Resistive or inductive load and relay rating

I'm wanting to switch a 400W 240V electronic MH balast from a relay,
and possibly a small switchmode 12V/5V supply from either the same or a
different relay. As it could affect relay selection, I was wondering
whether these type of supplies should be classified as resistive or
inductive? I could play safe and go for a heavy duty power relay having
excessive capacity, but would prefer not to over-design the solution if
a smaller PCB relay with, e.g. 10A 240Vac resistive rated contacts,
would do. If relevant, the initial power drawn by the balast is around
100W, ramping to 400W over a couple of minutes. Maybe these are
obviously inductive, but I'm unsure with these type of supplies. Also,
I suppose that the balast need not contain a transformer at all, but
it's sealed in epoxy so I'd need to ask the manufacturers about that.
Advice most welcome!

Nick wrote:
> I'm wanting to switch a 400W 240V electronic MH balast from a relay,
> and possibly a small switchmode 12V/5V supply from either the same or a
> different relay. As it could affect relay selection, I was wondering
> whether these type of supplies should be classified as resistive or
> inductive? I could play safe and go for a heavy duty power relay having
> excessive capacity, but would prefer not to over-design the solution

<snip>

Some suggestions:

1 - If you are unsure of the load or if it could possibly change once
the product has been released, design for the worst possible case.

2 - Relays often times have an inductive rating so you may want to
consult the data sheet or ask the manufacturer if it doesn't

3 - There are things you can do to help increase the contact longevity,
such as use of a form of snubber across the contacts. This can include
an RC, an MOV, etc.

4 - While I haven't used it, I have read application notes on adding a
zener diode in with the normal flyback diode. The idea is that the
zener delays the point at which the flyback diode begins to conduct
which in turn results in a more forcefull opening of the contacts. The
increase opening force is supposed to make the contacts open faster,
which should help extinguish the arc sooner.

Nick wrote:
> I'm wanting to switch a 400W 240V electronic MH balast from a relay,
> and possibly a small switchmode 12V/5V supply from either the same or a
> different relay. As it could affect relay selection, I was wondering
> whether these type of supplies should be classified as resistive or
> inductive? I could play safe and go for a heavy duty power relay having
> excessive capacity, but would prefer not to over-design the solution if
> a smaller PCB relay with, e.g. 10A 240Vac resistive rated contacts,
> would do. If relevant, the initial power drawn by the balast is around
> 100W, ramping to 400W over a couple of minutes. Maybe these are
> obviously inductive, but I'm unsure with these type of supplies. Also,
> I suppose that the balast need not contain a transformer at all, but
> it's sealed in epoxy so I'd need to ask the manufacturers about that.
> Advice most welcome!

Hi, Nick. Don't do it. Both loads have an inductive component, and as
with all inductive loads, your problem is going to be turning it off.

A big difference between relays that are made to switch inductive loads
and those that aren't is the draw length of the contacts. Relays are
inherently "snap action", but unless you can quickly draw the relay
contacts far enough apart to extinguish the arc, you'll end up arcing
for an extended period of time. That will burn up your contacts almost
immemdiately. And this has nothing to do with the size of the
contacts, as well as very little to do with contact force (the primary
considerations in developing contact current ratings).

Another difference between the two relays is contact composition.
Inductive-rated relays do have alloy contacts which are made to better
endure the high temperatures of arcing.

Relays made to switch inductive loads do have higher current coils,
because they require more pull to draw the contact further and faster.
They're also more expensive. But they're worth it. Just as a WAG, I'd
suggest you get a relay that's rated to switch 1/4HP to be safe. For
just one ballast, I'd doubt you'd need to go with a small lighting
contactor. That probably would be overdesign. I'd also recommend a
socketed relay, so the user can swap it out if the relay doesn't live
up to expectations.

If nothing else, you can just try it. Switch your relay on and off
with a five second on/five second off cycle using a 555, and just see
how long it lasts. I would think if you can get 10,000 cycles (this
should take less than a week of testing, turning it on at the beginning
of the workday and turning it off when you leave), you can start to
have some confidence in the switching.

But I'm sure you'll find that a relay made to switch only resistive
loads is going to cause you nothing but problems in the future.

Good luck
Chris

Thanks for the quick replies. I too remember seeing a paper about a
zener on the coil side a while back when looking at best practice on
driving relays. I'm going to look up on that again.

I'm the user in ths case, but I still really don't want a design fault
in what is supposed to be a safety oriented cicuit! Fused contacts
could cause precisely what I'm trying to avoid, so the advice is duly
noted, and the relay I had in mind is almost certainly inadequate. In
general I agree on testing a large number of cycles, but it could take
a long time in this case (or perhaps not if the relay really was
destined to fail) as once powered down, the ballast/lamp should not be
powered again until several minutes have elapsed, giving the bulb time
to cool sufficiently to allow a restrike.

The idea is to use a 12F675 to control the ballast/lamp, cooling fans
and other bits to give over temperature protection and controlled
operation when powering up or shutting down the system. For good
measure on the ballast circuit I was planning to put in a bimetalic
switch too.

I'm going to do some more relay research now.