Contactor Vs Relay ?

[Adrian Brentnall]

HI Folks Some advice please. I'm working on the control gear for a homebuilt glass-fusing kiln. It's using a commercial digital controller, which provides a switched output @ 12v DC. Final load is two resistive wire elements, each rated @ 240v / 15A.

I can either feed the controller output through a small relay, to provide a switched 240v AC, or source a high-current Relay / Contactor that will accept the 12v. At the moment, I'm leaning towards the 'slave relay' approach, as I don't know what current the control board can supply.

I'd rather not go the fully electronic route with a solid state relay for the final device - I rather like things that go 'clunk' .

Looking at CPC, high powered relays are available, as are contactors, but the relays tend to be cheaper. What's best in this application? - is a contactor 'overkill' - or berely 'prudent' ?

Thanks for any informed comments. Adrian

[Andy Burns]

Dad used solid state relays on his kiln controller, instead of going "clunk", the specialist fuses that protect the relays liked to go "fot" unfortunately they were more expensive than the relays themselves.

I think he did eventually find a previous owner had incorrectly wired the elements, and now it's reliable.

[dennis]

Aren't contactors just relays with extra bits to drive high inductive loads?

Being a resistive load a relay should be fine.

[Bill Wright]

Double pole relays rated at 240V/15A per pole or better will do it. Use both poles in parallel. I made a gadget for switching the caravan electrics automatically between two supplies using relays, and it's been fine for 10 years.

Bill

[Andrew Gabriel]

Contactors are relays which can pass higher currents on the NO contacts, because the high magnetic coil force is used to generate a high contact pressure. Whilst a contactor can also have NC contacts, they can only pass a lower current because the contact force is just from the return spring.

Relays generally have NO and NC contacts identically rated, and therefore they can't go up to the same current ratings and life expectancy. The coil is significantly lower power, and the whole thing can therefore be much smaller.

I would go with a contactor for your application (assuming the kiln is something that fires for some time, like a pottery one).

[Adrian Brentnall]

HI Andy Thanks for your comments. I've always been a bit wary of high-power semiconductors protected by expensive fuses. Many years ago at university we used a big homebrew lighting rig, and when the explosive fuses blew in that (which they did fairly often, with a 'crack' like a gunshot and a bright purple flash), they often took out the triacs as well...

I'm sure that ssr's should be more reliable, but the luddite in me likes an electro-mechanical device - if for not other reason that you can hear it working...

[Adrian Brentnall]

Ah - that's interesting - thanks

A typical firing cycle will be three or four hours, but, as you're aiming for a particular profile of ramp-hold sequences over time, the elements aren't 'on' for the whole of that period.

I'm torn between installing the whole control in 'a box', or using a stripped-down consumer unit, which would have the advantage that din-rail-mounted contactors and mcb's can be easily mounted. I may even use a separate box to house the controller, its mains transformer, the over-temperature 'watchdog' and the low-power relays - leaving the 'real power' stuff in a separate CU..

Thanks Adrian

>

[Adrian Brentnall]

HI Bill I'm always a bit nervous of the 'commoned-contacts' approach, as Murphy's Law suggests that one set of contracts will operate slightly in advance of the other set, and will end up making or breaking the lion's share of the current.

I'm deliberately over-engineering this project! (Partly because I want it to be reliable, and partly for safety reasons - as it's a box of heat, operating unsupervised, in a timber workshop. For this reason I'm forgoing the usual 'let's see what's in the junk box' approach, and speccing / buying new components. Call me paranoid!

Thanks Adrian

[dennis]

Why not look for a solid state power controller..

You can get a 50A rated one for about £50.

They come in mark space control and in phase control. I think you want mark space for your application.

The power follows a DC input voltage of about 0-5V so you can drive it with a Pi or similar.

Don't forget to fit a heat rise fire detector to trip the supply.

[chris French]

In message , Adrian Brentnall writes

I'm interested in the homebuilt kiln, do you have any links?

Thanks

Chris French

[Adrian Brentnall]

Hi Chris Drop me a direct email (the email address is valid). I've got lots of links from the research I did last year - but it all depends on what you're wanting to do with it

Adrian

[newshound]

I built a controller for a small brewery "hot liquor tank", basically four 3 kW immersion heaters. I used octal base relays (notionally adequately specced) in the first version but they didn't last all that long. Solid state ones have been fine.

[Bill Wright]

I did wonder about that, but it doesn't seem to have happened. Since I wrote before I've remembered that I made a unit that switches in two sets of batteries on different time delays, the idea being that the alternator wouldn't find itself supplying 24V to the vehicle battery plus two 110Ah batteries both of which might theoretically be flat. That device used 20A commoned contacts relays and it's had some right hammer over the years, and it's been fine.

How's your insurance?

You need to be paranoid by the sounds of it.

Bill

[The Natural Philosopher]

Non inductive loads like heaters is the one case where I WOULD use SSRs without hesitation.

The fuses will not protect the SSRS but they will stop the wiring catching fire.

SSRS fail short circuit though,so its important to have a full mechanical DP isolator behind them for fault repair purposes.

[Andrew Gabriel]

It's not just the make or break order, but inevitable difference in contact resistances means that the current will be badly balanced between them (albeit, picking the better one at any one time;-) You can mitigate this to some extent by making sure there's a length of parallel wires connected to the contacts rather than paralleling up as close to the contacts as possible - the resistance of that wire will help reduce the ratio of differential contact resistances.

Relays are not usually rated for paralleling up contacts.

Another common mistake in this area is using a 3-phase contactor on single phase. A 3-phase 4-wire contactor rated at, say, 100A, will have 4 x 100A contacts. People will sometimes use this on a single phase (where the load doesn't require phase rotation), and overlook that the 100A neutral contact is now carrying 300A. Even when the contactor is only switching the 3 phases and the neutral is permanently connected, the connections and neutral conductors through the device are still only rated 100A.

[mick]

As the load is resistive, look at the AC1 rating of the contactor, not the usual AC3 "motor" rating. You'll use a much smaller contactor that way. If you are going for "over engineering" then don't use a conventional relay. The contacts run hotter at the same current rating. A contactor-relay is ok as it is really a small contactor with the contact numbers changed! If you can get a contactor with a 12vDC coil then that's what I would recommend. It'll be a lot cheaper SSRs by the time you've got heatsinks sorted out.

[Adrian Brentnall]

Thanks for the advice! I need to design some 'relay logic' so that the 'over-temperature' circuit can latch out if it trips - and so it's a question of just where things go from 12v coils to 240v coils...

It'd be nice to run 12v control signals everywhere as I have 12v available.

Thanks Adrian