solid-state relays - how to spec ?

Adrian Brentnall brought next idea :

Solid state is fine and would only need to be rated for a resitive load. As would a relay/contactor or even an proportional SCR system. I've not seen mercury wetted for a long time?

Any relays would need to be AC rated.

Dad has a 10kW(ish) kiln, the controller for that uses solid state relays to do ramps and holds, the recommended fuses to protect the relays cost more than the relays themselves - at one time it was blowing quite a few fuses, I think it turned out to be an issue with the wrong heating elements.

Opto coupled zero voltage switching SSRs will be best.

Pick a SSR that is rated at double or more than the running current to allow margin for inrush currents into cold elements. Don't even think about using them in parallel for extra current handling

- one always comes on first!

Mount on a heatsink, fins vertical - check the data sheet but you possibly need something like 2 degrees C per watt.

For when a volt drop aint a problem, I'd go solid state BUT not if its going to be in ahot place.

They really are trivially easy to make or buy BUT they cant handle temps much over 150C even derated.

e.g. this

is only rated to 80C operating.

But its a nice thing. Zero crossing switch and will switch 15A mains.

The mercury ones are mecury in glass bottles that tilt so that the mercury covers the contacts inside to switch on. They were usually used in explosive atmospheres (eg coal mines, petrol pumps etc) in day s of yore. Expensive & haven't seen any for a long time.

I've been involved with control panels for more years than I care to remember now... lol

You must NEVER EVER rely on a SSR to isolate the supply or cut things off in an emergency - it may not turn off. Don't share one fuse as you won't be able to get close enough protection.

How I'd do it:

It may be an idea to delay one element slightly on initial switch-on, allowing the first one to get fairly hot first (with consequent increase in resistance and reduction in total startup current). Don't delay too long unless there is no danger from uneven heating. After startup the SSRs can switch together as the thermal inertia will keep the heater resistance up. Sometimes the delay is done using another "kiln sitter" type unit in reverse, to enable the second heater only when the kiln temperature is above a certain point.

It would be good to use a "kiln sitter" as you described, killing the supply to the conventional contactor (which should use "stop" & "start" buttons with a holding contact to give failsafe operation).

Have fun. :)

I was usually on the low-power side - let somebody else worry about the 'big amps'...

No - I was planning on the 'safety' being an electro-mechanical device

Don't share one fuse as you won't

Good plan

OK - this is my 'safety' relay ? - and the 'stop' button is an 'emergency stop' - in series with the kiln sitter ? Do you have an example of how this would be wired ?

You'd prefer SSR's over a mechanical relay ? I was leaning towards electro-mechanical... but then, I'm a dinosaur!

Hmm - if I use the controller that I have 'spare' then it's only got a single output - so I'd need to look at some extra delay circuitry.

Assuming 3kw per element, I guess it's only the same as two electric fires switched on at the same time.. so maybe running two relays (rather than a single one with twice the rating) is making life complicated.

I think there could be issues with uneven heating - depending on the physical layout of the elements. Kiln is planned to be about 3ft 6" x 2 ft - haven't got as far as drawing element layouts yet.

I have various books & dvd's on order from people who have been this way before.... so I'll not be completely in the dark.

I'm looking at 'build-your-own' due to frustration with my existing commercial kiln, which tends to shed bits of firebrick onto the glass as it fuses... As I discovered the hard way, the controller is also rather 'dumb' - as a fault on the thermocouple resulted in the kiln trying to fire 'forever', as it waited for the temperature to register an increase.... Luckily, I caught it somewhere round-about the 980c mark - normal process temperatures are under 800 - so that's why I'm thinking about kiln-sitters as a second line of safety

The other motivation is to build a large kiln for round-about 1000 -

If you can point me at a circuit diagram then that would be great.

I'm planning on mounting the control gear remote from the 'hot bits' - so I guess conventional contactors would work OK ?

Hopefully! Thanks Adrian

Hi Harry I was leaning towards relays 'cos they 'clunk' in a satisfying fashion . Don;t know anything about Solid-state...

I was planning on over-rating any device by 2x or so - in the hope of extending its life. I guess a plug-in device might make for easy maintenance...

Thanks Adrian

I've worked on lighting rigs with explosive fuses - the fuse blowing was usually an indication that the control device had also gone west

My bad - I was going to do the real maths later..

Any reason 'why' you'd prefer SSR over electro-mechanical ?

I was planning on keeping the control gear remote from the 'hot bits' - probably in a wall-mounted cabinet, with cabling to the heating elements and thermocouples - so I can keep the heat away from the controls

Fairly pricey, though the SSR. The same people do various 30A relays for under a tenner...

Some blurb here from one kiln manufacturer

It does mention long hold-times in a hot ambient environment - which isn't really what I'll be doing - so it's probably overkill!

Zero voltage switching and more suitable for the number of high power on/off operations that you propose Also any further control you might want to do is done a low voltage and low power. A few milliamps will control 10s of amps with ease.

Ah - thanks - I see! The 'spare' controller that I was planning on using outputs 12v - but who knows what I might end up using in the future ?

Are these things fairly sturdy nowadays ? I guess I can always add an indicator of some sort to replace that reassuring 'clunk' from a real relay...

Thanks

I'd use SSRs. Ten or more years ago I built a heater controller for a brewery with four octal base relays each controlling a 3 kW immersion heater. They didn't last particularly long, and I replaced them with SSRs which have proved much more reliable. The ones I used were about 2 inches square, from RS (because they had an account). They need a heat sink.

The other thing you might consider for your kiln is a "policeman", i.e. a second thermocouple and temperature controller set a bit higher than your normal target temperature and arranged to trip the supplies in the event of a control system failure.

A watchdog!

plus mercury is toxic so a risk if the glass bottles break.

There is legislation relating to electronics and certain metals. Google for ROHS and then REACH (which is ROHS on steroids)

SO expect you can't buy this particular technology anymore

You can get SSRs with an indicator LED built in.

OK - good 'evidence' in favour of SSRs - thanks

Yes - they're known as 'kiln-sitters' - and they're designed to do just what you say - preventing the whole lot from 'meltdown' in the event of a controller / power 'relay' failure. Having a background in software control systems - I was a bit surprised to see that the (commercial) controller on my current kiln didn't have any logic to handle 'heat on but temperature not increasing'. When the thermocouple failed, soon after installation, the thing just went on heating - luckily I spotted it and pulled the plug before it got much above the high 900 c's....

The Big Kiln will have an independent 'kiln sitter' to avoid that - might even end up with a pic-based controller with a little bit more intelligence too!

Thanks