Relay chatter

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
jwax
Posts: 2150
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Relay chatter

Post by jwax » Sun Jul 22, 2007 8:20 am

I have a small Transmitter keyfob/receiver on around 300 MHz. The receiver takes a digital code from the TX, decodes, and closes a relay if the digital code is correct.
Next, I hook up a 12 volt motor to the relay contacts, and use the receiver 12 vdc to also activate the motor (100 mA).
Unless I put 100 MFD across the relay, I get relay chatter. Any way around using the big cap? I'm out of room. There is a diode across the relay.
At first I thought the motor current was pulling the supply voltage down, but I still get chatter with a big battery.
Any ideas?
John

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Relay chatter

Post by MrAl » Sun Jul 22, 2007 8:52 am

jwax wrote:I have a small Transmitter keyfob/receiver on around 300 MHz. The receiver takes a digital code from the TX, decodes, and closes a relay if the digital code is correct.
Next, I hook up a 12 volt motor to the relay contacts, and use the receiver 12 vdc to also activate the motor (100 mA).
Unless I put 100 MFD across the relay, I get relay chatter. Any way around using the big cap? I'm out of room. There is a diode across the relay.
At first I thought the motor current was pulling the supply voltage down, but I still get chatter with a big battery.
Any ideas?
John
Hi John,

When you say:
I put 100 MFD across the relay
do you mean you put the capacitor across the relay coil or across
the relay contacts?

Also, is there any way you can check to see if the receiver is putting
out a series of pulses instead of a single change of state?
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Post by Robert Reed » Sun Jul 22, 2007 8:52 am

Well, you are probably already aware of this, but the only source of chatter I can envision is a serious loss of DC feeding the relay circuit or a low frequency chopping of the DC on the drive to the coil. Have you scoped this out for a closer investigation? Seems like this should pin point the problem for further action.

User avatar
jwax
Posts: 2150
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Post by jwax » Sun Jul 22, 2007 9:58 am

I had first put the 100 MFD across the relay COIL, but then tried it across the contacts (motor). Coil connection works, across contacts did not.
I can use a smaller cap across the coil (10 MFD) if I add another cap, 0.1 MFD, across the motor!
Maybees this is why we usually see a tiny cap across small motors in CD players, tape drives, etc. I'm thinking the motor noise is confusing the digital decoding? Scrambling, in effect?
When there's chatter, all the scope shows is...chatter! Even the drive to the relay, hence me thinking I have a scrambler (motor noise) going on.
Appreciate your inputs, guys! :grin:

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Post by Robert Reed » Sun Jul 22, 2007 11:04 am

You could try disconnecting the motor temporaily while looking at the recieved level at the coil driver or look at same point with no recieved signal and then externally run motor and note any changes. If motor is indeed the problem - isolate the power supplies ( common coupling and ground loop ) or distance the motor from circuit ( radiated interference).

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Post by Chris Smith » Sun Jul 22, 2007 3:48 pm

I had a remote switch like yours built out of a toy car RC that did the same thing.

I suspected RF may have played a role as it seemed to stop if they were seperated by distance.

I just dampened the coil like you did.

User avatar
haklesup
Posts: 2934
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Sun Jul 22, 2007 8:01 pm

A couple thought s come to mind.

First, with an inductive load like a motor you can get a voltage spike for the exact same reason you would across the coil. A parallel diode would help with this and is applied in the same way (DC motor only of course). The cap would also help across the contacts but you say you needed it across the coil so,

Relay chatter is often a mechanical problem (physical bounce of the contacts). ALmost all relays suffer from some bounce. A large cap would slow the rise time of the coil voltage maybe cushioning the blow. Maybe, Its a reach I admit. Varying the coil voltage a bit may yield a voltage at which it performs better.

Frankly, just change out the relay for a different form and see how that goes. A reed relay may be better suited for less bounce.

The purpose of a small value cap across a motor is to supress EMI noise at its source (the motor). Rather than to clean the power going to it, It helps keep noise from the motor from contaminating the supply rails. If the motor is defective it may require an even bigger cap to supress noise.

On a scope, the critical tell would be if the chatter voltage remains inside the supply voltage or if it spikes to a higher value. Only inductive or capacitive interaction can get a higher voltage. Simple mechanical chatter or bounce would remain strictly between the supply voltage.

User avatar
jwax
Posts: 2150
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Post by jwax » Mon Jul 23, 2007 4:48 pm

I'll have to live with the 10 MFD across the coil (to hold the relay on during receiver drop out), and a 0.1 MFD across the motor to minimize noise-causing drop-out. Very tiny package to be mixing noisey motor with digital signals, but this will fly! Thanks all!:grin:

User avatar
Lenp
Posts: 1371
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Post by Lenp » Tue Jul 24, 2007 5:57 am

Years ago we had some equipment that suffered relay chatter from being induced on to the coil voltage. We 'fixed' it with a cap across the coil and a small adjustment to the relay's armature spring tension. Depending on the characteristics of the noise a diode in series with the relay coil voltage may be of value. Also, look upstream and add the capacitor to the relay driver input which is likely a transistor's base, A smaller cap may be acceptable. Adding caps to the relay contact side will do nothing unless there is noise being generated by the motor itself.

Last thought, if it is possible, change the relay to an AC type. AC relays have a shading ring on the magnet pole so that they don't chatter on AC.

Len

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Tue Jul 24, 2007 7:05 am

Hi,

I know this reply is comming in a little late, but just wanted to
add a few notes.

I have to agree with most of the other posts in that interference
can be causing the problem in the first place. There are various
ways the circuit can generate RFI which can cause positive
feedback which could cause the relay to be turned on and off
several times instead of just 'off'.
The relay coil should have at least the diode, but i think you have
that already. I think someone mentioned this already, but the
motor also has a coil of wire in it and once power is disconnected
(through the relay contacts) it would develop a nice big voltage
spike which could even arc the contacts, so a cap across the
contacts could help too, not only to prevent RFI, but to make
the contacts last longer also.
An ac relay sounds like an interesting idea, but then you would
have to have an ac supply voltage to turn the relay on, and that
would require something that could control ac as right now i think
your circuit only can control dc, so that would take a little triac
circuit or else another smaller dc relay. Also, the noise period
could be as short as one half cycle, in which case the ac coil
might end up being turned on and off just like the dc one, although
at the end of the half cycle (zero crossing) the relay would probably
stay off.

As long as it is working now i guess you're good to go :smile:
LEDs vs Bulbs, LEDs are winning.

User avatar
haklesup
Posts: 2934
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Tue Jul 24, 2007 11:17 am

I said:
The purpose of a small value cap across a motor is to supress EMI noise at its source (the motor). Rather than to clean the power going to it, It helps keep noise from the motor from contaminating the supply rails. If the motor is defective it may require an even bigger cap to supress noise.
That last sentence was an oversimplification. The criteria for selecting a bypass cap is partially based on the frequencies you are trying to shunt and the amplitude of the noise. You select the value to create a high pass filter with cutoff freq less than your noise and select the caps ESR to make that filter as low impedance as you can.

noise amplitude and cap value are not linked as they might be in the other application of supporting a sagging signal or supressing noise at the point of load.

One possibility to check out is that your power supply is under powered (WRT current) and it sags when the relay closes. Excessive resistance in the leads to the relay may induce a similar voltage drop. Also check for inductive crosstalk in your cable. If you have your relay control lines bundled next to logic inputs it could happen.

Another thought is 555 to shape that multiple pulse control signal into a single pulse of minimum width. But you said you were space limited so.........

What chip are you using to drive the motor? If its 74xx you might try a different family. ACT is a good higher power version of HC and HCT and has TTL compatible inputs.

One final thought, Don't ground one end of the relay and drive the other. Treat it like a differential signal and return the relay to a ground close to the driver chip. Send a second ground down the cable for anything else.

User avatar
jwax
Posts: 2150
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Post by jwax » Wed Jul 25, 2007 3:46 am

This is a very small, battery powered device- no AC available here.
The 12 VDC relay is driven by an 18-pin DIP IC.
Without the cap across the coil, and a cap across the motor, a large battery still results in chatter. The supply voltage is not dipping- the relay drive is being chopped by RF interruptions from motor noise. The caps fix that.

User avatar
Lenp
Posts: 1371
Joined: Thu Oct 26, 2006 8:11 pm
Location: Maryland
Contact:

Post by Lenp » Wed Jul 25, 2007 4:44 am

An ac relay sounds like an interesting idea, but then you would have to have an ac supply voltage to turn the relay on, and that would require something that could control ac...
]

Mr. AL,
My suggestion was to change only the relay not redesign the circuit.
An 'AC' relay will work on 'DC'. The DC coil current, for the same AC voltage is usually higher, and the pull in/drop out values are lower.
Alas, a DC relay operated on AC is called a 'Buzzer!

Len

User avatar
haklesup
Posts: 2934
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Post by haklesup » Wed Jul 25, 2007 11:31 am

What is the part number of that 18 pin DIP?

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Wed Jul 25, 2007 12:30 pm

Lenp wrote:
An ac relay sounds like an interesting idea, but then you would have to have an ac supply voltage to turn the relay on, and that would require something that could control ac...
]

Mr. AL,
My suggestion was to change only the relay not redesign the circuit.
An 'AC' relay will work on 'DC'. The DC coil current, for the same AC voltage is usually higher, and the pull in/drop out values are lower.
Alas, a DC relay operated on AC is called a 'Buzzer!

Len
Oh ok Len, that's cool, but then why would you expect the
ac relay to behave that much different than the dc one?
LEDs vs Bulbs, LEDs are winning.

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 5 guests