Why was i electrocuted?

[Relayer]

Hello treez,
Is your PSU a variac? i.e. 0 to 240V AC variable supply? Or is it a 1 to 1 isolation transformer?
Regards,
Relayer

 

[treez]

its a variable AC supply..isolated output

 

[betwixt]

After going through the previous posts, I think the conclusion has to be that the isolated AC supply was not at he same voltage as the Earth pin on the 'scope socket and you became the discharge path when you joined them. Even if the AC supply output was 100% isolated there could be no guarantee it had no offset from Earth and in any case both lines carrying voltage couldn't both be at the same Earth potential. If there wasn't some offset from Earth potential it implies the output isn't truly isolated as it SHOULD float.

What you experienced was a grounded scope (check - I bet you find almost zero Ohms between the BNC outer and the power plug Earth pin) and an AC supply that charged its output through stray capacitance to a high voltage. What caused the shock was that capacitance discharging through your finger.

The solution is to Earth one side of the AC supply output, either directly or through a resistor then to take scope measurements using two high impedance probes in "invert and add" scope mode so you take a differential measurement. Doing it that way and on a static dissipative mat meets all the safety requirements and also presents minimal load to the circuit under test.

Brian.

 

[treez]

Thanks, i hope i explained it right, the BNC of the coax probe was actually connected to the scope socket when i touched the BNC of the coax probe, and then got electrocuted....that BNC of the coax probe should have been hard connected to earth ground...because as you know it was connected to the scope's earth via the socket "outer bit".

But i will try and measure with DMM from BNC of scope to scope's earth pin on its mains plug...because if that isnt connected then that explains it. I now wonder if our contractor took the mains plug of the scope apart and disconnected the earth so as to reduce the chance of the scope ever getting blown up...eg blown up by say connecting a non-diff probe to a non isolated mains circuit.

 

[Relayer]

Hello treez,

its a variable AC supply..isolated output

Somehow I find it hard to believe that an auto-transformer can have isolated output.



As you can see from the diagram above, an auto-tranny can't be isolated.
Therefore you will have no isolation between input and output.

Your scope should have been plugged into a 1:1 isolation transformer. Especially if
you use it on equipment, projects , appliances etc. that uses switch-mode designed
power supplies.

You could make your own isolation transformer using two identical trannys back to back.
Say, as an example: two 240V AC primaries to 12V AC secondaries... The secondary voltage
isn't critical, but the two transformer secondaries must be the same.
Depending on your current needs, will depend on the transformer VA ratings.
If you do proceed to make your own, then you MUST include a
fuse in the mains input side of one of the transformers. Current rating will depend on the VA, but
to ensure reliability of the fuse, it should be a slow-blow type.
Regards,
Relayer

 

[KlausST]

Hi,

An autotransformer has primary winding and secondary winding combined, therefore it can't be isolating.
Autotransformers can be variable or not.

Usual tranformers have separated primary and secondary winding. They are isolating.
Usual transformers can be variable or not.

Klaus

 

[FvM]

Presently, it hasn't even been stated if "isolated AC power supply" is a low frequency transformer or an electronic AC supply (isolated transformer). Discontinued product or not, it should be possible to determine the supply type.

I previously guessed that excessive leakage current could be the problem, which wouldn't be surprising for an electronic supply. A variable safety transformer ("variac") without additional output filters should have a low leakage current < 1 mA.

Regarding oscilloscope grounding, all general purpose oscilloscopes that I know have the probe input connected to power ground (protective earth). Manuals have safety advices like below:

CAUTION The probe ground lead is connected to the oscilloscope chassis and the ground wire in the power cord. If you need to measure between two live points, use a differential probe. Do not negate the protective action of the three-conductor power cable by disconnecting the oscilloscope from ground. The oscilloscope must remain grounded through its power cord for safety.

 

[Kajunbee]

The variac I bought has no isolation. I found one identical to it that specifys isolated output. Not sure how there providing isolation, couldn't find a schematic on it.

- - - Updated - - -

On the product description it says built in isolation to prevent interference.

- - - Updated - - -

Is it possible that it's not true electrical isolation.

 

[treez]

Thanks, we are using a Lisun King Isolated output AC PSU...but our model has gone obsolete.
Its and inverter based AC source with isolated output.

 

[c_mitra]

You can get a shock even with a very small current- say 100 uA or less. However, with a dry skin you just feel a tingle. Currents ~1mA can be fatal (voltage does not matter; if the current enters the bloodstream and enters the heart, even 10-20V can be fatal). If you can measure the voltage with a digital multimeter (that may have an input impedance of 1M) that it is rather serious. If the voltage can be felt (as a shock) but cannot be measure it surely comes from some leakage somewhere.

Even isolated PSU can have some leakage (it is just a matter of degree) but it may be worth investigating the details.

 

[FvM]

Its and inverter based AC source with isolated output.

That was my assumption. Typically an inverter based AC source needs output filters to comply with EMC regulations, the Y capacitors can easily cause leakage current in the mA up to tens of mA range.

 

[treez]

You should use a static disipative mat as a conductive one will still transfer charge quickly, which is exactly what you don't want for any kind of ESD protection.

Thanks, so what are the conductive mats for? They are on sale.
I say its “conductive”……….by that I mean if a metal live terminal and a metal neutral terminal are touched to it, about 2 inches apart, then the mat starts smoking….thats how conductive it is. Is that too conductive? Is this maybe why we have had so many failures of our product after re-working the PCBs on these types of mats?
We do connect them to the mains earth via the press-stud.

You need a leakage current below 1 mA to safely touch the floating circuit.

Thansk, this isolation transformer doesnt say what its leakage current is....it quotes a standard, but we cant afford the standard. Do you know why, with such an important parameter, they dont state it here..

http://carroll-meynell.com/portable-isolation-transformers/item/isolation-transformer-cm750-230

 

[KlausST]

Hi,

If the mat starts to smoke it is wy too conductive.
There is a good chance that this caused additional failures ...after rework...

Throw it away and buy a good one.
Usually the distributors for antistatic laboratory equippment should have information about what and how to use.

In our laboratory we don't have a single part connected directely to earth GND..there is at least 1MOhm series resistor.

And the mats seem to be made of multiple layers. I assume there is an inner layer with relatively good conductivity to feed the charge to the earth connection. Outer layers seem to be dissipative...you may safely put 230V circuits on it without too high current...not against earth and not against other devices on the mat. And not against your body...in case your body touches the mat.

If the mat is too conductive it increases the risk for damage caused by ESD high currents.
Imagine you carry an electonic equippment and place it on the mat...maybe your body (and the equippment) is charged by walking in the floor. Then you place the device on the conductive mat. The low resistance generates high currents. The situation more equals an ESD_machine_model than an ESD_human_body_model. --> The destructive energy is much higher..

Klaus

 

[betwixt]

Rule of thumb: if you push two DMM probes into a mat 10 mm apart and the resistance is less than 5M it is too conductive.

The idea of the mat isn't to ground the equipment, it's to prevent the build up of static charge and the series resistor to ground is to keep the residual voltage close to zero volts compared to surrounding objects, especially the operator, table top and floor.

I recall a post graduate student with a string of letters after their name but no practical experience spreading Aluminum foil across the workbench top, grounding it to to prevent static then placing a rather expensive processor board on it and plugging it in to a power supply. It proved a perfect example of the low melting point of Aluminum and why fire extinguishers should be on hand. Needless to say, the board was quite literally toast.

Brian.

 

[treez]

Rule of thumb: if you push two DMM probes into a mat 10 mm apart and the resistance is less than 5M it is too conductive.

Thanks, but does that only apply to the outer layers of the mat, which are high resistance?.....the inner core of the mat is a very low resistance to earth ground i thought?

I now wonder what our highly conductive mat is actually for? It has press studs on it for connection to the yellow earthing "plug"...so its obviously an "ESD Mat".

Do you believe that this highly conductive ESD mat is at least better than just using the varnished wooden surface of the desk to work on?

 

[c_mitra]

... using the varnished wooden surface of the desk to work on?

It is the varnish that is the culprit; not the wood. If you remove the varnish, you will probably not need any highly conductive mat...

- - - Updated - - -

... spreading Aluminum foil across the workbench top...

Aluminum foils have been advertised as cure all problems in real life...

Young engineers are very fond of shine and glitter...

 

[Kajunbee]

Do you have humidity control in the work area. As Cmitra as pointed out to me humidity is a major contributor to levels of static buildup. I looked into it and 40-60 % seems to be widely agreed upon. To high or to low humidity can also contribute to component failure and poor solder joints.

- - - Updated - - -

i would think when the furnaces are operating humidity control would be more of a problem. They seem to dry the air out even more. This never made since to me. I understand how the AC removes moisture from the air. The heater doesn't do this. It just circulate the same air over and over again.

 

[betwixt]

I now wonder what our highly conductive mat is actually for? It has press studs on it for connection to the yellow earthing "plug"...so its obviously an "ESD Mat".

There are applications where a conductive mat is required. For example some high voltage safety tests specify that equipment sits on a conductive surface while being zapped with several KV to test insulation. BSI caught me on that one years ago because I used a dissipative instead of conductive surface to flash test some enclosures. You can also safely use a conductive mat to assemble equipment, the golden rule is never to power it up on a conductive surface.

Brian.

 

[c_mitra]

They seem to dry the air out even more. This never made since to me. I understand how the AC removes moisture from the air. The heater doesn't do this. It just circulate the same air over and over again

Air conditioning in a part of the building removes moisture from the air. The cold air might have 90% RH value but actual moisture may be very less. The same hold true in the open in winter- the RH may be 80% but actual moisture content is low.

When the cold air is heated to a higher temp (say 27C), the moisture content does not change but the RH value drops to 40% or sometimes less. That might be problematic. Very simple hygrometers (often comes with room thermometers) are inexpensive and lasts practically indefinitely- they are very useful.

The higher the temperature (inside - outside) the lower is the relative humidity and the air gets drier. Humidifier is the only solution.

- - - Updated - - -

Do you believe that this highly conductive ESD mat is at least better ...

A conducting surface underneath a non-conducting layer acts as a giant capacitor: if you consider the basics, the charge and the voltage is actually present in the dielectric.

By grounding the conducting layer, you are actually making the problem worse. Just trying to sell a refrigerator to an Eskimo!

Well, you will ask the obvious: how to ground the beast? That is for the experts.

Again, I repeat: a constant high voltage is not damaging. Damage is caused when this voltage is discharged at a fast rate (high dV/dt causes differential potentials at different places within the device)

 

[treez]

Thanks , when we power the PCBs , they are on an earthed heatsink, which stands on top of the "conductive mat".
Sometimes when i am soldering an unpowered PCB, i have the PCB resting on the "conductive mat". I find it almost impossible to imagine that this conductive mat would be more of an ESD problem than a "dissipative ESD mat". The "conductive mat" is after all, connected to mains earth via the press stud.....i just cant see how its going to get induced up to 1000's of volts. I cant see how soldering a PCB which is resting on this "conductive mat" is going to mean ESD poblems.