mounting thermal fuses or snap switches on hotend

Afternoon all,

Anyone mounted thermal fuses or switches onto the e3D v6 hotend or heatsink? Mind sharing how you did this?

Thanks.

Slightly off-topic, but I just had an idea;

Wouldn't it be possible to add a feature to the printer boards so that the boards can measure the resistance of the heater? This way, if the printer sees that the temperature and resistance suddenly start to deviate from each other, it can do a safety stop?

I assume that the resistance of a heater would drop as it gets hotter. We probably cannot use the heater resistance as a trusty value to do temperature readings on, but I think it should be possible to at least add a safety feature this way.

---

http://www.marinusdebeer.nl/

Most heater elements use nichrome for the element although some may use manganin or constantan instead. All of these have a very low temperature coefficient of resistance but like almost all metals it is positive - the resistance increases with temperature. Heaters where the resistance is used both for heat and for measuring the temperature often use pure nickel for the element.

edit:

To get back to the ops question. It should be possible but there are two important considerations. Thermal fuses do have some self heating when wired in series with the heater so you will need to have a fuse rated at a little more than you would wish to trip out at. The other point is that you need good thermal contact but like will need to electrically insulate the fuse from the body of the hot end block - one way that occurs to me is to make a new block with a hole to take the thermal fuse and a sleeve of mica or similar to give electrical insulation.
Mike

Edited 1 time(s). Last edit at 05/12/2019 10:14AM by leadinglights.

Discussion around the topic is welcome!

There are a few failure modes to consider.

If the thermistor looses contact with the heater block the heater won't respond as expected and theefore go into fault.

If the MOSFET fails closed circuilt (common failure mode) the temperature should exceed limits and and heater fault raised.

If the heater fails and shorts out it should blow the fuse for the heater before any damage is done to the wiring.

If the heater circuit breaks open the heater won't respomd as expected and heater fault will be raised.

Heater fault will drop the PS_ON/ATX_Enable signal which will drop relays cutting the live and neutral legs of the PSU that feeds the heaters.

I can think of two failure modes which aren't covered at the moment which are:

Controller lock up
Hot end fan lock up.

I've never had the former but I have caught the latter which led to the hotend fan duct getting very hot and smelly. A thermal fuse in the hotend would stop everything as a last resort and a snap switch in the heat sink would open and drop the relays controlled by PS_ON if the heat sink got too hot after the fan failled.

It's a good idea!

You can get up to 300C in a typical thermal fuse case style.

As leadinglights said, you'd need to electrically insulate it from the heat block (the case is connected to one of the leads), but easily screwed to the side of a heat block with a little aluminium square block drilled with a hole for the fuse I should think, or just clamped to the heat block with a strip of aluminium. Use a bit of heat sink compound to conduct the heat well, simply wire the fuse in series with the heater element and that would be that I would have thought.

Fuse ..
[www.ebay.co.uk]

Square bar ..
[www.ebay.co.uk]

It's not hard to make a new heat block purpose made if you have some drills, a hack saw and a 6mm coarse thread tap ..

[www.ebay.co.uk]

Edited 1 time(s). Last edit at 05/12/2019 02:22PM by Pippy.

Fair point about it being relatively simple to make a heater block. If they're [edit: the fuses] a similar size to the heater cartridge then you could end up with something that looked more symmetrical about the nozzle axis than the standard V6 blocks! I'll have a little read about it. I think 300 is too low for some nylons PCs and almost certainly PEEK, but it could make processing PLA, ABS, PETG and flexibles far more failsafe.

Edited 1 time(s). Last edit at 05/12/2019 01:34PM by WesBrooks.

Maybe use some aluminium solder as the thermal fuse, run it along the side of the heat block simply stuck down with thermal adhesive tape ? You can get it with set melting points of 220C'ish to 500C'ish.

Edited 3 time(s). Last edit at 05/12/2019 06:30PM by Pippy.

With regards the snap switch I was thinking of something like this.

[uk.rs-online.com]

Bolted directly to one of the upper fins. I would use this to short out the circuit powering the relays blowing a DC side fuse. I would resistor limit the short current to just over double the fuse raiting to avoid drawing unnecessary current from the PSU.

The Airpax 6700 series bimetal thermostats that you point to are only available to a bit more than 100°C. While it is possible to use thermal equivalent of a resistor chain to put the switch where it will operate, this is dificult and unreliable in practice. Any switch with the contacts held at a temperature of much above 200°C will also be unreliable - making a switch with the bimetalic element in the hot zone and the switch contacts somewhere much cooler would work. For ideas it may be worth looking at the way Weller Magnestat soldering irons work - a switch controlled by the attraction of a magnet to a special iron slug - the slug looses its magnetic attraction sharply at the "Curie point" allowing the current to be switched off.

A quick edit: The "crowbar" protection that you propose will be very slow acting because the snap switch will take a long time to sense the overheat and the fuse will also take a significant time to blow even at twice its rating.

Mike

Edited 1 time(s). Last edit at 05/13/2019 05:25AM by leadinglights.

Yes my weller solderings work like that leadinglights.

I guess it's finding a magnet that regains it's magnetism when it cools.

Quote
Pippy

I guess it's finding a magnet that regains it's magnetism when it cools.

The magnet is on the end of a magnetic rod that has a higher Curie point - mild steel will do. The slug itself is an alloy of (I think) Iron, cobalt and aluminium

Mike

@leadinglights. Discussion on a few failure modes at once. My last sugheation was at the top of the hot end heatsink and 70 or 80 to detect hotend fan failure which would cause some nastyness well before the hotend exceeded temperature specifications. My bad for not being a little clearer.

Edited 1 time(s). Last edit at 05/13/2019 05:56AM by WesBrooks.

Quote
leadinglights
A quick edit: The "crowbar" protection that you propose will be very slow acting because the snap switch will take a long time to sense the overheat and the fuse will also take a significant time to blow even at twice its rating.

At twice rated currwnt fuses are normally specified to blow near instantly. That's from a human perspective rather than a MOSFET perspective. Blowing the fuse is only so you can see what has happened after the event. My relays don't self reset. That said you could crowbar the heater circuit directly as I think you suggest, perhaps by accident. But that would risk blowing the heater control MOSFETs as they are susceptible to short spikes.

Edited 1 time(s). Last edit at 05/13/2019 06:20AM by WesBrooks.

A couple of points. Firstly I am in full support of the idea - I particularly liked your list of reasons to incorporate a thermal fuse in the hotend. Having said that, I spent many years in product development and know that thermal protection is particularly difficult and even the textbooks get it wrong. I can also be wrong and have seen many products that work well where parts of the design cause my inner design guru to wince.

Quote
WesBrooks
......

At twice rated currwnt fuses are normally specified to blow near instantly. That's from a human perspective rather than a MOSFET perspective. .....

A fairly typical datasheet even for quick blow fuses is below - the times here are not short even by human perspective.

[docs-emea.rs-online.com]

I used the term "Crowbar" to mean any system where a fuse or other component is destroyed to protect the rest of the equipment.

Mike

No worries,Ijust though you were thinking that I was trying to orotect from a heater control of the cartridge failure with my last suggestion rather than fan failure.

Those specs are quite different to that of the automotive fusesI had been working with for my truck loom. I thought 2 was a good rule of thumb. Looks like that is really more like 3!