when looking at circuits. Is this a bad practice vs using an IT? (I don't own an IT either, for now).

I do believe I probed that big resistor with my DMM and it delivered 230V.

[capt bullshot]

Not using the ground clip of the oscilloscope probe is in general a bad idea. A common earth connection can provide the signal ground for some rather rough measurements, but this is not the normal use case. The "normal" way is to always connect the ground clip to a suitable circuit GND.

There are special circumstances that will give you halfway usable results (gnd clip not connected), including low frequencies and high signal levels. In general, not using the gnd clip will bring all kinds of interesting distortion and interferences to the scope if the signals are small (less than maybe 5V) and of higher frequency than maybe audio range.

Most circuits don't require an isolation transformer, as they are safely isolated from mains anyway (by their power supplies). If the circuit isn't, I'd recommend to stay away until you've gained more experience.

The big resistors provide safe isolation from mains, although this technique isn't common anymore. Don't know whether it's common, but from what I've seen, this kind of radio uses the very small current from mains through these resistors to make "Sensor" buttons work. If you touch the button, a small current flows from mains through your body that is detected by a simple circuit. Because the resistors are of high value (1 MOhm ballpark), the current is small enough you won't notice it. For the same reason (very high value resistor) these resistors don't provide a suitable ground path for scope measurements.

[SilverSolder]

Using an isolation transformer is a good idea - it will make you feel much more confident and comfortable, even if you "could" measure without one...   

[Renate]

Normally such a small radio with a transformer would be safe, but it looks like they are feeding some clock signal from off the line so you should be careful.

In a normal repair facility you'd have an isolation transformer which you'd use even in normal circumstances (in case there was a primary secondary short).
If you can afford it, pick up two transformers, like 230 VAC to 12 VAC (or 24 VAC or 6 VAC or whatever).
Connect the primary of one to line, connect the two secondaries together, connect the primary of the second one to the power cord of the radio.
The efficiency is not good and if the transformers are small you won't be able to play at full volume, but that's not a big deal.

Since you want to fix the clock the line signals may be where you have to start and an isolation transformer will make you happier.

[DennisCA]

I found this on a 2nd hand site, sold as an isolation transformer, 30 euros:


Put it in an enclosure and hook up two outlets and I should have an IT right?

[SilverSolder]

Good find!  - it's not often that you see a toroidal isolation transformers - they are superior to the normal types.

Warning - When you mount it in your enclosure - don't accidentally create a 1 turn loop with the bolt through the middle and your enclosure!

[Old Printer]

Make sure you are using at least a 10X probe, and if still concerned get a 100X one for good measure.

[Renate]

That's a nice transformer and 1000 VA to boot.
You're going to have to start up your own repair shop!

[Brumby]

Please be VERY careful if you are thinking of using an isolation transformer.

The obvious benefit brings with it some risks that are not immediately obvious.  The general wisdom is to NOT use an isolation transformer if there is any other way of performing the measurement ... and this starts with the question: "Do you really need to measure this on a scope?"

A set of appropriately rated differential probes is by far the better option.

[DennisCA]

Differential probes are kind of expensive though. But I don't know I need to use an oscilloscope on the high voltage circuit, I was mostly considering using that on the low voltage parts.

But I also know that on the whole I will most likely need an IT for future projects as well, so I might as well get this one. I have written a list of tools to keep a look out for on the 2nd hand market, such as a variac, signal generator (into the RF range) and current limiter seems like some basic useful tools. I have no funds for buying new equipment so I am willing to spend a long time waiting for them to pop up, like this transformer did.

I have fixed 10x probes for this scope. It's an analog 40mhz scope (hameg 404-2).

[HB9EVI]

Indeed you can cause you trouble with the usage of an isolation transformer - most obvious is the fact, that as soon as you clip the GND of the scope in a circuit, there is no isolation anymore. That's not a big problem, if you're probing in a low voltage circuit, but doing so in a 230V circuits means you and the scope are in the same danger like without an isolation transformer.

like it was said already, the right way to go working on mains voltage level is a differential probe.

if you're probing the on secondary side of the device's transformer, so there where you measured 8-9V not much can happen to you or the scope; but beware that there is mains voltage on the same pcb and sometimes the way to get in touch with it is shorter than it seems.

[freda]

Don't you normally power the scope AND the  DUT from the isolation side ?

what bothers me most, is that say you had a brain fart and did something stupid,
the RCD protection on your mains won't trigger, will it ?

[james_s]

Don't you normally power the scope AND the  DUT from the isolation side ?

what bothers me most, is that say you had a brain fart and did something stupid,
the RCD protection on your mains won't trigger, will it ?

No, why would you do that?

It's true that an isolation transformer defeats the RCD. I don't consider this a serious issue though, in much of the world we don't even use RCDs except for outlets in wet locations.

[Kalvin]

You could use a lab power supply to feed voltage into the clock's circuitry. In this way you would not have to deal with the mains voltage at all while probing the circuitry.

[freda]

Don't you normally power the scope AND the  DUT from the isolation side ?

what bothers me most, is that say you had a brain fart and did something stupid,
the RCD protection on your mains won't trigger, will it ?

No, why would you do that?

It's true that an isolation transformer defeats the RCD. I don't consider this a serious issue though, in much of the world we don't even use RCDs except for outlets in wet locations.

well mains powered scope, mains earth lead connected to chassis, well is on my old tektronix, and I did that on the very rare
occasion I was checking non isolated mains powered stuff in way past.
Sure I got battery powered scope now, even the power brick is isolated, no earth, so nothing to worry about with these as they
are by default "floating".

[Ian.M]

If a scope has a ground pin on its power in connector:

*NEVER* float your scope

Competent engineers who thought they knew what they were doing, and who had taken what they thought to be reasonable precautions against contacting a live part have been killed doing exactly that.  e.g. https://groups.io/g/TekScopes/message/10754

A CAT II 600V rated differential probe from a reputable manufacturer isn't likely to be cheap, but is still far cheaper than a funeral!

If it does not have a ground pin on its power in, read its manual carefully.  It *may* be rated for floating measurements up to 50V from ground, but *NOT* for mains voltages.  Very few scopes, even handheld, are rated for full floating measurements Cat II 600V or better on mains voltage circuits.  If you can touch the metal shell of any BNC connector or any other connector with a single channel connected via its original probe lead, you can bet your handheld scope is not CAT II rated.

Except in the rare case of scopes designed and intended for fully floating operation (e.g. Fluke ScopeMeters), your scope was designed under the assumption that no hazardous voltages would be present on its chassis (or front panel PCB ground plane). You don't know how good the insulation is between the scope chassis or ground plane and your finger at the supposedly insulated control you are touching.    e.g. conductive dirt may have built up on the sides of a button, or in the hole for the grub-screw of a knob, from fingerprint oils and ambient dust.  If your scope isn't properly grounded, and you are using it on a non-isolated mains derived circuit, all it takes is one bad mains spike at the wrong moment i.e. while you are pushing a button or holding a knob, and there's a significant risk of one dead engineer.

[m k]

Long long ago I had to connect the protective ground to get anything out from an old Tek and single ended probe.
Possibly the whole measured circuitry was totally unknown.
Maybe the isolated one was the scope and not a DUT.

I've also measured mains without any protection but that is finally pretty stupid thing to do.
It's not the mains but the probe and where it actually is when it isn't.

BTW, around here and for repairs the isolator transformer used to be a law.

[PKTKS]

(..)
BTW, around here and for repairs the isolator transformer used to be a law.

IT is ... the best rule of them all.

ISOLATE all your gear on the bench to repair nasty stuff..
you do not know how bad the DUT is ..

And safe rules in these cases are not handy.

I have my bench ISOLATED (floating) for decades..
so far  so good nothing weird happened.

Just think ahead and be safe

REPAIR   and R&D  are 2 very different things


Paul

[freda]

I gues

If a scope has a ground pin on its power in connector:
..................

I guess I generalized too much in my post. As yeah I was referring to a scopemeter as I have, actually its a bit old, since its labeled by Phillips.
But I do have another scope, with 4channels, color display and a lot of fun to use and intended for digital and general low voltage scoping.
I appreciate your post Ian.M to remind myself never to use it in any voltage stressfull way.

[Kalvin]

In my understanding the DUT (device-under-test) shall be floating and thus connected to mains using an isolation transformer, and the measuring instruments shall be connected to the protective ground. NOT another way around.

And do not solder live high voltage circuits even when the DUT is floating (or, you THINK it is floating).