Why doesn't my electrical relay work?

---

Request for Question Clarification by aht-ga on 09 Feb 2004 12:11 PST

Some basic clarification requests, to see if there is any possibility
of helping you:

- what is the electrical relay used for?
- what is the relay connected to?
- how do you know it isn't working?
- what have you already tried for troubleshooting?
- are there any specifications you can provide for the relay?

---

Clarification of Question by hparti-ga on 09 Feb 2004 12:32 PST

The relay is part of a solar heating system.  It is connected to the
fan in the plenum (forced hot air).  Sometimes (not every time) the
fan does not turn on when the thermostats are calling for heat
although other portions of the system seam to know that the thermostat
is calling for heat (Temptrol damper, humidifier).  When the problem
occurs if I tap on this electrical relay with the back end of a
screwdriver I can hear the relay click and in turn the fan comes on
and the problem is solved.  I have replaced the electrical relay with
a brand new relay of the exact same type.  The new relay works the
same way as the old relay did.  I do not have the exact specs on the
relay here at work but I have them at home so I can get them - it is
an older (1987) 120V Industrial grade relay - that is currently made
by Tyco but used to be made by another company.

---

Request for Question Clarification by aht-ga on 09 Feb 2004 13:16 PST

hparti-ga:

Based on what you have described, there are a couple of potential
causes that come to mind right away:

- environmental conditions are beyond spec: most relays are rated to
perform within certain temperature and humidity ranges, you should
check to make sure that your relay is rated for the environment it is
physically located in

- insufficient control power: the solenoid inside the relay that
actually mechanically throws the contacts from open to closed,
requires a certain voltage/current to operate. You mention that the
relay is 120 VAC, what is the control voltage and can you be certain
that the control voltage is correct? Even if the relay works when
brand-new, age, wear, and corrosion can cause the power requirements
to go up over time in order to throw the contacts, so if the
controller that is sending the signal to the relay is outputting at
the lower range of the relay's rated control signal requirements, then
over time it will stop activating. Hitting it with a screwdriver is
just enough to unstick the solenoid.

Do you know anything about the controller that is using this relay to
activate the fan?

Regards,

aht-ga
Google Answers Researcher

---

Clarification of Question by hparti-ga on 09 Feb 2004 13:54 PST

My next step was going to be to get a voltmeter out and check voltage
and amperage at the relay.  I assume the low amperage side is the side
I need to measure, and I also assume that the ratings on the relay
will tell me what it is expecting. Right?

I am not 100% certain as of yet but I believe the Temptrol controller
is what is turning the relay on/off.  I am not 100% certain because
there is some chance that there is something in between.  Another
thing I was going to try was replacing the Temptrol controller - I
have 2 heating zones each with its own Temptrol controller so I was
going to try to switch them to see if the problem disapears.  Another
potentially interesting piece of information is that if I switch the
master thermostat from auto-fan to fan-on the fan does turn on - my
only assumption here has been that this is turning on the fan some
other way but I guess it could just be going through another path
(also connected to the relay) that actually works consistently.

To answer the 2nd question posted from leoj-ga - I replaced the relay
because I thought it had to be the problem and was either just old
and/or broken.  This was also the information that 2 different heating
technicians had given me - figure out which relay is having a problem
and replace it...

---

Request for Question Clarification by aht-ga on 09 Feb 2004 14:07 PST

The heating technicians speak from experience; relays are the usual
culprit in this sort of situation, they will eventually wear out
anyway, and are usually the cheapest component to replace.

If you are able to manually and repeatedly activate the fan from the
Temptrol controller without any problems, then chances are that the
problem is in the Temptrol controller itself. As leoj-ga mentioned in
his comment, the cause could be anything from a marginal wiring
connection to a bad solder joint. Switching the controllers will help
to diagnose if the problem is only a marginal wiring connection (in
which case, installing/uninstalling the controllers will help result
in you tightening up the connections anyway), or if it is a fault in
the controller itself. Check the voltage across the contacts of the
controller, then again at the contacts of the relay. If there is a
wiring problem, you should be able to note a marked difference.
Checking current flow requires that your ammeter be in-line with the
circuit, so easiest to check this at the controller. Compare the
current flow when you manually turn on the fan, to the current flow
when the controller turns on the fan automatically (get a can of
compressed air to help you cool down the controller so that it
automatically turns on the fan).

May I ask what you are seeking in terms of a proper Answer to your
question? The answer to "why doesn't my electrical relay work" is
essentially all of the causes mentioned above... but I'm guessing that
isn't necessarily what you are looking for now! Please advise how you
wish for me to proceed in Answering this question for you.

Thanks,

aht-ga
Google Answers Researcher

---

Clarification of Question by hparti-ga on 09 Feb 2004 14:29 PST

I think most of what I was looking for was in your last response.  I
am really looking to move backwards from the relay now to determine
the problem.  I want to be able to measure something at the input side
of the relay to confirm that something before it is off - is it
voltage or amperage (both?) that actually causes the circuit to close
in a relay?  Is it obvious from the specs on the relay what it expects
for both?  I think the answers to those questions give me enough
knowledge to continue to troubleshoot.

Thanks.

hparti-ga:

I can definitely help you with the last two questions.

First, here is a link to the "How Stuff Works" website, where
amazingly enough there is an article about relays:

How Relays Work
http://electronics.howstuffworks.com/relay.htm


And, here is a webpage with more information than you should ever want
to know about solenoids:

Solenoid Technical Data
http://www.solenoids.com/solenoid-technical-data.html


In summary, there is a solenoid (wire coiled around an iron core to
amplify the magnetic field) inside the relay. The magnetic field
created by this solenoid depends upon the voltage applied across it,
and the current that flows through it. More specifically, the current
flow is more important than the voltage, but since voltage and current
are interrelated by the impedance of the design, both matter.

On a typical industrial relay, the design voltage and current for the
control side, and the maximum voltage and current for the load side,
should either be marked directly on the relay casing, or should be
something that can be looked up based on the relay's model
designation. The control requirements should be lower than the load,
but this is not always the case (relays are sometimes used simply to
isolate a control circuit from the rest of the circuit).

Here are some examples:

http://relays.tycoelectronics.com/datasheets/gp.stm
(click on the datasheets for some of these units to see how they are marked)


If you look at the datasheet for the KUP93 series, for example:

http://relays.tycoelectronics.com/datasheets/GP_Power_Relays/KUP93DS.pdf

You will see a chart that shows the 'nominal coil voltage', the 'must
operate voltage', and the 'nominal coil current'. What these mean, is
as follows:

nominal coil voltage - the design voltage for the coil, control source
should be at this voltage

must operate voltage - the coil will still operate as long as the
control source voltage is at least this voltage, but anything below
this will be insufficient

nominal coil current - the amount of current that flows through the
coil at the nominal coil voltage


Not all relays will have this information easily accessible, but you
may have some luck looking up the information on the Tyco website
above. As well, all relays should come with this sort of application
specifications either printed on the casing, or as a paper insert in
the box that it came in.

You will want to look up the voltage and current requirements of the
relay, then verify at the relay that there is sufficient voltage
(measured across the coil) and sufficient current (measured in-line
with the coil, meaning you will need to disconnect one of the wires
leading to the coil, and carefully wire it through the ammeter to the
relay.

Good luck with this, and I hope that this information helps!

Regards,

aht-ga
Google Answers Researcher

This begs the question:  Why did you replace the old one?

If it is the same type, but different manufacturer, or even sometimes
from the same manufacturer, its not working in the same situation that
the other one did is an indication that the wiring leading to it is
somehow marginal - they probably changed something to save cost that
keeps it within the original spec, but not as good as the previous
one.  If your drive circuit was only just barely switching the old
one, then it would be reasonable for it to have issues with the new
one.

The thing you need to figure out is why is it having trouble.  It may
be a marginal design, it may be some loose terminals in the path
creating excessive resistance, or it may be a defective part in the
circuit.  Without examining the hardware, no one can tell you
conclusively which.  My advice would be to do what you can - make sure
all  the terminals are properly tightened, or if they are soldered
that there is no evidence of a cold joint or a broken or partially
broken wire.

Also, a good way to find a high resistance joint is with one's finger
inside a latex glove.  Hot things are bad.  :)  (Make sure you only
touch one place at a time and are not grounded either.)

Good Luck.