2
\$\begingroup\$

If I have a zero-crossing SSR switching 240VAC (e.g. this LS240D12), will the zero-crossing detection function if the load side is shorted?

I know that zero-crossing SSR's can't be used with DC loads because the load never crosses zero. What I'm wondering is if with an AC load the SSR only cares that the voltage is zero, or if it actually needs the voltage to cross zero.

E.g. in the following schematic:

schematic

If the switch is already closed and the SSR receives an activation signal on the control side will it actually activate, or will it only activate once the switch is opened?

I'm asking because I'm hoping to use a combination of SSR and mechanical relay in parallel (per advice received elsewhere). The idea is to prevent arcing in the mechanical relay, and to prevent heat buildup in the SSR.

To turn on: SSR activates at zero crossing, small delay, mechanical relay activates, small delay, SSR deactivates.

To turn off: SSR activates, small delay, mechanical relay deactivates, small delay, SSR deactivates at zero crossing.

My concern is that the "turn off" procedure described above may not in fact reduce arcing in the mechanical relay if the SSR fails to truly activate without alternating voltage on its outputs.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ A hybrid relay is a good idea. And the selected SSR also uses an SCR-pair design (like I used with an MCU with software I wrote.) I can't tell from the datasheet, but I'm guessing that if the control input is held active, then the triggering circuitry will automatically retrigger at the start of each AC half-cycle. If so, then you want to activate it first, then activate the relay and wait long enough for the relay to fully settle down, then remove the control input to the SSR. I guess I'm not sure why your question about crossing vs at zero matters in this case, though it will be crossing. \$\endgroup\$ Jun 19, 2023 at 1:21
  • \$\begingroup\$ Thanks @periblepsis -- I'm concerned about the reverse operation of turning off the power (the "To turn off" operation above): the relay is already on/closed, then the SSR is "turned on" but in reality (judging from other answers here as well) it doesn't really turn on, as it sees a shorted load on the output (because it is parallel with the relay) so the triggering circuit doesn't activate. Then when the relay clicks open, there will possibly be arcing (unless the SSR extremely quickly turns on as the relay is opening.) Others suggested an instant SSR instead and that seems wise, despite EMI? \$\endgroup\$
    – user272901
    Jun 19, 2023 at 5:57

2 Answers 2

3
\$\begingroup\$

From the internal block diagram, we can see how it works:

enter image description here

The power to trigger the thyristor gates comes from current flowing through the load near the zero crossing.

So, if you apply the appropriate control voltage to the SSR then open the contact then the voltage could rise very quickly and not allow a long enough pulse to actually trigger the thyristors. I suspect it will be erratic and the erratic behavior will depend on the load, any snubber, and the exact time during the AC cycle that the contact happens to open.

Confining our thoughts to the linked datasheet, the random trigger type of SSR would likely be better for this application (there would be more EMI, but less than just a relay contact, and the thyristor power dissipation and relay contact life issues would be mitigated).

\$\endgroup\$
4
  • \$\begingroup\$ Thanks! -- so a random-fire SSR like the CX240D5R here: mouser.com/datasheet/2/657/… -- on page 4 it shows that the "R" means "Random Turn-On"... but given the schematic on page 3 (top left) do you think I can trust that the triggering circuit will work with a shorted load side? Seems like it might have the same issue of requiring AC on the load to power the triggering? \$\endgroup\$
    – user272901
    Jun 19, 2023 at 5:58
  • 1
    \$\begingroup\$ It should trigger the thyristors whenever the voltage across the contact is more than 10 or 20V, rather than trying to trigger in the narrow (in time) zone between (say) 10V and 20V. It’s not ridiculously narrow when the voltage is rising in a sinusoidal manner, but when the contact opens the voltage can rise very quickly. That’s my expectation anyhow. \$\endgroup\$ Jun 19, 2023 at 9:51
  • \$\begingroup\$ Ok so if I follow correctly, this means the relay will still have a chance of arcing, but since the SSR won't have to wait until the zero crossing a random-fire SSR (as opposed to zero-crossing) will lower the chance of serious arcing. I.e. a random-fire SSR might be the best I can do with a pre-made SSR module in a parallel relay setup like this? \$\endgroup\$
    – user272901
    Jun 19, 2023 at 19:53
  • 1
    \$\begingroup\$ The idea is that the relay will do very little arcing but there might be a bit more EMI than ideal. Yes, I think it's the best you can do with that constraint. Be sure to consider the failure modes. If the relay fails to close then the SSR will dissipate the full power since it is carrying the load current. It may fail short as a result. It may fail short as a result just because it feels like it. \$\endgroup\$ Jun 20, 2023 at 7:12
2
\$\begingroup\$

That data sheets says it has an SCR output stage, an SCR cannot be started with 0V across it.

As for whether shorted output will trigger the zero crossing detector it's impossible to tell. maybe it will maybe it will not. it's probably easier to use the immediate trigger version.

So the question now becomes will the SCR restart fast enough or stop slow enough to mitigate relay contacts bouncing. some of that depends on the trigger circuit. if the datasheet does not suggest using this SSR in parallel with a mechanical relay it may be unsuited for that task.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks -- does this mean that no SSR modules like the one I listed which use an SCR, even the "random fire" type, will work with shorted load contacts? Is there an alternative variety of random-fire (or zero crossing) SSR that can still operate with a shorted load? \$\endgroup\$
    – user272901
    Jun 19, 2023 at 6:03
  • 1
    \$\begingroup\$ nothing happens instantly, it depends how fast the SCR resets, if it's slow enough it will cover the bounces \$\endgroup\$ Jun 19, 2023 at 6:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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