Stupid voltage leak question

[ImpulsiveJames]

Ok, I know this is a dumb post right off the bat, but I'm hoping to minimize waste without any REAL danger.

I bought an espresso machine awhile back and have been using it for a year or two. Recently I was barefoot on our garage cement floor and got a small tingle while touching certain metal parts of the machine while it was on.... The unit it grounded with a 3 prong plug. My multimeter is measuring 12v AC between the metal on the unit and the ground pin on a power strip.

I understand water as electricity don't mix, but when you have a smaller voltage leak in the 12-20v range, is this lethal in the real world? Any advice or ways to find exactly where the leak is originating?

[AVGresponding]

Voltages that low aren't normally hazardous, due to the internal resistance of the human body. People with certain medical conditions might be more vulnerable.

I'm reluctant to say whether you have a fault or not, as I'm not a US sparky, and am not intimately familiar with the normal operation and failure modes of your systems.

[Gregg]

Perhaps the ground wire isn’t connected internally; a dubious “feature” of many imported items; an easy check from the ground pin of the plug to the metal of the device with an ohm meter.  Another possibility is a poor or missing ground at the wall receptacle. 
If voltage is present with it turned off; that is another story. 

[wraper]

My multimeter is measuring 12v AC between the metal on the unit and the ground pin on a power strip.

That metal could be totally insulated and the only reason you get that measurement is tiny leakage through tiny capacitance through air. Generally you get some voltage measurement even with free floating multimeter leads unless multimeter input resistance is relatively low. I suggest measuring current instead of voltage.

[ejeffrey]

Check again with the dmm between the chassis and ground in current measurement mode.  It's probably only a couple dozen microamps of current.  If so, it's not a touch hazard as is.  However, then do what Gregg says and check for continuity.  If you unplug the device from the wall and measure continuity from the metal chassis and any other exposed metal parts (such as pipes, the boiler tank, etc) to the ground pin on the plug and it should show < 1 ohm.  If not, those parts aren't grounded and the 3 pin plug is mostly just for show then.  If you look around inside you might see if it goes anywhere.  I don't know if there are specific rules for coffee makers, but I would think that at least the boiler and the pipes going in and out should be grounded since the heating element is at 120V and water is, as you say, conductive.

[Zero999]

You shouldn't feel a tingle from an appliance like that, which should be earthed. It's not the same as a double insulated power supply, with a Y capacitor bridging the primary and secondary sides.

Test the continuity between the exposed metallic parts and the earth pin of the mains plug.

[SiliconWizard]

The "tingle" when touching is indication that there was probably more than just a few µA that was circulating. Obviously not enough to be lethal though (the OP's family would know  ). A tingling sensation would be probably something in the tens or hundreds of µA.

Rather than a fault with the machine, it's just probably a potential difference between the earth of your installation and your garage floor. Which is not particularly surprising nor necessarily an indication of a faulty installation.

[nightfire]

This is the so called "Berührungsstrom" in german (Touch current in english?)- due to VDE regulations in germany it is allowed to be 0,5mA from a non-grounded metal part of a device to earth, measured over a 2kOhm resistor to simulate body resistance.

Reasons for this have been named, likely whether some kind of induction to non-grounded metal plates (I once had a cheap desk lamp that managed to do 30 Volts at the casing), or the grounding of  the coffeemaker or outlet is broken.

If the multimeter is sensitive enough, measuring those milliamps would be some first step.

[Zero999]

The question is whether the current leakage is occurring from a part of the device which is supposed to be earthed, or not. I suspect it should be earthed and there shouldn't be any current leaking to the user. If it's faulty, then there's a risk of death and serious injury.

[soldar]

Do not assume anything. Check and test everything. Let me tell you two things which happened to me.

Years ago I bought a two story house in a development with several hundred identical houses. After a couple years I noticed a computer case, which was supposed to be grounded, gave me that tinging feeling. After some tests I discovered all the outlets in the upper story were not grounded. All power distribution in the upper floor was done from a register at floor level which had a tube directly to a register in the lower floor. All the wire bundles had been prepared in advance and the electrician just joined all the earths in each box. BUT the earth cable between the two floors had been cut short so it was not showing in one of the registers and had not been connected. I had to pull through a longer cable and that way connect the earth to the cables in the upper floor.

So, measure and make sure protective earth is connected to the electric outlet and then to the metal parts of the device.

Another story: many many years ago I worked at a place where there was a plaster wall which gave that tingling sensation if you touched it with your hand and there was a pipe there which also did it. We just lived with it for as long as I can remember and we had theories but I think there was some pinhole leak in the pipe which caused the dampness and some electrical wire or connection was also wet.

So it could be the floor or wall which is at fault too.

[ebastler]

In the specific case of the OP's espresso machine, it all boils down to the question:

• Are those metal parts (by design) meant to be grounded, because there is a risk that they could otherwise come in contact with "hot" mains through just a single fault in the machine? If that is the case, the tingle upon touching them suggests that the parts are not actually grounded, and that's a safety issue that needs to be fixed immediately.

• Or are those parts double-insulated from mains inside the machine, and were never meant to be grounded? In that case, a tingle due to capacitive coupling of some voltage is normal. The machine may still require a properly grounded plug to keep other touchable metal parts safe, but the "tingle" does not give any indication that there is something wrong with that grounding.

In practice, I would assume that functional metal parts which are in direct contact with the heater will need to be grounded. E.g. in a typical portafilter machine, the boiler and the group head. But if the machine's outer enclosure is mostly a plastic structure, with some decorative metal parts glued on, those might well be double-insulated and not grounded.

To tell whether specific metal parts were meant to be grounded or not, one could disassemble the machine and look for provided connection points or loose ground wires. But a judgement call "from the outside" may be good enough in some clear-cut cases. @ImpulsiveJames -- which parts specifically caused the tingle when touched? Could you share the specific brand and model, and/or a photo of your machine?

[Siwastaja]

In the specific case of the OP's espresso machine, it all boils down to the question:

• Are those metal parts (by design) meant to be grounded, because there is a risk that they could otherwise come in contact with "hot" mains through just a single fault in the machine? If that is the case, the tingle upon touching them suggests that the parts are not actually grounded, and that's a safety issue that needs to be fixed immediately.

• Or are those parts double-insulated from mains inside the machine, and were never meant to be grounded? In that case, a tingle due to capacitive coupling of some voltage is normal. The machine may still require a properly grounded plug to keep other touchable metal parts safe, but the "tingle" does not give any indication that there is something wrong with that grounding.

In practice, I would assume that functional metal parts which are in direct contact with the heater will need to be grounded. E.g. in a typical portafilter machine, the boiler and the group head. But if the machine's outer enclosure is mostly a plastic structure, with some decorative metal parts glued on, those might well be double-insulated and not grounded.

To tell whether specific metal parts were meant to be grounded or not, one could disassemble the machine and look for provided connection points or loose ground wires. But a judgement call "from the outside" may be good enough in some clear-cut cases. @ImpulsiveJames -- which parts specifically caused the tingle when touched? Could you share the specific brand and model, and/or a photo of your machine?

And here we have the correct answer. Usually a lot of unnecessary noise in threads like this.

[Zero999]

And here we have the correct answer. Usually a lot of unnecessary noise in threads like this.

Except, if you re-read the thread carefully, you'll find the correct answer. I also dispute part about parasitic capacitances.

Or are those parts double-insulated from mains inside the machine, and were never meant to be grounded? In that case, a tingle due to capacitive coupling of some voltage is normal.

I doubt that it's parasitic capacitances, because they'll be far too small to pass enough current to cause a tingling sensation. Y capacitors of course can do that, but not the tiny capacitance between say a metal grip tray and some live parts inside the case. This shouldn't happen, if it has an earth on the mains lead.

[ebastler]

Except, if you re-read the thread carefully, you'll find the correct answer. I also dispute part about parasitic capacitances.

Yes, you pointed out the relevant question before. But I felt that you also jumped to a particular conclusion, which may be the wrong one; hence thought that I should elaborate a bit.

Considering the boundary conditions under which the tingle was felt (barefoot on a cement floor), I would certainly not rule out harmless capacitive coupling to a metal part that is floating by design. What constitutes a "tingle" is a very subjective assessment; it might just be the 60 Hz "vibrations" which are typically felt. Let's wait whether the OP comes back with more details about his espresso machine and the specific metal parts that caused the sensation.

[nctnico]

You shouldn't feel a tingle from an appliance like that, which should be earthed. It's not the same as a double insulated power supply, with a Y capacitor bridging the primary and secondary sides.

Test the continuity between the exposed metallic parts and the earth pin of the mains plug.

Agreed. On a grounded device, all metal parts should be connected to the ground pin. If not, then the unit is faulty and should be checked / replaced. If the coffee maker is OK, then the power cord the next suspect. It is also not unheard of that a grounded outlet doesn't have the ground pin connected.

[Zero999]

Except, if you re-read the thread carefully, you'll find the correct answer. I also dispute part about parasitic capacitances.

Yes, you pointed out the relevant question before. But I felt that you also jumped to a particular conclusion, which may be the wrong one; hence thought that I should elaborate a bit.

Considering the boundary conditions under which the tingle was felt (barefoot on a cement floor), I would certainly not rule out harmless capacitive coupling to a metal part that is floating by design. What constitutes a "tingle" is a very subjective assessment; it might just be the 60 Hz "vibrations" which are typically felt. Let's wait whether the OP comes back with more details about his espresso machine and the specific metal parts that caused the sensation.

The reason why I doubt it's parasitic capacitance is because I can touch one side of a toroidal 1kVA transformer with one hand and earth with another and don't feel a thing and such a transformer will have far greater parasitic capacitance than a coffee maker. Mechanical vibration is a possibility, but doubt it would have raised the same level of alarm, because it would be audible i.e. the pump running.

I prefer to err on the side of caution, especially given the wording of the original post indicates it was written by a beginner..

[ebastler]

On a grounded device, all metal parts should be connected to the ground pin.

I very much doubt that. As I said a few posts above, e.g. some decorative metal panels glued onto the plastic enclosure might well be unconnected to anything (and rightfully so, since they are double-isolated from mains). At the same time, the boiler and group head need to be grounded, and hence the machine comes with a grounded plug. My little portafilter espresso machine has a full metal enclosure which obviously is grounded, but the metal drip tray cover is not, since it resides outside of anything electrical. Etc...

[Siwastaja]

On a grounded device, all metal parts should be connected to the ground pin.

Not true at all. The world is full of appliances which are grounded but have floating metal parts; in fact I would dare to say 99% have some, it's very difficult to ground every metal piece. In no legislation they are illegal. Decorative stainless steel surfaces on knobs etc. are a great example.

Of course for any such floating metal part, it has to conform to double insulation requirements (or something similar; exact details depend on country). Every metal part which is user-accessible AND which can touch a live (neutral or line) wire due to a single fault has to be grounded to protective earth. But this does not comprise every metal part.

[Nominal Animal]

On some non-electrical "devices", like the aluminium pencil tray on a whiteboard, you can sometimes find static dissipation resistors (1 - 2 MΩ) to a protective earth (true ground potential, or any much larger metal structure).

Fully isolated conductive surfaces (anodized and brushed aluminium in particular, having that isolating oxide layer on top) can get statically charged to very high relative potentials in dry air conditions (e.g., indoors during cold winters here in Finland), so if they are intended to be touched by humans, static dissipation resistors can be used to stop the painful zaps one would otherwise get.  (My "best zap yet" was an arc ~ 10 cm long from a door handle to my middle finger, which numbed my hand to my elbow for about a quarter of an hour.  Painful.  Very dry air, plus static charge accumulating clothing, socks, and carpeting, all contributed.)

This is a particular problem with class II and class III (or SELV, separated extra-low voltage, powered devices), since they do not have a ground conductor to the actual device/appliance at all.

There are two mechanisms – capacitive coupling and static charge accumulation – such isolated but highly conductive surfaces can cause a tingle or a zap when touched by a hunam.  Capacitive coupling carries a low AC current, sub-milliamp range, via the surface through the human to a ground potential, and produces a tingle.  Large enough potential difference causes a single zap, which equalizes the potentials.  The zap is typically in the kilovolt to a few dozen kilovolt range, but with very little current, in a very short pulse. (The instant wattage can be surprisingly high for a very small fraction of a second.)

Non-electrical devices are easy, as they can be safely grounded via a large enough resistor, but very conductive decorative panels on electrical devices gives me the heebie-jeebies: I don't know how to deal with those properly, unless I've built the device/enclosure myself, and understand the grounding and protection scheme.  (I only do class III/SELV devices, hobby-grade, though: routers, single-board computers, and such.)

I suspect, but have not verified, that a paint, coating, or replacement panel that is only slightly conductive (i.e., "static dissipative"; sufficient resistance) is the only generally safe option in all cases.  This dissipates some of the electrical power, and spreads out the zap pulse in time, leading to lower voltage and current on contact, helping with both the tingle and the zap, compared to a completely isolated well-conducting surface.

Plastics and metals are easy to prep – sand, then use proper primer for the plastic/metal at hand, then thin layers of surface paint or laquer with fine sanding (600 grit and better) in between to remove bumps and hairs, with optionally a final linear pass with a scouring pad to get a brushed steel/aluminium effect; and not even the spouse will complain –, so it is only a question of finding a surface paint or lacquer in a rattle can form, providing a slightly conductive, static dissipative surface.

Does anybody know of any specific paints or lacquers for this, for home/hobby use?  I'm particularly interested in durable conductive carbon acrylic/polyurethane aerosol paints in rattle cans myself.

[soldar]

The reason why I doubt it's parasitic capacitance is because I can touch one side of a toroidal 1kVA transformer with one hand and earth with another and don't feel a thing and such a transformer will have far greater parasitic capacitance than a coffee maker. Mechanical vibration is a possibility, but doubt it would have raised the same level of alarm, because it would be audible i.e. the pump running.

I prefer to err on the side of caution, especially given the wording of the original post indicates it was written by a beginner..

Many years ago the steel kitchen sink was giving me the tingling.  The sink was electrically isolated because it had PVC drain. After some confusion I determined the sink was touching some aluminum foil which, in turn, was touching a toaster oven. The all-metal toaster oven had two heating elements which were metal tubes with, I assume, resistive wires inside, isolated from the outside metal tube which was held in place and electrically connected to the metal structure of the oven. Measuring resistance between the internal heating element and the external metal tube the reading is infinite resistance. But there was enough capacitance to give that tingling. The oven was supposed to be connected to earth and that would solve the problem but my apartment at that time did not have protective earth wire in the supply. This was long time ago.  This was a 6th floor apartment and the building did not have earth wiring.

I bought some heat resistant flexible tubing and put it between the metal tube and the metal clamps which hold it in place and that resolved the problem.

I think it might be asbestos but at that time asbestos was not yet dangerous.  I still have and use this toaster oven but now it is grounded.

Just an example if capacitive leak. With the oven grounded there is no problem but with the metal housing floating it does give the tingles.

[nctnico]

On a grounded device, all metal parts should be connected to the ground pin.

I very much doubt that. As I said a few posts above, e.g. some decorative metal panels glued onto the plastic enclosure might well be unconnected to anything.

That is semantics. Ofcourse this is about metal parts which are a structural part of the appliance which are close to mains voltages and where double isolation can't be guaranteed. Should go without saying. Either way, if touching an appliance gives you a tingling sensation, it is time to unplug it and have it checked / replaced.

[Zero999]

The reason why I doubt it's parasitic capacitance is because I can touch one side of a toroidal 1kVA transformer with one hand and earth with another and don't feel a thing and such a transformer will have far greater parasitic capacitance than a coffee maker. Mechanical vibration is a possibility, but doubt it would have raised the same level of alarm, because it would be audible i.e. the pump running.

I prefer to err on the side of caution, especially given the wording of the original post indicates it was written by a beginner..

Many years ago the steel kitchen sink was giving me the tingling.  The sink was electrically isolated because it had PVC drain. After some confusion I determined the sink was touching some aluminum foil which, in turn, was touching a toaster oven. The all-metal toaster oven had two heating elements which were metal tubes with, I assume, resistive wires inside, isolated from the outside metal tube which was held in place and electrically connected to the metal structure of the oven. Measuring resistance between the internal heating element and the external metal tube the reading is infinite resistance. But there was enough capacitance to give that tingling. The oven was supposed to be connected to earth and that would solve the problem but my apartment at that time did not have protective earth wire in the supply. This was long time ago.  This was a 6th floor apartment and the building did not have earth wiring.

I bought some heat resistant flexible tubing and put it between the metal tube and the metal clamps which hold it in place and that resolved the problem.

I think it might be asbestos but at that time asbestos was not yet dangerous.  I still have and use this toaster oven but now it is grounded.

Just an example if capacitive leak. With the oven grounded there is no problem but with the metal housing floating it does give the tingles.

I doubt it had anything do to with the capacitance. There are other reasons why the toaster leaked whilst on the drainer, yet your meter measured open circuit, such as moisture ingress into the oven, or it not being purely resistive, i.e. the resistance of some materials drops, with increasing voltage and the meter uses a much lower test voltage than the mains.

On a grounded device, all metal parts should be connected to the ground pin.

I very much doubt that. As I said a few posts above, e.g. some decorative metal panels glued onto the plastic enclosure might well be unconnected to anything.

That is semantics. Ofcourse this is about metal parts which are a structural part of the appliance which are close to mains voltages and where double isolation can't be guaranteed. Should go without saying. Either way, if touching an appliance gives you a tingling sensation, it is time to unplug it and have it checked / replaced.

I agree. The internal connections should not leak to external metalwork like that. It's a sign something is wrong.