Question re: power relays & RC snubbers

[Adam Tenenbaum]

Hi,

I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A. Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads. Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts? If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

Thanks,
Adam

 

[Tom Biasi]

"Adam Tenenbaum" <ac740@torfree.net> wrote in message
news:3acad9f1-8ff7-41a3-b27e-508c8487eb05@h5g2000yqh.googlegroups.com...

Hi,

I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A. Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads. Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts? If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

Thanks,
Adam

Hi Adam,
You are correct, usually resistive loads do not cause this condition.
If the relay contacts are not wiping properly (the slight slide when the
contacts touch) this could happen.
If the contact spring is weak, this could happen.
Is the 10 ohms the cold resistance?
Tom

 

[Ross Herbert]

On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac740@torfree.net>
wrote:

:Hi,
:
:I'm trying to help diagnose a problem; the application is a simple
ower relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
:I was surprised to see significant charring on the relay contacts, as
:the relay is rated for 30A. Everything I have read about the topic
i.e. in this group + elsewhere) suggests that this effect only really
:happens with inductive loads. Is it possible that a normal resistive
:load could cause high enough inrush currents to char these relay
:contacts? If so, I'd appreciate advice for selecting appropriate
:values for the resistor/cap (without having a scope available to track
:dv/dt).
:
:Thanks,
: Adam


Could you supply relay make/model details please?

Usually, only minimal arcing will occur on relay operation in situations such as
yours. However, if the contacts are insufficiently tensioned, contact bounce may
be prevalent, and considerable arcing may occur during operation. Also, if the
relay is too slow to release, contact break time may be excessive, thus
prolonging any arcing. The faster a relay can both operate and release with
minimal contact bounce, the tendency to arcing will be reduced.

 

[Rich Webb]

On Thu, 29 Jan 2009 22:57:46 GMT, Ross Herbert <rherber1@bigpond.net.au>
wrote:

On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac740@torfree.net
wrote:

:Hi,
:
:I'm trying to help diagnose a problem; the application is a simple
ower relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
:I was surprised to see significant charring on the relay contacts, as
:the relay is rated for 30A. Everything I have read about the topic
i.e. in this group + elsewhere) suggests that this effect only really
:happens with inductive loads. Is it possible that a normal resistive
:load could cause high enough inrush currents to char these relay
:contacts? If so, I'd appreciate advice for selecting appropriate
:values for the resistor/cap (without having a scope available to track
:dv/dt).
:
:Thanks,
: Adam


Could you supply relay make/model details please?

Usually, only minimal arcing will occur on relay operation in situations such as
yours. However, if the contacts are insufficiently tensioned, contact bounce may
be prevalent, and considerable arcing may occur during operation. Also, if the
relay is too slow to release, contact break time may be excessive, thus
prolonging any arcing. The faster a relay can both operate and release with
minimal contact bounce, the tendency to arcing will be reduced.

With this in mind, the problem could well be on the coil side, not the
load side. What kind of inductive spike protection is the OP using on
the coil? If it's the common practice of using only a diode then that
could be where the problem is. See
<http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3264.pdf> for
one of several discussions of this.

--
Rich Webb Norfolk, VA

 

[Adam Tenenbaum]

On Jan 29, 4:24 pm, "Tom Biasi" <tombi...@optonline.net> wrote:

"Adam Tenenbaum" <ac...@torfree.net> wrote in message

news:3acad9f1-8ff7-41a3-b27e-508c8487eb05@h5g2000yqh.googlegroups.com...

Hi,

I'm trying to help diagnose a problem; the application is a simple
powerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on therelaycontacts, as
therelayis rated for 30A.  Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads.  Is it possible that a normal resistive
load could cause high enough inrush currents to char theserelay
contacts?   If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

Thanks,
 Adam

Hi Adam,
You are correct, usually resistive loads do not cause this condition.
If therelaycontacts are not wiping properly (the slight slide when the
contacts touch) this could happen.
If the contact spring is weak, this could happen.
Is the 10 ohms the cold resistance?
Tom

Yes -- it was measured as 11.3 Ohms cold, and it's referred to as a
1400 W coil (reference of 110V).

Adam

 

[Adam Tenenbaum]

On Jan 29, 5:57 pm, Ross Herbert <rherb...@bigpond.net.au> wrote:

On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac...@torfree.net
wrote:

:Hi,
:
:I'm trying to help diagnose a problem; the application is a simple
owerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
:I was surprised to see significant charring on therelaycontacts, as
:therelayis rated for 30A.  Everything I have read about the topic
i.e. in this group + elsewhere) suggests that this effect only really
:happens with inductive loads.  Is it possible that a normal resistive
:load could cause high enough inrush currents to char theserelay
:contacts?   If so, I'd appreciate advice for selecting appropriate
:values for the resistor/cap (without having a scope available to track
:dv/dt).
:
:Thanks,
:  Adam

Could you supplyrelaymake/model details please?

Usually, only minimal arcing will occur onrelayoperation in situations such as
yours. However, if the contacts are insufficiently tensioned, contact bounce may
be prevalent, and considerable arcing may occur during operation. Also, if therelayis too slow to release, contact break time may be excessive, thus
prolonging any arcing. The faster arelaycan both operate and release with
minimal contact bounce, the tendency to arcing will be reduced.

It's a Finder relay, part# 66.82.8.110.0307 -- rated for 30A, DPST NO,
switched on 110VAC with AgCdO contacts. I wasn't the owner of the
device in question when the first relay failed in this way, but now
that I am, I'm trying to prevent it from happening again.

http://www.findernet.com/en/products/detail.php?codice=668281100307&lang=en&gruppo=gruppo1

Adam

 

[Adam Tenenbaum]

On Jan 29, 7:17 pm, Rich Webb <bbew...@mapson.nozirev.ten> wrote:

On Thu, 29 Jan 2009 22:57:46 GMT, Ross Herbert <rherb...@bigpond.net.au
wrote:



On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac...@torfree.net
wrote:

:Hi,
:
:I'm trying to help diagnose a problem; the application is a simple
owerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
:I was surprised to see significant charring on therelaycontacts, as
:therelayis rated for 30A.  Everything I have read about the topic
i.e. in this group + elsewhere) suggests that this effect only really
:happens with inductive loads.  Is it possible that a normal resistive
:load could cause high enough inrush currents to char theserelay
:contacts?   If so, I'd appreciate advice for selecting appropriate
:values for the resistor/cap (without having a scope available to track
:dv/dt).
:
:Thanks,
:  Adam

Could you supplyrelaymake/model details please?

Usually, only minimal arcing will occur onrelayoperation in situations such as
yours. However, if the contacts are insufficiently tensioned, contact bounce may
be prevalent, and considerable arcing may occur during operation. Also, if the
relayis too slow to release, contact break time may be excessive, thus
prolonging any arcing. The faster arelaycan both operate and release with
minimal contact bounce, the tendency to arcing will be reduced.

With this in mind, the problem could well be on the coil side, not the
load side. What kind of inductive spike protection is the OP using on
the coil? If it's the common practice of using only a diode then that
could be where the problem is. See
http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3264.pdf> for
one of several discussions of this.

--
Rich Webb     Norfolk, VA

Thanks Rich (& everyone else I just mass replied to),

I don't believe that there's any type of protection on the coil side,
but this problem (at least in that application note) seems to be DC
related. Does an equivalent issue exist for AC? (The relay in
question is being switched by AC from a pressurestat). The only other
source of EMF that I can think of is a motor; however, it's not being
switched by the relay, but they do share common/neutral.

Adam

 

[Phil Allison]

"Adam Tenenbaum"

I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A. Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads.

** Nonsense.

Breaking significant current causes arcing across the contacts.

Breaking the 120 volts AC supply to a 12 amp resistive load will create a
visible blue flash every time.

If it were a 120 volt DC supply, then the blue flash would not extinguish
and the relay would be destroyed in a few seconds.

Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts?

** Breaking causes arcing.

If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

** For 10 amps, you will need a very large cap to suppress arcing.

Something is fishy, cos the Finder relay you describe elsewhere is rated for
100,000 operations at rated AC load.



...... Phil

 

[Ross Herbert]

On Thu, 29 Jan 2009 16:50:57 -0800 (PST), Adam Tenenbaum <ac740@torfree.net>
wrote:

:On Jan 29, 5:57 pm, Ross Herbert <rherb...@bigpond.net.au> wrote:
:> On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac...@torfree.net>
:> wrote:
:>
:> :Hi,
:> :
:> :I'm trying to help diagnose a problem; the application is a simple
:> owerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
:> :I was surprised to see significant charring on therelaycontacts, as
:> :therelayis rated for 30A.  Everything I have read about the topic
:> i.e. in this group + elsewhere) suggests that this effect only really
:> :happens with inductive loads.  Is it possible that a normal resistive
:> :load could cause high enough inrush currents to char theserelay
:> :contacts?   If so, I'd appreciate advice for selecting appropriate
:> :values for the resistor/cap (without having a scope available to track
:> :dv/dt).
:> :
:> :Thanks,
:> :  Adam
:>
:> Could you supplyrelaymake/model details please?
:>
:> Usually, only minimal arcing will occur onrelayoperation in situations such
as
:> yours. However, if the contacts are insufficiently tensioned, contact bounce
may
:> be prevalent, and considerable arcing may occur during operation. Also, if
therelayis too slow to release, contact break time may be excessive, thus
:> prolonging any arcing. The faster arelaycan both operate and release with
:> minimal contact bounce, the tendency to arcing will be reduced.
:
:It's a Finder relay, part# 66.82.8.110.0307 -- rated for 30A, DPST NO,
:switched on 110VAC with AgCdO contacts. I wasn't the owner of the
:device in question when the first relay failed in this way, but now
:that I am, I'm trying to prevent it from happening again.
:
:http://www.findernet.com/en/products/detail.php?codice=668281100307&lang=en&gruppo=gruppo1
:
:Adam


The data sheet indicates that the maximum AC current which can be broken for
resistive loads (AC1 rating) is 22.7A (2500VA/110V), so provided that the load
is purely resistive (slightly inductive is OK), your stated approx 10 Ohms load
shouldn't stress the contacts at all. Assuming that the load is about 10 Ohms
resistive the maximum current required to be broken is 11A which is only 50% of
the rated maximum breaking current.

I would suggest that you remove the relay and jury rig it to operate from the
same AC voltage and connect a known 10 Ohm resistive load to operate via the
contacts - a 1.2kW resistive load would be an ideal test. If you don't see any
appreciable arcing on operation or release then this will indicate that in the
real life situation there is some other load factor (inductive) which you don't
know about.

If possible you might be able to use both contacts in parallel to share the load
current if all else fails.

 

[Adam Tenenbaum]

On Jan 29, 11:31 pm, "Phil Allison" <philalli...@tpg.com.au> wrote:

"Adam Tenenbaum"



I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A.  Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads.

**  Nonsense.

Breaking significant current causes arcing across the contacts.

Breaking the 120 volts AC supply to a 12 amp resistive load will create a
visible blue flash every time.

If it were a 120 volt DC supply, then the blue flash would not extinguish
and the relay would be destroyed in a few seconds.

Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts?

**  Breaking causes arcing.

If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

**  For 10 amps, you will need a very large cap to suppress arcing.

Something is fishy, cos the Finder relay you describe elsewhere is rated for
100,000 operations at rated AC load.

.....   Phil

The relay is cycling at about 1 min intervals, so if the device was
running for 4 hours/day, the expected life at maximum rated load (i.e.
based on 100,000 operations) is a little over a year. I believe this
is about how long it lasted, but I guess I expected a bit longer life
(since it's running at half its maximum rated current) and a less
spectacular failure (completely black contacts). Having said that,
shouldn't I still be able to reasonably extend the contact life /
suppress arcs with a simple RC snubber circuit paralleling the
contacts?

Adam

 

[Adam Tenenbaum]

On Jan 30, 6:43 am, Ross Herbert <rherb...@bigpond.net.au> wrote:

On Thu, 29 Jan 2009 16:50:57 -0800 (PST), Adam Tenenbaum <ac...@torfree.net
wrote:

:On Jan 29, 5:57 pm, Ross Herbert <rherb...@bigpond.net.au> wrote:
:> On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum <ac...@torfree.net>:> wrote:

:
:> :Hi,
:> :
:> :I'm trying to help diagnose a problem; the application is a simple
:> owerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
:> :I was surprised to see significant charring on therelaycontacts, as
:> :therelayis rated for 30A.  Everything I have read about the topic
:> i.e. in this group + elsewhere) suggests that this effect only really
:> :happens with inductive loads.  Is it possible that a normal resistive
:> :load could cause high enough inrush currents to char theserelay
:> :contacts?   If so, I'd appreciate advice for selecting appropriate
:> :values for the resistor/cap (without having a scope available to track
:> :dv/dt).
:> :
:> :Thanks,
:> :  Adam
:
:> Could you supplyrelaymake/model details please?
:
:> Usually, only minimal arcing will occur onrelayoperation in situations such
as
:> yours. However, if the contacts are insufficiently tensioned, contact bounce
may
:> be prevalent, and considerable arcing may occur during operation. Also, if
therelayis too slow to release, contact break time may be excessive, thus
:> prolonging any arcing. The faster arelaycan both operate and release with
:> minimal contact bounce, the tendency to arcing will be reduced.
:
:It's a Finder relay, part# 66.82.8.110.0307 -- rated for 30A, DPST NO,
:switched on 110VAC with AgCdO contacts.  I wasn't the owner of the
:device in question when the first relay failed in this way, but now
:that I am, I'm trying to prevent it from happening again.
:
:http://www.findernet.com/en/products/detail.php?codice=668281100307&l....
:
:Adam

The data sheet indicates that the maximum AC current which  can be broken for
resistive loads (AC1 rating) is 22.7A (2500VA/110V), so provided that the load
is purely resistive (slightly inductive is OK), your stated approx 10 Ohms load
shouldn't stress the contacts at all. Assuming that the load is about 10 Ohms
resistive the maximum current required to be broken is 11A which is only 50% of
the rated maximum breaking current.

I would suggest that you remove the relay and jury rig it to operate from the
same AC voltage and connect a known 10 Ohm resistive load to operate via the
contacts - a 1.2kW resistive load would be an ideal test. If you don't see any
appreciable arcing on operation or release then this will indicate that in the
real life situation there is some other load factor (inductive) which you don't
know about.

If possible you might be able to use both contacts in parallel to share the load
current if all else fails.

Strangely enough, the manufacturer already did this with one pair of
contacts on each side of the load -- the circuit would work equally
well with a single pole relay. To be honest, I'm not sure why they
did this -- I was under the impression that when you parallel relay
contacts, unless the timing is perfectly synchronized (which it rarely
is), only one of them will be opening wet. If all else fails, I was
considering shorting one of the wire pairs + using only one set of
relay contacts for its lifetime and then switching to the other.

Removal + testing is of course a good idea, but I think the closest I
can get to that large a power resistor would require either some stove/
oven elements or a lot of lightbulbs in parallel

Adam


Adam

 

[Phil Allison]

"Adam Tenenbaum"



Removal + testing is of course a good idea, but I think the closest I
can get to that large a power resistor would require either some stove/
oven elements or a lot of lightbulbs in parallel


** Got an electric jug in the kitchen ??

Wot a goose.



...... Phil

 

[Phil Allison]

"Adam Tenenbaum"

** For 10 amps, you will need a very large cap to suppress arcing.

Something is fishy, cos the Finder relay you describe elsewhere is rated
for
100,000 operations at rated AC load.

The relay is cycling at about 1 min intervals, so if the device was
running for 4 hours/day, the expected life at maximum rated load (i.e.
based on 100,000 operations) is a little over a year.


** Christ almighty - there is NO WAY that device should have a damn relay
in it.

Ever heard of a thing called a TRIAC ????


I believe this
is about how long it lasted, but I guess I expected a bit longer life
(since it's running at half its maximum rated current) and a less
spectacular failure (completely black contacts).


** Alter your dumb expectations - pal.


Having said that,
shouldn't I still be able to reasonably extend the contact life /
suppress arcs with a simple RC snubber circuit paralleling the
contacts?


** No.

Now piss off.


....... Phil

 

[Tom Biasi]

"Adam Tenenbaum" <ac740@torfree.net> wrote in message
news:3acad9f1-8ff7-41a3-b27e-508c8487eb05@h5g2000yqh.googlegroups.com...

Hi,

I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A. Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads. Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts? If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

Thanks,
Adam

Adam,
What is the heating element used in?
Tom

 

[Adam Tenenbaum]

On Jan 30, 9:41 am, "Phil Allison" <philalli...@tpg.com.au> wrote:

"Adam Tenenbaum"

** For 10 amps, you will need a very large cap to suppress arcing.

Something is fishy, cos the Finder relay you describe elsewhere is rated
for
100,000 operations at rated AC load.

The relay is cycling at about 1 min intervals, so if the device was
running for 4 hours/day, the expected life at maximum rated load (i.e.
based on 100,000 operations) is a little over a year.

** Christ almighty  -  there is  NO WAY that device should have a damn relay
in it.

    Ever heard of a thing called a  TRIAC ????

I believe this
is about how long it lasted, but I guess I expected a bit longer life
(since it's running at half its maximum rated current) and a less
spectacular failure (completely black contacts).

** Alter your dumb expectations  - pal.

Having said that,
shouldn't I still be able to reasonably extend the contact life /
suppress arcs with a simple RC snubber circuit paralleling the
contacts?

**  No.

  Now piss off.

......   Phil

OK.. I'll bite the trollbait.
- Yes, I've heard of triacs -- my first reaction was to swap out the
relay with an SSR, but the owner/customer received a warrantied relay
replacement, so I was trying to mitigate the effect & extend the life
of this relay for as long as possible.
- No, I didn't design the original circuit, yes, it is dumb to use a
mechanical relay in this application, and also, this is my first time
dealing with relays switching large currents on AC mains.

Thanks for the "friendly" advice.. it seems that I can either keep
replacing the relay on a yearly basis or go with my gut + use the SSR.

Cheers,
Adam

 

[Adam Tenenbaum]

On Jan 30, 4:36 pm, "Tom Biasi" <tombi...@optonline.net> wrote:

"Adam Tenenbaum" <ac...@torfree.net> wrote in message

news:3acad9f1-8ff7-41a3-b27e-508c8487eb05@h5g2000yqh.googlegroups.com...

Hi,

I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A.  Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads.  Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts?   If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

Thanks,
 Adam

Adam,
What is the heating element used in?
Tom

Hi Tom,

It runs a steam boiler in a semi-commercial espresso machine.

Adam

 

[Jasen Betts]

On 2009-01-30, Adam Tenenbaum <ac740@torfree.net> wrote:

On Jan 29, 11:31 pm, "Phil Allison" <philalli...@tpg.com.au> wrote:
"Adam Tenenbaum"

The relay is cycling at about 1 min intervals, so if the device was
running for 4 hours/day, the expected life at maximum rated load (i.e.
based on 100,000 operations) is a little over a year. I believe this
is about how long it lasted, but I guess I expected a bit longer life
(since it's running at half its maximum rated current) and a less
spectacular failure (completely black contacts). Having said that,
shouldn't I still be able to reasonably extend the contact life /
suppress arcs with a simple RC snubber circuit paralleling the
contacts?

Adam

think about using a solid-state relay, they work really well with
resistive loads, and don't have the same wear mechanism.

 

[Jamie]

Adam Tenenbaum wrote:

On Jan 29, 11:31 pm, "Phil Allison" <philalli...@tpg.com.au> wrote:

"Adam Tenenbaum"




I'm trying to help diagnose a problem; the application is a simple
power relay switching a ~10 Ohm heating element load at 120VAC/60 Hz.
I was surprised to see significant charring on the relay contacts, as
the relay is rated for 30A. Everything I have read about the topic
(i.e. in this group + elsewhere) suggests that this effect only really
happens with inductive loads.

** Nonsense.

Breaking significant current causes arcing across the contacts.

Breaking the 120 volts AC supply to a 12 amp resistive load will create a
visible blue flash every time.

If it were a 120 volt DC supply, then the blue flash would not extinguish
and the relay would be destroyed in a few seconds.


Is it possible that a normal resistive
load could cause high enough inrush currents to char these relay
contacts?

** Breaking causes arcing.


If so, I'd appreciate advice for selecting appropriate
values for the resistor/cap (without having a scope available to track
dv/dt).

** For 10 amps, you will need a very large cap to suppress arcing.

Something is fishy, cos the Finder relay you describe elsewhere is rated for
100,000 operations at rated AC load.

..... Phil


The relay is cycling at about 1 min intervals, so if the device was
running for 4 hours/day, the expected life at maximum rated load (i.e.
based on 100,000 operations) is a little over a year. I believe this
is about how long it lasted, but I guess I expected a bit longer life
(since it's running at half its maximum rated current) and a less
spectacular failure (completely black contacts). Having said that,
shouldn't I still be able to reasonably extend the contact life /
suppress arcs with a simple RC snubber circuit paralleling the
contacts?

Adam
Replace it with a solid state relay.

http://webpages.charter.net/jamie_5"

 

[Ross Herbert]

On Fri, 30 Jan 2009 04:57:29 -0800 (PST), Adam Tenenbaum <ac740@torfree.net>
wrote:

:On Jan 30, 6:43 am, Ross Herbert <rherb...@bigpond.net.au> wrote:
:> On Thu, 29 Jan 2009 16:50:57 -0800 (PST), Adam Tenenbaum <ac...@torfree.net>
:> wrote:
:>
:> :On Jan 29, 5:57 pm, Ross Herbert <rherb...@bigpond.net.au> wrote:
:> :> On Thu, 29 Jan 2009 07:55:20 -0800 (PST), Adam Tenenbaum
<ac...@torfree.net>:> wrote:
:>
:> :>
:> :> :Hi,
:> :> :
:> :> :I'm trying to help diagnose a problem; the application is a simple
:> :> owerrelayswitching a ~10 Ohm heating element load at 120VAC/60 Hz.
:> :> :I was surprised to see significant charring on therelaycontacts, as
:> :> :therelayis rated for 30A.  Everything I have read about the topic
:> :> i.e. in this group + elsewhere) suggests that this effect only really
:> :> :happens with inductive loads.  Is it possible that a normal resistive
:> :> :load could cause high enough inrush currents to char theserelay
:> :> :contacts?   If so, I'd appreciate advice for selecting appropriate
:> :> :values for the resistor/cap (without having a scope available to track
:> :> :dv/dt).
:> :> :
:> :> :Thanks,
:> :> :  Adam
:> :>
:> :> Could you supplyrelaymake/model details please?
:> :>
:> :> Usually, only minimal arcing will occur onrelayoperation in situations
such
:> as
:> :> yours. However, if the contacts are insufficiently tensioned, contact
bounce
:> may
:> :> be prevalent, and considerable arcing may occur during operation. Also, if
:> therelayis too slow to release, contact break time may be excessive, thus
:> :> prolonging any arcing. The faster arelaycan both operate and release with
:> :> minimal contact bounce, the tendency to arcing will be reduced.
:> :
:> :It's a Finder relay, part# 66.82.8.110.0307 -- rated for 30A, DPST NO,
:> :switched on 110VAC with AgCdO contacts.  I wasn't the owner of the
:> :device in question when the first relay failed in this way, but now
:> :that I am, I'm trying to prevent it from happening again.
:> :
:> :http://www.findernet.com/en/products/detail.php?codice=668281100307&l...
:> :
:> :Adam
:>
:> The data sheet indicates that the maximum AC current which  can be broken for
:> resistive loads (AC1 rating) is 22.7A (2500VA/110V), so provided that the
load
:> is purely resistive (slightly inductive is OK), your stated approx 10 Ohms
load
:> shouldn't stress the contacts at all. Assuming that the load is about 10 Ohms
:> resistive the maximum current required to be broken is 11A which is only 50%
of
:> the rated maximum breaking current.
:>
:> I would suggest that you remove the relay and jury rig it to operate from the
:> same AC voltage and connect a known 10 Ohm resistive load to operate via the
:> contacts - a 1.2kW resistive load would be an ideal test. If you don't see
any
:> appreciable arcing on operation or release then this will indicate that in
the
:> real life situation there is some other load factor (inductive) which you
don't
:> know about.
:>
:> If possible you might be able to use both contacts in parallel to share the
load
:> current if all else fails.
:
:Strangely enough, the manufacturer already did this with one pair of
:contacts on each side of the load -- the circuit would work equally
:well with a single pole relay. To be honest, I'm not sure why they
:did this -- I was under the impression that when you parallel relay
:contacts, unless the timing is perfectly synchronized (which it rarely
:is), only one of them will be opening wet. If all else fails, I was
:considering shorting one of the wire pairs + using only one set of
:relay contacts for its lifetime and then switching to the other.
:
:Removal + testing is of course a good idea, but I think the closest I
:can get to that large a power resistor would require either some stove/
ven elements or a lot of lightbulbs in parallel
:
:Adam


It is often possible (with most relays) to make minor adjustments to ensure
close synchronising of parallelled contacts. The Finder relay appears to have a
sealed casing so you can't even see the contacts let alone make adjustments.
However, in an application such as yours minor overlap in contact timing
(parallel operation) would not cause arcing to be any more severe than would
occur due to normal contact bounce in my opinion.

As Phil suggests and electric kettle or jug would make an ideal test load.

 

[Ross Herbert]

On Sat, 31 Jan 2009 00:23:24 GMT, Ross Herbert <rherber1@bigpond.net.au> wrote:

It occurred to me that you could also try connecting both relay contacts in
series to increase the effective contact gap at opening which is where the
majority of arcing will occur.

A search of literature on arc suppression capacitors for a load such as yours
indicates that a value of about 13uF would be required. Since it would need to
have a rating of 300Vac this would make it bigger than the relay itself so it is
not really a goer in my opinion. The cap could remain charged to line potential
so that when the contacts closed again it would cause a high discharge current
which would result in an arc of its own. A series resistor of say 5 - 10
ohms/10W would need to be included in series with the cap to limit this
discharge current.

A search of patents revealed a fairly novel approach using a thyristor in
conjunction with a standard relay to suppress arcing. Note, you may have to
create a free account to view.
http://www.freepatentsonline.com/3555353.html