The device produces output exceeding 55kV in short pulses. The current while claimed to be low by the manufacturer, measures three to four amps in pulses of about 100nS every time the device fires. I measured this with a Tektronix current probe and oscilloscope. They may be intending to say that the AVERAGE output current is low.

The electrode tips that they sell have an air gap insulator section which insulates the low voltages (110 or 220V) from the tip but passes the high voltage sparks with little loss.

The high voltage output is required for it to function in its intended application.

The manufacturer does not seem to have UL or CE claimed anywhere.

Also, how would you test the safety of this device, and what standards would apply?

  • 3
    \$\begingroup\$ Safety standards have carve-outs for things that are necessarily somewhat unsafe in order to do their jobs. For example, you wouldn't have to prove a gas range can't start a fire, but you might have to show that it won't turn itself on unless its knobs are turned, or that it won't leak gas when it's off. \$\endgroup\$ – The Photon Jul 3 '17 at 0:26
  • \$\begingroup\$ Then why go to the trouble of producing the tips with the plastic spark gap? It seems like they are trying to comply with a standard but I don't understand which one. \$\endgroup\$ – user103218 Jul 3 '17 at 0:57
  • \$\begingroup\$ TBH I don't even know what a "vacuum leak and holiday detector" is or how it works. But presumably different tips are used for different circumstances or they wouldn't offer so many different tips. \$\endgroup\$ – The Photon Jul 3 '17 at 0:59
  • 2
    \$\begingroup\$ Page 5 of the manual talks about what safety directive is followed and why there is no CE mark. \$\endgroup\$ – The Photon Jul 3 '17 at 1:05
  • \$\begingroup\$ Ok, so the tip then presumably has to meet some requirement in EN61010-1? \$\endgroup\$ – user103218 Jul 3 '17 at 1:09

How can a device like the ETP BD-10A pass electrical safety testing?

  • First what does the manual say? ( see below)
  • Second what is a relevant standard document that defines these requirements?
  • EN61010-2017 $726

BD-50EV User Manual ------

The Model BD-50EV generates a high voltage corona of approximately 500 kHz. However by the very nature of its design, it will produce electromagnetic interference (EMI) as a result of its operation. Electric arc welders, for example, are another product that by its very nature and mode of operation produces EMI.

As a result, the Model BD-50EV cannot meet the European Union Electromagnetic Compatibility (EMC) Directive 89/336/EEC, and cannot be CE marked. It does, however, meet EN61010-1:1993 Safety Requirements for Electrical Equipment for Measurement, Control and Laboratory Use, following the provisions of the Low Voltage Directive 73/23/EEC, as amended by 93/68/EEC

I see that this project detects vacuum leakage by an arc detection of a defined gap and voltage (50kV) . I suggest all you need is 5kV by reducing the 1.4mm gap or less using a tungsten tip sparkplug to achieve 500V PD @ 1Atm and thus >5kV PD or BDV @ <0.08 Atm. PD is the precursor of BDV and behaves like a unijunction discharge before the pC (picocoulomb) arc discharge across the gap. Sparkplugs are also very inexpensive and have tungsten tips for longevity.

Partial Discharge (PD) is a "Type Test" for high voltage insulation that is a better way to detect vacuum or contaminants than BDV. ( Research will tell you much more about this topic)

See Paschen's curve below which I recall applies to parallel smooth gap at with >300V arc @ 0.1mm or 3kV/mm. Raising either air pressure, or vacuum increases the breakdown voltage (BDV) while changing to sharp pins reduces the BDV by 50%. Dust or contaminants may reduce more and the purest insulator dielectrics better than air may raise BDV it up to 10~30x air BDV.

Thus below 0.080Atm BDV rises sharply from 10x BDV @ 1Atm so depending on your vacuum failure threshold a smaller gap than 3kV for a typical sparkplug, by reducing the gap to 500V then use 5kV for a threshold of >0.08 Atm.

Electrode variations of shape and sharpness and contaminants can cause variations and dielectric contaminants may cause Relaxation Osc. Partial Discharge. (PD) start around 1 pulse/minute and rising with voltage or loss of vacuum.

enter image description here

Having done these types of tests before you must limit the discharge energy and ensure there are no explosive particles or hydrocarbon fluids that can generate explosive gas (H2 and methane).

Your only obligation is HIPOT and LEAKAGE testing for AC powered instruments and a safe insulation ground, EMI shield and safe wiring & procedures. Commercial Instruments used by qualified personnel have more relaxed requirements. ESD tests are advisable for immunity and safety with 100 discharges to any point outside the box up to 7kV no fault and 15kV no damage.

This is what I would use.

Tungsten tip, threaded nuts with gasket. 6 threads min. Test BDV at 1 atm. then test at Vaccum fail threshold and choose that voltage.

Here 1.6mm gap in air 1 atm., PDIV = 3kV+/-0.1kV BDV slightly higher or same if clean. Then use a full length of auto. carbon conductor sparkplug wire to limit current and EMI. enter image description here

If you want to read new product compliance specs, EN61010-2017, you have to buy it.

1.2.1 Aspects included in scope

The purpose of the requirements of this standard is to ensure that HAZARDS to the OPERATOR and the surrounding area are reduced to a tolerable level.

Requirements for protection against particular types of HAZARD are given in Clauses 6 to 13, as follows:

  • a) electric shock or burn (see Clause 6);
  • b) mechanical HAZARDS (see Clauses 7 and 8);
  • c) spread of fire from the equipment (see Clause 9);
  • d) excessive temperature (see Clause 10);
  • e) effects of fluids and fluid pressure (see Clause 11);
  • f) effects of radiation, including lasers sources, and sonic and ultrasonic pressure (see Clause 12);
  • g) liberated gases, explosion and implosion (see Clause 13). Requirements for protection against HAZARDS arising from REASONABLY FORESEEABLE MISUSE and ergonomic factors are specified in Clause 16.

RISK assessment for HAZARDS or environments not fully covered above is specified in Clause 17.

NOTE Attention is drawn to the existence of additional requirements regarding the health and safety of labour forces.

  • \$\begingroup\$ I'm not really interested in the rationale for using 50kV. That is what it is. The output tips on their web site all have a plastic insulating section which contains a spark gap. The obvious purpose is to let out the high voltage (not actually dangerous) while blocking low voltage current from the outlet. The output socket is connected through a coil of about 70 ohms to the line neutral. So, this spark gap has to meet some standard, but which one would it be, and what would the pass/fail limits be? For example, if it's shorted it's an obvious fail but what if it leaks a microamp at 1kV? \$\endgroup\$ – user103218 Jul 12 '17 at 13:57
  • \$\begingroup\$ I believe all you are asking for is IEC 61010-1:2017 which costs $$$ \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jul 12 '17 at 14:13

You're slightly misunderstanding safety testing. Some hardware is tested to be safe only to the user. The common Glock pistol and electric chairs are extremely safe for the operators. They will have been tested in multiple scenarios. Admittedly, the Glocks will have had much much more extensive testing. And electric chairs probably don't have any formal certifications as they're half arsed Frankenstein creations, but I'm not sure. For your gadget, 55KV is pretty low on the spectrum.

CE marking only applies to European Free Trade Area & European Union countries. It doesn't apply to the US, unless they want to sell this device in the aforementioned area. The manual clearly states that it passes Low Voltage Directive 73/23/EEC, whilst failing the EMI regulation. So it can't have a CE mark, and my understanding is that it can't be legally sold (or imported) to the EU. But it's being sold in the US and not here anyway. CE compliance isn't really on topic here anyway as we're not Eurocrats /solicitors.


In terms of actual safety rather than safety regulations, high frequency signals such as this are much less of an electrocution concern due to the Skin effect. High power RF is more likely to cause burns rather than electrocution as the current will tend to flow on your skin rather than through your heart. In this case, it sounds like it generates short pulses with a low duty cycle, so I would expect that the overall danger is relatively low. The pulse energy and average power are both fairly small even if the peak power is rather high. That doesn't mean that it is completely safe or that you should touch the electrodes, but you may be overestimating how dangerous it is.

As far as safety regulations go, that is somewhat off topic here, but mostly what The Photon said. Safety regulations usually have mechanisms for exceptions based on what a device needs to perform its function. So you can make a high voltage power supply, but you need to take reasonable precautions to make sure that an operator doesn't accidentally touch the HV terminals. Depending on the application, that might involve an interlock, a mechanical shield, a fault current sense, or simply a warning label. You still need to provide proper mains isolation: just because your device is a high voltage generator doesn't mean you get to skip safety requirements that are not part of your devices operation.

  • \$\begingroup\$ I suggest you work the equations for skin effect at 500kHz, and with the rough impedance of meat, then get back to me. \$\endgroup\$ – user103218 Jul 8 '17 at 16:02
  • \$\begingroup\$ There is no need to be snarky when you are asking for help, and if you already know all the answers, don't bother asking questions at all. \$\endgroup\$ – Evan Jul 8 '17 at 17:18
  • \$\begingroup\$ The skin effect statement is ludicrously wrong, and totally irrelevant to my question. \$\endgroup\$ – user103218 Jul 8 '17 at 18:01
  • \$\begingroup\$ The OP said that it would be relevant, and I suggested he do the math. How is that snarky? In any event, I am looking for the rationale and regulations behind the plastic spark gap, which they put on every one of their output devices, so there must be a reason for it. The only apparent reason is to prevent the user from coming into contact with the internal wiring, which is nominally the neutral lead of the plug, but of course exposing a user to contact with neutral isn't acceptable practice. \$\endgroup\$ – user103218 Jul 8 '17 at 18:09
  • \$\begingroup\$ Then you question is not clear. If your question is "why do they have to prevent contact with mains wiring" the answer is because that is required. \$\endgroup\$ – Evan Jul 8 '17 at 18:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.