Feature Request Consideration - Relay Operability Check & Relay Failsafe Runaway Protection

Hey dev team!

I think of this as a "relay failsafe runaway protection" to prevent unregulated current to designated circuitry. It would require logic to be developed and some minor circuit addition. I'm a mechanical engineer, but not very knowledgeable in the software development.

The common problem:
I've been slow to want to upgrade to 110v heated bed, however I am about to do it. My main concern and probably many others' as well is the relay- electromechanicals don't play nice with PID, and SSR's typically fail 'closed' from my understanding. That last part is what I'm concerned about- even if my printer stops and signals a problem and stops the print and cuts power, the current still flows, right? That could still overheat the elements and cause a hazard or fire quickly. Perhaps I don't know enough about this part- but as I understand it at the time I'm writing this, even cut signal to the MOSFETS/SSR's, unregulated current still flows.

Solution proposal:
In my mind, it seems feasible to do a repeating 'operability check' of a SSR and install a 'failsafe' electromechanical relay ahead of the SSR (or entire system) that would be maintained 'closed' while the logic is reading safe. This could be multiplied out for all hot ends. I imagine it could be integrated to protect boards with onboard MOSFETS as well that are prone to failure.

The way I see it working would be like so:


1)
a) A feedback circuit is placed on the high voltage end of the SSR to the bed (reads voltage/no voltage and gets fed back into control board).
b) An electromechanical relay would be set to supply power to the SSR (or entire system, but I'm not sure how it would initially power on)- this electromechanical relay would be turned on via the control board as soon as the control board is turned on and runs a logic check.
2) Feedback confirmation from 1a would be checked every time the SSR is actuated. Control board would cycle the switch intentionally off/on quickly every 30 seconds for long draws to confirm relay hasn't failed during heatup. Good signal confirmation maintains signal to electromechanical relay, keeping power supplied. This would be operation as normal, and electromechanical relay would typically be switched "on" and "off" only as many times as the main control board is, retaining it a typical long life.
3) On an unexpected signal from 1a (signal doesn't change to 'false' in certain about of time (.5 sec?) when control board issues power cut signal to SSR, the control board immediately knows the SSR has failed or is failing. This can send it to immediate and full shutdown mode (or user select- make noise and allow user to override with a click?). This mode would cut signal to electromechanical relay ahead of SSR (or entire system). (With EEPROM enabled, EEPROM could be used to store a message to the user?)

These steps are over simplified but I wanted to present the concept without trying to hammer out all the details. Am I way off base here, or is this something that might be doable and practical?

Thanks for your consideration, and thank you all for your enormous contribution to the maker world!

Edited 1 time(s). Last edit at 02/25/2020 01:20PM by awinn17.

Not exactly what you're asking, but you might want to take a look at SUICIDE_PIN. This is a pin that gets switched whenever a kill condition occurs - for example in case of a thermal runaway. You could connect your relay to the SUICIDE_PIN.

You could also use the PS_ON pin in a similar way - switch your primary power supply (12V or 24V) using PS_ON. Use the output of the power supply to power the relay. If a thermal runaway occurs the PS_ON pin will be toggled, shutting down the power supply and therefore the relay.

That is totally new to me. I thought I'd looked into everything safety related. That is indeed very similar and could be utilized to a similar end. Thanks!