See if this diagram helps make sense of the latching leakage path.

On the left, both controller outputs are closed, so both relays close and all works as expected.

On the right, controller number 2 has opened its output, but the voltage from the left side of relay 2's contacts leaks through (the blue path) which continues to apply power to the whole right leg which is meant to be controlled by controller 2.

With power still present in this leg on the right, relay 2's coil is powered, so the contact doesn't release. Since the contact doesn't release, the leg on the right is still powered. Can you tell that this paragraph is a recursive loop that never ends?!?! It's latched!

Once this latching state has occurred, the only end to it is if BOTH controllers open their contacts at the same time. That would remove the source of power for the latching behavior.

 

Thanks! It makes perfect sense now.

I wish I would have seen it coming. I spent a bunch of money and jumped through a bunch of hoops because I didn't have a grip. I rewired per Albert's instructions and now all seems well in my little refrigerated world.

I'm glad I found this place and I'm glad you folks were kind enough to help me fix the problem AND help me understand what the heck was happening.

I couldn't have done it it without you.

 

You're welcome and if it's any consolation the problem isn't obvious from the diagram.
Thanks for the beer!

 

Virtual beers all around!

 

You're welcome and if it's any consolation the problem isn't obvious from the diagram.
Thanks for the beer!

Yeah, good catch on that one. I had looked it over several times and didn't see the issue until you pointed it out. Well done!

 

The circuit in post 15 does indeed remove the unintentional latching circuit, and also reduces the loading on the contacts of the controllers. I just now saw this thread or I would have responded sooner. I am guessing, also, that those are not just LED indicators connected across the AC line.
What would have made the problem much simpler to diagnose would be to have drawn the circuit in a "ladder diagram" format instead. Most industrial control circuits are drawn that way and it makes understanding the operation much easier. It is also called "ladder logic", if you want to research it. It is a quite different style of circuit presentation.

 

The circuit in post 15 does indeed remove the unintentional latching circuit, and also reduces the loading on the contacts of the controllers. I just now saw this thread or I would have responded sooner. I am guessing, also, that those are not just LED indicators connected across the AC line.
What would have made the problem much simpler to diagnose would be to have drawn the circuit in a "ladder diagram" format instead. Most industrial control circuits are drawn that way and it makes understanding the operation much easier. It is also called "ladder logic", if you want to research it. It is a quite different style of circuit presentation.

Well, this is just my opinion, but I'm glad it was drawn with a circuit schematic like I'm used to, not ladder logic. Ladder logic may be more fitting for this project, but schematics are the universal language of this forum, and it seems to me like a lot more people can participate and help out when schematics are used. Maybe I'm way off base here, or maybe I'm just biased because I don't already know ladder logic!

 

Industrial control schematics, even with the older relay logic were usually drawn in ladder format.
Once used to it, it is generally much easier to follow a schematic that way.
Max.

 

Well, this is just my opinion, but I'm glad it was drawn with a circuit schematic like I'm used to, not ladder logic. Ladder logic may be more fitting for this project, but schematics are the universal language of this forum, and it seems to me like a lot more people can participate and help out when schematics are used. Maybe I'm way off base here, or maybe I'm just biased because I don't already know ladder logic!

Ladder logic is just a different arrangement of the circuit schematic, and you would have no problem following it if you are able to follow the one in the post. While the name is intimidating, the actual application was originally to help the less experienced follow the circuit functions.

 

Ladder logic is just a different arrangement of the circuit schematic, and you would have no problem following it if you are able to follow the one in the post. While the name is intimidating, the actual application was originally to help the less experienced follow the circuit functions.

Cool, I've been meaning to look into it and just hadn't gotten around to it. I'll check out some ladder logic and see how things go. Thanks!

 

I'm lucky I found someone to cad draw the lame schematic I posted. Since then I've added another relay controlled by the output of the refrigerator control that interrupts the outputs of the humidifier and dehumidifier while the refrigerator is running.

 

I'm lucky I found someone to cad draw the lame schematic I posted. Since then I've added another relay controlled by the output of the refrigerator control that interrupts the outputs of the humidifier and dehumidifier while the refrigerator is running.

I wouldn't call that schematic lame by any stretch. There's a bit of an art to making schematics as clear, clean, meaningful, and easy to understand as possible, and nearly every schematic involves compromises and leaves room for improvement in some way or another.

The schematic you provided captured your circuit well enough for us to see what was going on, so that's a success in my book.

If you have a continued interest in creating your own original designs, it might be good to learn more about creating your own schematics, and there are a variety of free programs that make it reasonably easy to do so. Either way, you certainly don't need to worry about what you and your friend provided here.

 

Indeed! Your original schematic was not lame. It was accurate although it did not point out how things could interact. That is why I suggested the ladder format.
There is a lot to creating drawings that help the folks building the device avoid errors, as I have discovered over the years. some things that are conventions, just make the flow of signals easier to follow, while others are helpful in relating a drawing to physical reality. Sort of like writing for others to read, in that some writers are easier to follow than others. ( I hope this makes sense to you, I know what I mean)