Safety relay for 5500W heating elements

[stephelton]

I'm working on an electric brewery that is entirely controlled by a Raspberry Pi. I can control a 45A 240V SSR from the pi without a problem. However, I'm concerned about safety with respect to the SSR's leakage current and the potential of it to fail (as discussed on theelectricbrewery.com).

I'd like to use a mechanical relay, also controlled from the Pi, so I'll need something along the lines of:

- Raspberry Pi's 3.3V GPIO pin controls a transistor
- transistor directs 5V to a mechanical relay capable of switching 30A @ 240V
- mechanical relay allows SSR to power heating element

I'm having a difficult time finding a suitable mechanical relay. I'm wanting a SPST NO relay with coil rating of 240V 30A that can be operated with 5V.

Is this simply impractical? Perhaps the coil for such a hefty contact would require a lot more than what the Raspberry Pi can provide (5V and < 1A)? In that case, I suppose I'll need yet another relay to send either 12VDC or 120VAC to the larger relay's contacts...

If it's not clear by this point, I'm not very experienced in electrical engineering, so I'd appreciate any input

 

[beluedog]

I'm having a difficult time finding a suitable mechanical relay. I'm wanting a SPST NO relay with coil rating of 240V 30A that can be operated with 5V.

Is this simply impractical? Perhaps the coil for such a hefty contact would require a lot more than what the Raspberry Pi can provide (5V and < 1A)? In that case, I suppose I'll need yet another relay to send either 12VDC or 120VAC to the larger relay's contacts...

I dont think you will find something like that

I would use 1-channel 5v Relay Module For Arduino or similar to control a low amperage 120V Hot to energize the 220V 30A contacter

make sense?

 

[stephelton]

yeah, that's pretty much what I was getting at... so one more relay in the whole mix...

 

[beluedog]

the more the merrier

 

[jeffmeh]

A few questions/comments:

1) Why a SPST contactor, rather than a DPST contactor to cut power to both hot legs?
2) Generally, the purpose of the contactor after the SSR is to be able to be certain that there is no power to the element. A NO mechanical contactor can provide this certainty, if you can be certain that the coil is not energized to close the contactor. You have to be able to trust anything on the input side of the coil to behave correctly, so using an SSR (that can fail closed) to control the coil of the mechanical contactor defeats the purpose. Why not use a mechanical switch to control the contactor, so you can be sure the element is off when you want it off?

 

[stephelton]

1) Why a SPST contactor, rather than a DPST contactor to cut power to both hot legs?

Good idea.

2) Generally, the purpose of the contactor after the SSR is to be able to be certain that there is no power to the element. A NO mechanical contactor can provide this certainty, if you can be certain that the coil is not energized to close the contactor. You have to be able to trust anything on the input side of the coil to behave correctly, so using an SSR (that can fail closed) to control the coil of the mechanical contactor defeats the purpose. Why not use a mechanical switch to control the contactor, so you can be sure the element is off when you want it off?

I'm not controlling the mechanical relay with the SSR. Both relays would be independent and controlled with different circuits (effectively, different GPIO pins). The end result is that both relays must be energized for the element to be turned on. The mechanical relay would only be energized when the software knows it's safe for the element to be turned on, while the SSR would be used to control the duty cycle of the element (and not wear down the contacts of the mechanical relay).

 

[stephelton]

What if I took a different approach and attempted to detect when the SSR has failed? If I could detect 240V when I know that the SSR is off, I could display an error to the operator explaining that a SSR failure has been detected.

I still need some way to shut off the 240V in that case, and that still doesn't prevent leakage current from doing any damage...

What kind of device would allow me to do this? I'd essentially need to be able to read 3.3V on yet another GPIO pin...

 

[stephelton]

How bad is the leakage current, anyway? Wikipedia says it's on the order of micro-amps. Even if it was 1 milli-amp, we're only talking about 0.24W, which doesn't sound very dangerous to me. Should I be concerned about leakage current?

 

[jeffmeh]

Good idea.



I'm not controlling the mechanical relay with the SSR. Both relays would be independent and controlled with different circuits (effectively, different GPIO pins). The end result is that both relays must be energized for the element to be turned on. The mechanical relay would only be energized when the software knows it's safe for the element to be turned on, while the SSR would be used to control the duty cycle of the element (and not wear down the contacts of the mechanical relay).

OK. I was responding to the Arduino relay suggestion someone else made earlier, as I assume that is a solid state relay (SSR) rather than a mechanical one, and subject to leakage and failing closed. The latter is much more concerning than the former.

 

[beluedog]

OK. I was responding to the Arduino relay suggestion someone else made earlier, as I assume that is a solid state relay (SSR) rather than a mechanical one, and subject to leakage and failing closed. The latter is much more concerning than the former.

The relay I was suggesting is mechanical 5v dc coil switching 120v @ 10amps

 

[jeffmeh]

The relay I was suggesting is mechanical 5v dc coil switching 120v @ 10amps

That would work well, assuming of course that no glitches in any of the software, firmware, or solid state components of the Pi inadvertently power on the relay, lol. So the Pi sends a 5v DC signal that energizes this relay, which in turn sends an 120v DC signal to a 120v coil, 240v AC DPST relay, which closes so that the Pi can control the element through the SSR. Pretty neat.

 

[stephelton]

So the Pi sends a 5v DC signal that energizes this relay, which in turn sends an 120v DC signal to a 120v coil, 240v AC DPST relay, which closes so that the Pi can control the element through the SSR. Pretty neat.

You missed one part -- the Pi controls a transistor, which allows 5v to power the relay Actually, it does this for both relays. The GPIO pins that are controllable from the Pi are 3.3V.

So basically, we go:

(Safety Switching)
3.3V -> transistor -> 5V -> light duty mechanical relay -> 120VAC -> heavy duty mechanical relay -> 240VAC -> heating element circuit

and
(Duty Cycle Switching)
3.3V -> transistor -> 5V -> heavy duty SSR -> 240VAC -> heating element circuit

Only if both of these chains are energized does the element fire.

 

[jeffmeh]

You missed one part -- the Pi controls a transistor, which allows 5v to power the relay Actually, it does this for both relays. The GPIO pins that are controllable from the Pi are 3.3V.

So basically, we go:

(Safety Switching)
3.3V -> transistor -> 5V -> light duty mechanical relay -> 120VAC -> heavy duty mechanical relay -> 240VAC -> heating element circuit

and
(Duty Cycle Switching)
3.3V -> transistor -> 5V -> heavy duty SSR -> 240VAC -> heating element circuit

Only if both of these chains are energized does the element fire.

OK. I would clarify on the safety switching that both hot legs to the element go through the DPST mechanical relay, and on the duty cycle switching only one hot leg goes through the SSR. Looks good to me, FWIW.