Mechanical relay death in stuck in closed postion bed at 120c

Dougal1957

I use one from this range manually resettable thermal cutout

HTH

Doug

claustro

Thank you
Did you mount it with screw or silicone glue under the bed?
Should this works?
https://www.ebay.co.uk/itm/150-C-Nc-Termostato-Temperatura-Interruttore-Termico-Normalmente-Chiuso/263953401277?hash=item3d74d945bd:g:KNsAAOSwBahU-3-6:rk:34:pf:0

Stephen6309

@claustro My bed is a 110vac silicon heater from McMaster-Carr stuck to a 1/4" MIC6 plate. The thermal fuse is similar to these:
https://www.ebay.com/itm/10pcs-167-C-TF-Cutoff-250V-10A-12-4mm-L-D-Thermal-Fuse-Electric-Cooker-Fuses/183188846355

For the mounting, I used kapton tape to hold in place on the heater.

dragonn

That is exactly why you should use a mechanical relay for driving you bed... bad idea overall.
You have 24V, why you don't connect it directly too you Duet?

claustro

First of all thank you all for answering me.
some questions.

@ dragonn I used a mechanical relay but it failed in on position form what you are writing i can't understand if you judge a mechanical relay reliable or not.
I drive the bed by relay because it has a dedicate power supply separated from duet. ( 30Amp + 20Amps)
Also I was afraid it drain to much current for my duet that is 1.2 version and the bed is 300x300

The great majority of thermal switch are rated 10A at 240 will be this specs sufficient for the current drain from a kk3 heated bed?
Tomorrow I'll measure the exact current drained from the bed

dragonn

@claustro mechanical relay are just not designed for this use-case. They will fail fast when doing a lot of on-off cycles.
You should be an external mosfet module like for example https://www.aliexpress.com/item/3Dprinter-heat-control-MKS-MOSFET-for-heated-bed-printer-head-MOS-30A/32405884519.html they are much easier to get then SSR for DC and are reliable too.
And this will work with separated power supply too, only the ground of the power supplies need to be connected together (with this mosfet module probably does already).

AndreS

Those MOSFET-modules have an opto-coupler at the input. No need to connect minus of the power-supplies together. 2 pins from the power-supply, 2 pins to the heatbed, and 2 pins from the duet heater output. The last two can be thin cables, because the are only a few miliamps for the opto-coupler needed.

claustro

Is this connection scheme correct?
I bought an overrated thermal sensor but I prefer be ready in case of upgrade.

AndreS

From Duet to MOSFET-module use the upper terminal (bed+ and bed-). From power-supply to MOSFET Power- is at bottom, then Power+.

Mangy_Dog

Im using a mains heated Silicon matt on mine.... Ive put three levels of protection and control. A mechanical relay, that would fail open (ie off)
This is the the first layer, this is triggered by PS_ON, PS_ON will also kill power when thermal runaway is detected in software. This could happen if a therm has failed or if the SSR fails and is stuck on, The actual PWM power control is done by the SSR from the heat bed pin.

Using an SSR for the PWM protects the mechanical from wear, as these are designed to have fast power cycles, its basically a power transistor...

After that, a thermal fuse on the bed, if the software crashes and for some strange reason PS_ON is stuck on and heat signal is also stuck on (very unlikely but in the realms of possibility) the fuse will trigger when the temp hits around 140 with the fuse ive got.

So in this setup its incredibly unlikely a mains heat bed will cause a fire. Im also thinking of just adding a power fuse in the line as well to protect against possible shorts if anything damaged the power line in there... Though there is going to be 2 fuses, one on the kettle socket and one in the plug socket...

I was just thinking a lower rated fuse just for the bed. maybe 4 or 5 amps... My bed is 3 amps max... So anything above that would be a short or something else failing...

dc42

@mangy_dog said in Mechanical relay death in stuck in closed postion bed at 120c:

Im using a mains heated Silicon matt on mine.... Ive put three levels of protection and control. A mechanical relay, that would fail open (ie off)

Mechanical relays can also fail short circuit, due to either a mechanical problem (rare) or the contacts getting welded together, which is probably what happened to the relay that @claustro was using to control his DC bed heater.

...Using an SSR for the PWM protects the mechanical from wear, as these are designed to have fast power cycles, its basically a power transistor...

A DC-AC SSR uses a triac, not a power transistor. Triacs are very reliable, unless they are subjected to a prolonged current overload (causing them to overheat) or frequent current surges.

After that, a thermal fuse on the bed, if the software crashes and for some strange reason PS_ON is stuck on and heat signal is also stuck on (very unlikely but in the realms of possibility) the fuse will trigger when the temp hits around 140 with the fuse ive got.

A agree, a thermal fuse is a good protection mechanism to have.

Mangy_Dog

1 out of three isnt that bad

I would say though a mechanical welding closed is incredibly unlikely... The mains fuse will pop before a short could weld a relay im sure...

Triacs, ok lack of a better word.... But yes Triac not a transistor...

dc42

@mangy_dog said in Mechanical relay death in stuck in closed postion bed at 120c:

I would say though a mechanical welding closed is incredibly unlikely

It's much less likely when switching an AC mains heater than when switching a low voltage DC heater, because the current is much lower, and because the arc that is formed when the contacts start to separate gets extinguished as the mains voltage passes through zero. That's why the rated current of a relay or switch is typically higher when switching AC than it is for DC.