Author Topic: High voltage non bouncing relay  (Read 1213 times)

0 Members and 1 Guest are viewing this topic.

Offline CloudTopic starter

  • Regular Contributor
  • *
  • Posts: 194
  • Country: si
High voltage non bouncing relay
« on: December 17, 2019, 11:05:03 am »
After spending quite some time with Schaffner 438 it has broken (finally  ;D) and because for repair they want almost what we paid for used one I figured that ESD Simulator would be a nice project for a final year of university. I figured out that 15kV would be maximum voltage that we need but I am having hard time finding relays that have enough voltage rating and are non-bouncing. Currently my only candidate is gigavac G15 but it costs 800+$  |O
Are there any cheaper alternatives. Somewhere I read that you can use relays from defibrillator...
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: High voltage non bouncing relay
« Reply #1 on: December 17, 2019, 11:50:37 am »
After spending quite some time with Schaffner 438 it has broken (finally  ;D) and because for repair they want almost what we paid for used one I figured that ESD Simulator would be a nice project for a final year of university. I figured out that 15kV would be maximum voltage that we need but I am having hard time finding relays that have enough voltage rating and are non-bouncing. Currently my only candidate is gigavac G15 but it costs 800+$  |O
Are there any cheaper alternatives. Somewhere I read that you can use relays from defibrillator...

Relays with mechanical contacts bounce. The only exception are mercury-wetted types, but those aren't available in HV ratings. In fact, at this voltage you are mostly going to find vacuum relays and they are breathtakingly expensive if purchased new.

Defibrillators operate at 5kV or less, so a single relay from one isn't going to cut the proverbial mustard, and while it is possible - if not entirely kosher - to stack relays in series to increase the voltage rating, you would still need to add RC networks across each one to improve dynamic voltage balancing (that is, to reduce the voltage overstress seen by the first relay to open or the last one to close).

Probably a better approach to producing low-energy 30kV pulses is dumping charge from a capacitor into a small flyback or xenon strobe trigger transformer (or, if you have the available space, an old automotive ignition coil - which is just a flyback transformer).

 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 5050
  • Country: si
Re: High voltage non bouncing relay
« Reply #2 on: December 17, 2019, 12:15:05 pm »
Ah hey there fellow Slovenian forum member.

At these sort of voltages you could think about making your own in the form of a trigerrable spark gap.

All you need to do is build a mechanism that brings a set of rounded contacts together very quickly, but not close enough that they touch. With a millimeter of space you can get a few kV to jump across it fairly reliably. This won't bounce since they are not touching and once the arc strikes lots of current will flow from your capacitor so the hot arc will continue conducting well under 1kV until it cools off enough to disappear.

I seen this switching mechanism also used for discharging big capacitor banks with >1kJ of energy since if the contacts are big and chunky they can handle massive amounts of power, and because they don't touch means that they wont weld together. Also the highest spot on the contact is most likely to arc and vaporize it away so the contacts also "self level" with use.
 

Offline CloudTopic starter

  • Regular Contributor
  • *
  • Posts: 194
  • Country: si
Re: High voltage non bouncing relay
« Reply #3 on: December 17, 2019, 12:57:27 pm »
Defibrillators operate at 5kV or less, so a single relay from one isn't going to cut the proverbial mustard, and while it is possible - if not entirely kosher - to stack relays in series to increase the voltage rating, you would still need to add RC networks across each one to improve dynamic voltage balancing (that is, to reduce the voltage overstress seen by the first relay to open or the last one to close).

Probably a better approach to producing low-energy 30kV pulses is dumping charge from a capacitor into a small flyback or xenon strobe trigger transformer (or, if you have the available space, an old automotive ignition coil - which is just a flyback transformer).

I was thinking about stacking wet relays but since you can get them to 2kV that would mean 8 relays.

I am afraid that with coil or transformer the shape of signal wouldn't be according to HBM.

Ah hey there fellow Slovenian forum member.

At these sort of voltages you could think about making your own in the form of a trigerrable spark gap.

All you need to do is build a mechanism that brings a set of rounded contacts together very quickly, but not close enough that they touch. With a millimeter of space you can get a few kV to jump across it fairly reliably. This won't bounce since they are not touching and once the arc strikes lots of current will flow from your capacitor so the hot arc will continue conducting well under 1kV until it cools off enough to disappear.

I seen this switching mechanism also used for discharging big capacitor banks with >1kJ of energy since if the contacts are big and chunky they can handle massive amounts of power, and because they don't touch means that they wont weld together. Also the highest spot on the contact is most likely to arc and vaporize it away so the contacts also "self level" with use.

That would be a solution but I need it to trigger from 250V to 15kV.
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 10173
  • Country: gb
Re: High voltage non bouncing relay
« Reply #4 on: December 17, 2019, 01:38:07 pm »
You could always try a surplus mercury displacement relay (used for heavy surge loads like tungsten lighting). eg. https://www.ebay.co.uk/sch/i.html?_osacat=0&_odkw=mercury+relay&_from=R40&_trksid=m570.l1313&_nkw=mercury+displacement+relay

They're normally rated up to 600V AC/DC but I bet they would stand off a good few kV in practice. No bounce.

Alternatively, check out former Soviet Vacuum relays (some bounce) but high voltage standoff. eg. https://www.ebay.co.uk/sch/i.html?LH_PrefLoc=2&_sop=15&_osacat=0&_odkw=vacuum+relay&_from=R40&_trksid=m570.l1313&_nkw=soviet+vacuum+relay&_sacat=0

Maybe cheap enough to try, given the cost of the alternative.
Best Regards, Chris
 

Offline daqq

  • Super Contributor
  • ***
  • Posts: 2321
  • Country: sk
    • My site
Re: High voltage non bouncing relay
« Reply #5 on: December 17, 2019, 02:35:36 pm »
Take a look at the goodies from Behlke, fully solid stage, fully non-bouncy, insane powers. A bit sensitive to short circuits, but should not be an issue with a series resistor.
Believe it or not, pointy haired people do exist!
+++Divide By Cucumber Error. Please Reinstall Universe And Reboot +++
 

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: High voltage non bouncing relay
« Reply #6 on: December 18, 2019, 12:54:44 pm »
Depending on the amount of energy you're looking for, a small hydrogen thyratron may also be an option. A device like a HY-2 / 7621 can be operated up to 8kV and will switch up to 100A at the size of an egg  ;)... This device, equipped with a small step-up coil and a pulse forming network, will produce really nasty, fast 20kV spikes. An ESD simulator based on such a device is still on my "to-do-list"... ;D
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf