Relay contact ratings.

[Sylvia Else]

The defrost timer in my fridge is failing, and given the cost of a
'genuine' replacement part, I was pondering the option of making
electronic timer driving an electromechanical relay.

It probably won't happen, but when I was looking at relay specs, I found
that their ratings are usually specified as a highish reactive power,
and a much lower real power.

http://australia.rs-online.com/web/p/electromechanical-relays/1279550/

is typical, specifying in this case 300 W / 2500 VA.

I find this difficult to fathom. Imagine a purely inductive 2500VA load.
So no real power being switched, but it would arc like crazy. What am I
missing here?

Sylvia.

 

[Phil Allison]

"Stupider than Anyone Else"

http://australia.rs-online.com/web/p/electromechanical-relays/1279550/

is typical, specifying in this case 300 W / 2500 VA.

** The 300W figure refers to DC switching - ie 10 amps at 30 volts.

The 2500VA figure refers to resistive loads and 250VAC power.

There is nothing about inductive load switching.

See the data sheet.

Fuckwit.



.... Phil

 

[Sylvia Else]

"Stupider than Anyone Else"



** The 300W figure refers to DC switching - ie 10 amps at 30 volts.

The 2500VA figure refers to resistive loads and 250VAC power.

There is nothing about inductive load switching.

See the data sheet.

Care to point, in the data sheet, to where that distinction is drawn.

Sylvia.

 

[Trevor Wilson]

The defrost timer in my fridge is failing, and given the cost of a
'genuine' replacement part, I was pondering the option of making
electronic timer driving an electromechanical relay.

It probably won't happen, but when I was looking at relay specs, I found
that their ratings are usually specified as a highish reactive power,
and a much lower real power.

http://australia.rs-online.com/web/p/electromechanical-relays/1279550/

is typical, specifying in this case 300 W / 2500 VA.

I find this difficult to fathom. Imagine a purely inductive 2500VA load.
So no real power being switched, but it would arc like crazy. What am I
missing here?

Sylvia.

**Transformers and motors are notoriously nasty for relays to deal with.
TRIACs are a much better choice for such loads. Easy enough to design a
simple circuit, or, if you want a really easy way out, just buy a
suitably rated Solid State Relay (SSR). SSRs and VERY easy and safe to use.

 

[Sylvia Else]

**Transformers and motors are notoriously nasty for relays to deal with.
TRIACs are a much better choice for such loads. Easy enough to design a
simple circuit, or, if you want a really easy way out, just buy a
suitably rated Solid State Relay (SSR). SSRs and VERY easy and safe to use.

It's complicated by the fact that the defrost timer switch is a SPDT -
it switches between the compressor (via the thermostat, I presume) and
the defrost heater. In my relatively cursory search, I haven't found a
SPDT SSR. Could use two, I suppose, but a failure mode that leaves both
heater and compressor running seems more likely than with an
electromechanical relay.

The existing switch has contacts just as a relay would; they just don't
have an electromagnetic actuator.

The switch only cycles four times a day, so it's not so demanding in
terms of contact life.

Sylvia.

 

[mike]

The defrost timer in my fridge is failing, and given the cost of a
'genuine' replacement part, I was pondering the option of making
electronic timer driving an electromechanical relay.

It probably won't happen, but when I was looking at relay specs, I found
that their ratings are usually specified as a highish reactive power,
and a much lower real power.

http://australia.rs-online.com/web/p/electromechanical-relays/1279550/

is typical, specifying in this case 300 W / 2500 VA.

I find this difficult to fathom. Imagine a purely inductive 2500VA load.
So no real power being switched, but it would arc like crazy. What am I
missing here?

Sylvia.

You can often get more life out of a mechanical defrost timer by opening
up the
motor and oiling the bearings.

 

[F Murtz]

You can often get more life out of a mechanical defrost timer by opening
up the
motor and oiling the bearings.

I hope you are not being sarcastic.

 

[Sylvia Else]

I hope you are not being sarcastic.

It could be a reference to the absence of bearings - except to the
extent that the metal axle rotates in a hole in the plastic housing. At
least, that's the situation in the one I have.

Perhaps I shouldn't be concerned about having a "genuine" part, because
the original looks about as cheaply made as it could get while still
working at all.

Sylvia.

 

[Sylvia Else]

That makes more sense. RS Components don't help by reversing the order
of the power ratings relative to the voltage and current ratings.

Quoting power ratings seems redundant.

Sylvia

 

[Spehro Pefhany]

It's complicated by the fact that the defrost timer switch is a SPDT -
it switches between the compressor (via the thermostat, I presume) and
the defrost heater. In my relatively cursory search, I haven't found a
SPDT SSR. Could use two, I suppose, but a failure mode that leaves both
heater and compressor running seems more likely than with an
electromechanical relay.

The existing switch has contacts just as a relay would; they just don't
have an electromagnetic actuator.

The switch only cycles four times a day, so it's not so demanding in
terms of contact life.

Sylvia.

Something like the Omron G7L is properly rated for motor loads. You
could use a second relay for the heater.




Best regards,
Spehro Pefhany

 

[ehsjr]

The defrost timer in my fridge is failing, and given the cost of a
'genuine' replacement part, I was pondering the option of making
electronic timer driving an electromechanical relay.

It probably won't happen, but when I was looking at relay specs, I found
that their ratings are usually specified as a highish reactive power,
and a much lower real power.

http://australia.rs-online.com/web/p/electromechanical-relays/1279550/

is typical, specifying in this case 300 W / 2500 VA.

I find this difficult to fathom. Imagine a purely inductive 2500VA load.
So no real power being switched, but it would arc like crazy. What am I
missing here?

Sylvia.

The price of the relay at the url you posted is ridiculous.
Surely you can get something much cheaper and equally robust.
There are relays made for appliances like air conditioners
available from Digikey for < $5.00 so it's likely you can
get something where you are for much less than the ~ $82.00 at
the RS Australia site. You might be able to use an automotive
relay. Generally their contacts are rated at 30 or 40 amps at
12 VDC, and they'll handle 240 AC nicely.

Ed

 

[mike]

It could be a reference to the absence of bearings - except to the
extent that the metal axle rotates in a hole in the plastic housing. At
least, that's the situation in the one I have.

That could be a problem. Mine has a standard clock motor with metal
case and bronze-looking bearings.

A clock motor has about zero torque. Takes very little friction in
the first few gear reduction stages to stop it. hardened grease can
easily do it. Just cleaning out the dried up grease can make it work.

Mine ran for another decade after I cleaned/oiled it.
Replacement was about $13, but it failed after a week. Second replacement
timer lasted years before I retired the fridge.

 

[Phil Allison]

"kreed"

"TW"

Even though SSR's and Triacs are quite rugged, something like a surge
from a lightning strike or similar event could cause one or both of
these triacs/SSR's to short out,

** Nonsense.

If the max voltage rating of a triac is exceeded by a spike on the AC
upply - it simply turns on for the rest of that half cycle.



.... Phil

 

[Phil Allison]

"John Fields" .

The contacts are rated at 10A for either 250VAC or 30VDC, so 250VAC *
10A = 2500VA, and 30VDC * 30A = 300W.

The higher AC voltage is permitted because the arc on opening will
quench in, at most, 1/2 cycle when the voltage across the contacts
crosses zero, while with DC through the contacts the arc will persist
until the gap between the contact is large enough to quench the arc.

** Problem being, there is simply not enough gap available to break the arc
that forms if the DC ratings are exceeded.

For the type of relay in question, a permanent arc will form with DC if the
current flow is say 10 amps and the voltage across the contacts is over 40
volts. This means there is over 400 watts of heat, mostly being dissipated
into the contacts and destroying them in a few seconds.

In the world of high powered audio amplifiers it is still common to see such
relays used to protect speakers from DC high fault currents and turn on/off
transients. In the case of former, the relay is not capable of doing the
job.



.... Phil

 

[Trevor Wilson]

Even though SSR's and Triacs are quite rugged, something like a surge
from a lightning strike or similar event could cause one or both of
these triacs/SSR's to short out,

**Unlikely. In fact, IME, TRIACs (appropriately rated ones) are vastly
more reliable than relays when driving highly inductive loads. In fact,
I've been using several for around 30 years, without issue. That is not
to say that TRIACs cannot fail. They can and do and usually shorted.
Which, in Sylvia's case, may prove inconvenient.

and since fridges are turned on for

pretty much all their lives, this ensures that the fridge will be
connected to mains when something like this happens (unless you are
there to disconnect it at the first sign of lightning).

**IME, the vast majority of lightning problems occur via TV antennas.
Power line issues are massively over-stated. Again, the only time I can
pin point a power line "surge" as the direct cause of a problem was a
very long time ago, when a 5kV bearer fell across the 240VAC main
overhead lines. The damage was considerable and affected several blocks.

While same event might weld or vaporise relay contacts, its extremely
unlikely that both NO and NC could be connected to the common contact
simultaneously in any relay failure I can think of (short of the thing
being physically crushed). This is one good reason to use a relay.

**I agree. However, in terms of longevity, TRIACs win hands down.

 

[Trevor Wilson]

"John Fields" .

** Problem being, there is simply not enough gap available to break the arc
that forms if the DC ratings are exceeded.

For the type of relay in question, a permanent arc will form with DC if the
current flow is say 10 amps and the voltage across the contacts is over 40
volts. This means there is over 400 watts of heat, mostly being dissipated
into the contacts and destroying them in a few seconds.

In the world of high powered audio amplifiers it is still common to see such
relays used to protect speakers from DC high fault currents and turn on/off
transients. In the case of former, the relay is not capable of doing the
job.

**Tapco CP500 anyone? Crap design, with useless relay protection.

 

[Phil Allison]

"Trevor Wilson"

**IME, the vast majority of lightning problems occur via TV antennas.

** Plus the telephone line - modems and TAMs drop like flies when there is
a thunderstorm.

Power line issues are massively over-stated.

** It's an issue in many rural areas - the solution to which is fitting
varistors in the power box.

**I agree. However, in terms of longevity, TRIACs win hands down.

** Funny how microwave ovens all seem to have relays turning on the big
tranny.



.... Phil

 

[Spehro Pefhany]

The price of the relay at the url you posted is ridiculous.
Surely you can get something much cheaper and equally robust.
There are relays made for appliances like air conditioners
available from Digikey for < $5.00 so it's likely you can
get something where you are for much less than the ~ $82.00 at
the RS Australia site. You might be able to use an automotive
relay. Generally their contacts are rated at 30 or 40 amps at
12 VDC, and they'll handle 240 AC nicely.

Ed

I would not use a 12V automotive relay to switch mains voltage,
particularly inductive loads.. coil-to-contact dielectric strength of
those POS relays is << 1kV. I like to see 4kV or so. Are the housings
required to be flame retardant?

 

[swanny]

"John Fields" .

** Problem being, there is simply not enough gap available to break the arc
that forms if the DC ratings are exceeded.

For the type of relay in question, a permanent arc will form with DC if the
current flow is say 10 amps and the voltage across the contacts is over 40
volts. This means there is over 400 watts of heat, mostly being dissipated
into the contacts and destroying them in a few seconds.

In the world of high powered audio amplifiers it is still common to see such
relays used to protect speakers from DC high fault currents and turn on/off
transients. In the case of former, the relay is not capable of doing the
job.

Seen that happen. Contacts weld, amp and speakers go boom.

 

[Phil Allison]

"swanny"
"Phil Allison"

Seen that happen. Contacts weld, amp and speakers go boom.

** I have made a bit of a study of this in an attempt to find a solution.

Conclusions are:

1. There is no readily available relay capable of breaking the voltages and
currents involved if a 1000 wpc amp goes DC.

2. If you wire a changeover relay so it shorts the speaker and disconnects
the amp at the same time - chances are good the speakers will be OK, but
the relay contacts will be burnt by the resulting arc to ground. The amp
must have DC rail fuses if this is to work.

3. A large ( ie octal base ) relay with 4mm clearance contacts PLUS a
strong magnet next to the contacts CAN
work with amps up to about 400wpc. The magnet pulls at the arc and helps
break it.

4. A triac "crowbar" will also work but gives no on/off transient
suppression. The amp must be able to take a dead short without internal
failure and not have response down to subsonics or DC.




.... Phil