LED Project

[tooki]

I already gave you a complete hardware-based solution.

[Jbouss]

Thank you Tooki, I see that, sorry, it took me a couple times to understand what you were suggesting. Your suggestion appears to be a great solution the more I think it over. My problem: I needed to throw away the idea of having the switches in series and tapping into a weak ~1.8V signal and go in a totally different direction. I guess scrapping my original idea is where I got hung up:(

To clarify, you suggest using a DPDT relay per switch that is held closed by 3.3V(something I know the switch can handle based on the circuit) and if a switch opens the relay flips to a 12V supply that is connected to my warning light?

I will draw it up as best I can, and post it when I am done. Again appreciate all the patients and help with this.

[Jbouss]

Thank you, everyone, you have all been so helpful! I believe the best solution is to use a SPDT type C relay NO. Parts are on order, I will let everyone know how it works out.

I decided to use only one control box to monitor the 10 switches. I placed a short on the other 9 boxes, so they will still run their respective systems without issue. With one control box monitoring all the switches, I get a solid 3V out and will attempt to use that signal to activate the relay. If that doesn't work, I will try the relay NC with 3.3V running through the switches. Either way, I think this solution will work.

Revised schematic attached.
[ Specified attachment is not available ]

[tooki]

Thank you Tooki, I see that, sorry, it took me a couple times to understand what you were suggesting. Your suggestion appears to be a great solution the more I think it over. My problem: I needed to throw away the idea of having the switches in series and tapping into a weak ~1.8V signal and go in a totally different direction. I guess scrapping my original idea is where I got hung up:(

To clarify, you suggest using a DPDT relay per switch that is held closed by 3.3V(something I know the switch can handle based on the circuit) and if a switch opens the relay flips to a 12V supply that is connected to my warning light?

Correct. Here's what it looks like (channels 1,2, and 10 shown, repeat the channel segment for the remaining channels):

Note that EVERYTHING is shown isolated: 3.3V is totally isolated, 12V is totally isolated, and the relay contacts to the control boxes are totally isolated. You could easily connect the 3.3V and 12V grounds together, but I made them separate so that you can see how you could use this setup for complete isolation, for example if you wanted to use a 120V AC beacon instead of a 12V one, or if it needed to control the input of another control box of some type.

[Edit: redid schematic to make the isolation even clearer.)

[tooki]

Thank you, everyone, you have all been so helpful! I believe the best solution is to use a SPDT type C relay NO. Parts are on order, I will let everyone know how it works out.

I decided to use only one control box to monitor the 10 switches. I placed a short on the other 9 boxes, so they will still run their respective systems without issue. With one control box monitoring all the switches, I get a solid 3V out and will attempt to use that signal to activate the relay. If that doesn't work, I will try the relay NC with 3.3V running through the switches. Either way, I think this solution will work.

Revised schematic attached.
(Attachment Link)

That's a completely different approach unrelated to both your original and my proposed solutions. Can't say whether it'll work since the control box's function is unknown to me.

Is defeating the pressure switches on the other 9 control boxes wise? What are the safety and process ramifications of that? It's not something I'd do without FULLY understanding ALL the risks it brings.

My solution is fully safe, under the assumption that a pressure switch opening is the fault condition: if any switch fails, or the 3.3V power supply fails, or any relay coil fails, the control box sees the pressure switch go open. (The only fault condition not sensed here is relay contacts welding shut, which is extraordinarily unlikely to happen with just a simple 3.3V logic signal across them.) Failure of the 12V power supply causes the LED beacon to fail, but the control box would still see the pressure switch go open.

[Jbouss]

I really appreciate the schematic, I was having a hard time seeing the DPDT, and now it is much clearer.

Is defeating the pressure switches on the other 9 control boxes wise? What are the safety and process ramifications of that? It's not something I'd do without FULLY understanding ALL the risks it brings.
I am 100% that this is OK to do, I do it all the time for other tests. Thank you for verifying.
In fact, I use another set-up that uses only 1 control box to verify 10 switches for a different application, let's call that set-up B. I think I will test my solution on set-up B first.

For the 10 control box set-up, let's call that set-up A, I would have a single tower light like discussed but ideally there would also be smaller indicator LEDs for each control box, making troubleshooting a failure easier. I believe your schematic allows for this addition. This was issue #2 in my original post, but that really made things confusing, so I eliminated that idea for the time being. I have come a long way since then; this forum is a great resource. I am glad I decided to reach out, and look forward to following the discussion board.

I will post updates as my project progresses.

[tooki]

Yep, here's a bunch of places you could add indicator LEDs: one for each power rail, and then per channel one for the relay coil, one for the 12V relay contact, and one for the signal contact (assuming the control box can supply enough current for an LED.)




(Note: to avoid having to redraw the whole thing, I've shown the LEDs without resistors. In reality, every LED requires in series with it a current-limiting resistor appropriate for the supply voltage.)

[Cerebus]

It's worth pointing out that there are implied current limiting resistors in there for all those LEDs, and there's one vital component missing:



Without a flyback diode on those relay coils one is very likely to exceed the reverse breakdown voltage of those LEDs in a destructive fashion. Reverse breakdown voltages on LEDs are low, typically 5V or so, the reverse voltage pulse from a relay coil can easily exceed hundreds of volts. Also there's a minor potential for turning all the long wires involved into quite effective antennas at RF as they see the reverse voltage transient pulses from the relay coils as they're turned off.