Mosfet switch for DIY ESD gun

[pisoiu]

Not triggered.

I don't think  it will work without a trigger. At least not when you want to do something as close as possible to certification tests. Afaik, there are several ESD classes with various voltage levels. Cap needs to be charged at that level and discharged from that level. With a spark gap without trigger you cannot control discharge voltage, besides it can vary with humidity levels. And that for the air discharge test. Contact discharge would be impossible without a trigger.

[MichaelPI]

I don't think semiconductor switches are the right choice for an ESD gun. As far as I know, relays are used for that purpose. I attached a couple of pictures of the inside of an older Schaffner NSG ESD gun. The huge blue block should be the relay. The ball shaped tip of the ESD gun is used for air discharges (for contact discharges the tip is way more pointy). Dependent on the type of tests you want to perform, a trigger signal may not be required. For air discharges the discharge is introduced by bringing the ESD tip as fast as possible in contact with the DUT - you need to be fast, if the humidity is high, the tip could get discharges before you are close enough to the target. Professional equipment will tell you, that the discharge capacitors have already been bled out.
That usually means air discharge testing is way less reproducable because of the operator and the environment, besides that it is difficult to be fast and accurate. For contact discharge in which contact is made to e.g. a specific pin, there is really the need to switch the high voltage, if that is what you want.

Regards

[pisoiu]

My gut tells me that the huge block may contain the relay but also the HV generator. Nothing on the other board looks like a HV block and the connections between them are clearly low voltage type.
In the mean time I decided to use a vacuum relay, it was quite expensive but I need a good and fast solution. Therefore it cannot be cheap.
https://ro.mouser.com/ProductDetail/TE-Connectivity/K43C334?qs=sGAEpiMZZMtyK9GHuq88xIgA5rJPlzn8&countrycode=DE&currencycode=EUR
I also have been lucky enough to find on ebay a 500W adjustable power supply up to 30kv at good price, it's way more powerful than what I need, but at least I do not have to sped time building one.

[MichaelPI]

Yeah, you are right, after having a look at the ESD gun again - there is certainly more in the blue block than just a relay.

[TurboTom]

A hydrogen thyratron is a nice switch for an application like that. They are quite fast and permit repetition rates in the kilohertz range. the Russian TGI1-50/5 is available NOS for a fair amount in the "bay" and switches up to 5kV. With a suitable step-up transformer (air coil) and a pulse forming network, a very nice ESD generator can be built. These tubes are really evil...  But not for the faint of heart.

[PartialDischarge]

As I mentioned in a previous post, normally these voltages are generated by trigger type transformers, probably that is the blue box

[jonpaul]

Bonjour a tous, just see this now.

We designed and manufactured high voltage pulse ignitors for cinema arc lamps to 55kV and 12 kW, and did ESD an EMI compliance consulting in the 1980s..1990s.

The ESD pulses have well defined rise and fall time. Indeed a step-up pulse transformer is used to boost the capacitor discharge perhaps 3X ...10X.

The best switches are not the costly and power eating hydrogen thyratrons but much smaller cold cathode triggered spark gaps, krytrons, and enclosed gas gaps.

The transformers are like Tesla Coils with low capacitance and very few turns, using special insulated solid wire. The modules require vacuum potting with  a special  compound.

We used to get the parts and   ignitors from BAGTurgi in Switzerland and a specialist in Berlin, but that was decades ago.

Perhaps query Schaffner in Germany.

Hope this is useful info,

Bon Chance

Jon

[MichaelPI]

The blue box is certainly not a relay as I was thinking.  I checked the manual of a similar unit made by Schaffner (similar age / pretty old, lower voltage rating, air discharge tip). It is actually the high voltage generator (high voltage cascade).

If someone is interested in, I attached a photo of the schematic.

[pisoiu]

Thanks for the schematic. Looking at it, I'm quite sure that gun does not comply with IEC61000-4-2 or ISO10605 requirements so it's no more useful than a DIY piezo discharger from a regulatory point of view.
'HV Kaskade' I assume it's a diode/capacitor multiplier with a fixed polarity output. Both standards cited above requires test with positive and negative voltages, which is impossible with this configuration.
Not having a relay on output may be ok for air discharge tests but for contact discharge into DUT metallic parts or HCP/VCP, a discharge relay is a must. The installed tip is for air discharge. The other one, for contact discharge is more 'pointy'.

[MichaelPI]

The gun is older than the ISO 10605 standard, so it is not compliant. From a regulatory point of view any DIY tool and any tests you conduct, are basically worthless.
There is a chapter in the ISO 10605 in which the verification fixture for the pulse waveform is shown - just this thing combined with the required high bandwidth scope exceeds most probably the cost of the ESD gun. I would still prefer an old Schaffner NSG over a piezo discharger, if it's about precompliance.

[pisoiu]

Yep. I read about the test fixture. At some point I will have to test my device, after is completed. I do not have a DSO capable to measure that but I have a friend in a development company which has access to a capable $50k Tektronix. All I have to do is to convince him to let me discharge an ESD gun near it....

[jonpaul]

Rebonjour, use of expensive test equipment in ESD testing requires extensive shielding and protection, as the very fast rise HV pulses easily couple inductively or capacitively into probes, cables and input connectors.

We have blown up instruments this way.

Finalyy any electronics used to generate the ESD must be "bulletproof"

We used Transient absorbing Zeners, ferrite filters, calibrated gas spark gaps, metal can Faraday shielding, eg  in medical illuminator testing.

The cross coupling of HV pulses in a product or testng system can take weeks or months to debug.

If the product is for volume manufacture and shipping Internationallly, I suggest the OP find  a qualified laboratory with the required ESD probe and instruments to preform these tests.

USA: ETL, UL
EU: TUV, VDE

Kind Regards,

Jon

[pisoiu]

The final certification will be made at a qualified laboratory.
But the invoice exceeds 20k eur and I cannot risk going there without being as sure as possible that I will pass the tests.
This is why I need to do pre-compliance tests in house and I need tools as close as possible to those used for certification.