High power manually-reset thermal cutout solution

[atjb6014]

Hi

I am trying to create a thermal cutout circuit to protect a large (10sqm) silicone heat pad table. The total heat pad load is around 30A (split between seven 1kW heat pads), running off a single phase 32A supply. The heaters are controlled with Eurotherms and SSRs, which fail in the closed position. Because of this I want to add thermal switches, but in a manually-resetting circuit. Due to the size of the table, the switches will be very hard to reach (wedged underneath a 500kg steel plate), hence why I dont want to just use thermal fuses.

The max table temp is 90deg, so I was planning to use 110deg switches to switch off the power to the control box via a power relay, the power relay is initially switched on by use of a momentary switch. Once the relay is closed, the coil is powered by the mains supply, until a thermal switch breaks this circuit and powers off the relay. The control box will then not power back on until the thermal switches have closed again, and the momentary switch is manually pressed. See diagram below.

However, I cannot find a power relay to these specs. The link below is the closest I could find, but contact rating is only 30A.

https://uk.farnell.com/panasonic/he1an-ac240v/power-relay-spst-no-240vac-30a/dp/2503681

I am slightly worried that the right components don't exist because my design is wrong in some way. So if anyone can help with the correct components or a new design that would be amazing.

Thanks

 

[ElectroChem]

You are correct that general purpose relays top out at around the 20-30A mark.The component you're after is called a contactor. These are essentially bigger tougher relays with contacts designed for making and breaking large currents and voltages, and can be found in ratings of 1000A or more.

This part from Schneider is the kind I would pick for your operation. It's stated rating is 25A, but that is for motor loads that are harder to break, the datasheet shows it is good for 40A on resistive loads.

 

[atjb6014]

You are correct that general purpose relays top out at around the 20-30A mark.The component you're after is called a contactor. These are essentially bigger tougher relays with contacts designed for making and breaking large currents and voltages, and can be found in ratings of 1000A or more.

This part from Schneider is the kind I would pick for your operation. It's stated rating is 25A, but that is for motor loads that are harder to break, the datasheet shows it is good for 40A on resistive loads.

Ah that's perfect, so easy once you know!
Thanks a lot for the quick answer as well

 

[marconi]

The circuit you have drawn will work but after a few operations the push to make switch will fail unless it too is rated to carry the load current of 32A. You need a second set of NO contacts through which you pass the current for the load (control box) from the right contact of the main switch. The first set of NO contacts act as the hold on by bypassing the push to make switch. The thermal cutout is connected as you have drawn.

 

[atjb6014]

The circuit you have drawn will work but after a few operations the push to make switch will fail unless it too is rated to carry the load current of 32A. You need a second set of NO contacts through which you pass the current for the load (control box) from the right contact of the main switch. The first set of NO contacts act as the hold on by bypassing the push to make switch. The thermal cutout is connected as you have drawn.

ah I think I see your point. In normal operation, once the relay-come-contactor is switched 'on', the Eurotherm controllers take a couple of seconds to power up. So the momentary switch will be released long before any large current tries to pass through it, so the load should only ever pass through the contactor.
But if the SSRs are already broken, which is the entire purpose of this circuit, the load could be drawn immediately through the momentary switch as it is pressed.
Is this what you meant?

What type of NO contact would this second set be though? It should also be momentary otherwise it could continue to bypass the whole thermal cutout circuit, so surely it would be simpler to just use a 32A momentary switch as the one already in the diagram?

 

[marconi]

Is this what you meant? - Yes

I cannot reply in detail right now but will do so tomorrow because my wife has just returned home form an outpatient appointment. It would be good to know the purpose of this heating system.

One set of NO contacts bypasses just the momentary push to close contacts. The other set switches power to the controller. You can work out as a bit of fun the circuit.

 

[Lucien Nunes]

surely it would be simpler to just use a 32A momentary switch as the one already in the diagram?

If such a thing exists it will be very clunky!

No, do as Marconi says. All normal contactors will have an addtional contact available. Make the control circuit self contained, comprising the push to engage button, thermal switches, N/O holding contact and contactor coil. Fuse it down to 1A or whatever so that you can use convenient small cabling to the thermal switches.

Then simply feed the 32A power circuit through a separate pole of the contactor.

 

[atjb6014]

If such a thing exists it will be very clunky!

No, do as Marconi says. All normal contactors will have an addtional contact available. Make the control circuit self contained, comprising the push to engage button, thermal switches, N/O holding contact and contactor coil. Fuse it down to 1A or whatever so that you can use convenient small cabling to the thermal switches.

Then simply feed the 32A power circuit through a separate pole of the contactor.

Ahhh, something about this wording made it click in my head. I've also had no experience with contactors before. So I'm just running that coil holding 240V from a different contact on the contactor, rather than the one which the heaters are connected to, hence isolating it from the possible 32A load.

This is a neat solution, super grateful for your help I've attached a new schematic just to confirm it, and in case anyone else finds this thread looking for the same solution

@marconi the heated table is actually a large mould to infuse carbon fibre/syntactic foam/epoxy resin composite sandwich panels, which are used to replace the heavy aluminium cargo pallets in aeroplanes. So it just holds at infusion temperature, and then ramps slowly to a cure temperature for a few hours. Best wishes to your wife.

Thanks again.

 

[James]

As this is obviously in a commercial environment, I would recommend using an automation company or panel builder with suitable experience.

Also the relevant cosh data sheets and risk assessments should be provided to the designer so they are able to use the correct safety level to prevent fire / harm / injury etc.

i.e. if the effect of constantly heating the mould with material inside could cause noxious gasses or explosion / fire. the system of control would likely be different to if it Simpley meant that the part would be scrap.

 

[pc1966]

Typically you get add-on aux contacts. I have not checked comparability closely, but something like these:

A9C20842 | Schneider Electric iCT Series Contactor, 230 V ac Coil, 2-Pole, 40 A, 2NO, 250 V ac | RS

uk.rs-online.com

A9C15914 | Schneider Electric iACT Series Contactor, 2-Pole, 1NO + 1NC, 24 → 130 V dc, 24 → 240 V ac | RS

uk.rs-online.com

As already mentioned, you ought to have a small fuse (1-3A sort of thing) in series with the supply to the push-button and aux contacts just in case.
Finally (and in addition to James's advice about risk assessments, etc) you might want to have a couple of N/C sensors located over the area in case of local hot spots. They would be wired in series but do that at the control panel if the switches themselves are inaccessible. I.e. bring out both wires to each to a terminal strip (DIN rail or whatever) so if one fails you can identify and bypass it.

 

[pc1966]

This sort of thing:

 

[pc1966]

You can use the N/C contacts to light a warning lamp if tripped, etc, from the same fused supply.

 

[pc1966]

Actually the above circuits are all poor design in they have the thermal switches in the coil N return. While in these RCD days it ought to trip if there is a fault, that is not certain (depends on N-E voltage at CU, and coil sustain current, if the RCD happens to be working). So a fault to earth at one of those switches potentially bypasses the cut-out action which is a BAD THING.

This revised version puts the coil to N and all the switches to the L feed, that way a fault will blow the fuse (assuming things are earthed?) and cut the power to all so if fail-safe in that aspect:

 

[freddo]

I agree with the above. It is interesting to note that many class I appliances with heaters have the control switching the live and the thermal protection in the neutral.

Depending on the environment or what the heating pad is heating and so on, you may also need thermal links wired in series with each element. Don't ask me though, I don't get involved with designing safety systems!

 

[pc1966]

I was going to add that if this is all DIN rail mounted stuff you might want to look at fuse holders to match. If you use HRC fuses (not glass!!!) and the expected fault current is below 1.5kA then this is acceptable:

1SNA199208R1100 | Entrelec D Series Grey Fused DIN Rail Terminal, 1.5mm², Screw Termination, Fused | RS

uk.rs-online.com

Otherwise I doubt you will ever see too high a PFC for this:

DF101 | Schneider Electric Rail Mount Fuse Holder for 10 x 38mm Fuse, 1P | RS

uk.rs-online.com

C10G1 | Eaton 1A Ceramic Cartridge Fuse, 10 x 38mm | RS

uk.rs-online.com

 

[marconi]

If the heating elements each have their own thyristor switches then you will have to include an over temperature sensor to monitor each element t- the sensors wired as a daisy-chain. If the elements are all controlled by one thyristor switch then one might consider only monitoring only one element and regard it as a proxy for all the others - but I think/regard this to be poor design and each should be monitored , especially since the sensors - are relatively cheap. I think too you ought to have a digital thermometer actually measuring the temperature at various locations via selectable sensors placed over the heating table.

The advice on doing a risk assessment and failure modes is very good albeit it may seem like putting obstacles in your way - you want a working, reliable and safe product.

PS: I see my esteemed colleagues have said as much already - we are a chorus then

 

[James]

Just to add, many plastic extrusion machines have a similar control method and requirements.
might be worth having a chat with someone experienced in building them.