PTC Thermistor for doorbell?

[Gyro]

Wow, potentially up to 48W then, no wonder things got smelly!   Nearly 2W on the lamp too.

That's quite a switching load for the button contacts.

[james_s]

I found this which trips at 1.5A after 6 seconds

https://www.digikey.ca/en/products/detail/littelfuse-inc/RXEF075/5015922

Would it be suitable?

That's a polyfuse. I'd be looking at a PTC thermistor, not a polyfuse. The latter degrade fairly quickly after a few trip cycles.

[joeyjoejoe]

Hmm, Digikey only has PTC Resettable fuses... or sensor thermistors. What should I be looking for?

[Doctorandus_P]

The "PTC-resettable-Fuses" category seems to do the trick. There are many more devices there then the PPTC's, such as these:
https://www.vishay.com/docs/29087/ptccl265vseries.pdf

[joeyjoejoe]

I'm just thrown off by james_s comment that I picked the wrong thing... that was in the PTC Resettable fuses section. Is there a "type" i need to filter on or something? Or do I need to just read the literature for all the series?

[ArthurDent]

If the problem is a sticky doorbell switch, instead of adding complexity to such a simple circuit, why not just replace the faulty switch?

[Doctorandus_P]

The different types of PTC's seem to be intermixed in the same category.

The Plastic / or "Polymer" versions often start with 2 P's, as in: "PPTC".

https://en.wikipedia.org/wiki/Thermistor#PTC
https://en.wikipedia.org/wiki/Resettable_fuse

------------ 8<--------------- 8<--------------- 8<--------------- 8<--------------- 8<---
Why would anyone ever get the idea of replacing a faulty switch?
The Idea alone is just ludicrous!!!

[joeyjoejoe]

I did replace the switch. The goal is to prevent melting another doorbell. This switch was replaced less than 1 year ago by myself.

[joeyjoejoe]

Okay, so I want to find these PPTC's as these would be the desirable kind?

The different types of PTC's seem to be intermixed in the same category.

The Plastic / or "Polymer" versions often start with 2 P's, as in: "PPTC".

https://en.wikipedia.org/wiki/Thermistor#PTC
https://en.wikipedia.org/wiki/Resettable_fuse

[james_s]

If the problem is a sticky doorbell switch, instead of adding complexity to such a simple circuit, why not just replace the faulty switch?

I think adding a protection device is still a good idea. Devices have fuses and branch circuits have breakers not as a workaround for faults, but to prevent a fault from turning into something more serious. The next time the button fails and gets stuck there's no reason to burn out the doorbell and/or power transformer too.

[floobydust]

OP a car tail light like 'ol #1157 will also work, if you want to do this on the cheap.

[Gyro]

I'm just thrown off by james_s comment that I picked the wrong thing... that was in the PTC Resettable fuses section. Is there a "type" i need to filter on or something? Or do I need to just read the literature for all the series?

The Vishay ones that Doctorandus_P referenced are rated for 5000 trip cycles - I think that ought to see you out. 

[floobydust]

PTC fuses trip at lower currents after their initial trip. If I'm designing something with one, I will force them to trip a few times which gives a new hold/trip current value lower than a brand new part. I believe reflow also alters the crystalline structure so "surprise" they behave weaker in production.

[joeyjoejoe]

Okay, I'm gonna grab a PTCCL13H321HBE‎ which should trip in about 3 seconds of continous 2A current and call it a day!

[Doctorandus_P]

I was not clear in this, buy you do **NOT** want the PPTC variant, these have the lowest number of cycles.
When in doubt, just check the datasheets.

[james_s]

This is a very non-critical application though, lots of wiggle room, so don't stress over it too much. The very low duty cycle in normal use is a near ideal circumstance for this sort of protection scheme. If you find it eventually starts making the doorbell sound weak you can always try a different part. Nobody is likely to die if the doorbell stops working.

[joeyjoejoe]

Vishay is tested to 5000 trips. I think I'm good.

[Berni]

And the typical lithium phone battery is only rated to 1000 cycles (or even less for crappy ones) yet its still used everywhere.

If anything id say this is a better use for PPTC polyfuses than using them as an actual fuse. They suck as those as they lose quite a bit of voltage across them, they get hot and their response characteristic is kinda all over the place, especially if your device is expected to work over a wide temperature range. You would never want one of these to protect anything sensitive or critical, but they are the cheapest self resetting way to keep wires from melting and catching fire in case of a short circuit. But this slow lazy response is the perfect thing for this doorbell.

Sure you could solve it a lot more elegantly with a timing circuit and a relay and so on, but how much would that solution cost. Not only the BOM cost but the cost of time to figure it out, draw a design, order all the parts, put it together, test it etc... Time is money after all.

[james_s]

I'd argue that a PTC or PPTC is a more elegant solution than an electronic circuit. There is elegance in simplicity, it's a bit like the old $$$ fancy NASA space pen vs Russian pencil legend.

[mech_e]

Sorry to resurrect an old thread, but I'm in a very similar scenario as the original poster. Two follow-up questions:

1) Now that it's been a few years, how has this "upgrade" worked and would you do it differently?
2) For clarification, did you affix the PTC to the doorbell chime coil or to the transformer?

Any images would be helpful of how you attached the PTC.

And for the background story, I've got a decorative (read: expensive) doorbell button that the wife loves. For whatever reason, it has gotten stuck by the typical delivery man pressing it and running off. I think I've fixed the sticking issue, and then my wife pressure washes the house, likely gets the button wet, and shorts it again. So, now after two events, I don't want this to happen yet another time for whatever reason. I like this idea of a simple fuse to protect the equipment from overheating and melting.

Thanks - mech_e

[joeyjoejoe]

I put it with the transformer - no room near the doorbell.

Almost 5 years later, doorbell still working. Whether that's because of this device, or the button has never been stuck since, we'll never know

I can't remember how I connected it - probably with a Wago spring connector for simplicity.

[mech_e]

joeyjoejoe, thanks for the update and the info! It's far easier in my application to also put it with the transformer, so I'm leaning towards this option too.

[floobydust]

Did you guys use a Class II doorbell transformer? As I said earlier, these are energy-limited and have the primary thermal fuse. This change prevents house fires where the transformer or the wiring burned up. I had one old 1950's building do this, cotton covered wiring, cable staple or nail cut it etc. there was a fire.

An off the shelf transformer will have no fusing at all, so you must add it on both primary and secondary.

[mech_e]

My setup is pretty standard using a Heath Zenith 8V/16V/24V transfomer. I have a feeling this has a built-in fuse, which means maybe using the thermistor with the transformer may not do much since it's internal fuse may blow before getting too hot? Maybe the only real option then is to add the thermistor to the doorbell chime solenoid?

Or maybe I'm missing the point and you're saying to add two thermistors; one to the primary and then one to the secondary lines going to the doorbell chime from the transformer? I schematic would help me out since I'm pretty green with these types of components.

Thanks again!

[MarkS]

You're trying to fix the symptom and not the problem. Replace the doorbell switch. It should not stick.