electrical question

lee · #1 01-23-2003, 08:12 PM

I'm not sure this is the right part of the forum, but it is a tech question, and it's on an SVX, so...

Can somebody tell me what some causes for relay chatter/bounce are?

I have placed one in my SVX to bypass the Climate Control signal to the ECU because that part of the ECU is dead (everything else seems to function proper). I'm using the signal from the climate system that would go to the ECU to trigger a relay that provides ground to the relay under the hood for the A/C clutch - just what the ECU would do - except I lose the WOT cut-out by my method - I never drive that way anyway.

So, back to my question, when the system first kicks on the A/C there is a sequence of chatter, or bounce in the relay that damps out in short order. Don't know if it will help, but I'll include a circuit diagram I made of what I've done. FYI, this is a voltage follower circuit as I wanted to protect the climate system and wasn't sure it could trigger a relay anyway.

So, could I put in some capacitor and eliminate, or at least dampen, this bounce? On which side, i.e., low current trigger side, or high current clutch ground side?

edit: and I did this for like $5 vice the hundreds for a new ECU in case somebody is wondering

Beav · #2 01-23-2003, 10:32 PM

Out of curiosity....

I'm sure you checked the voltage on your trigger, didn't you? Many devices only send a 5volt signal...

Trevor · #3 01-24-2003, 03:05 AM

Reasons for a relay to " chatter Bounce ".

I can reply with some authority on this one as I operated a company which manufactured relays i.e. Relays NZ Pty Ltd.

As Beav has in mind if the supply voltage is below that required to close the relay it can chatter on the verge of closing.

The operating voltage is not constant D.C. and the relay is following what amounts to an intermittent signal.

A relay designed for DC operation will chatter if AC is applied.

The relay is faulty but this is highly unlikely.

The power dissipation specified for an LM324 in a 14-DIP package is 1310 mW. I do not know the resistance of the relay you are using but from the numbered terminals you have shown I would suspect that it is an automotive type and you could be pushing things and could be better off using a transistor to limit current.

In any event you will have to confirm full and constant voltage to the relay coil. Don't become fooled by a slow reading digital meter and treat any quick variation in reading as suspect.

lee · #4 01-24-2003, 09:01 AM

Beav,

Yes I did measure the voltage, and its at battery level when the A/C on signal is present, i.e., 12+, not 5V.

Trevor,

It's not an automotive relay. I wasn't sure the climate system computer would deliver true battery/alternator voltage or have sufficient power even though the spec for that wire is 0V=off, battery=on. The relay is what Radio Shack calls a miniature power relay rated to trigger at 7 to 9 VDC, and handle a two amp load. I didn't show it all, but I'm using this circuit to control ground on the trigger side of the stock A/C relay - which is still there - not the power handling side of the stock A/C relay. Here's a web reference to the part:

http://www.radioshack.com/product.as...5Fid=275%2D005

The relay bounce only lasts for about 1 second - it's a very fast click sound, almost a buzz, then it stops and the system works fine in all other respects - if I didn't have it installed under the dash I would never know this was occurring. It's been doing it since I installed it several months ago, just now getting around to addressing the issue.

I know little about relays except their basic functions...so can I put a small bipolar cap between the LM324 and the relay to damp the AC part of the signal, or do you still think I need to investigate a transistor solution? If a cap will work, I'm not sure about size and effect on the output of the LM chip - could use advise there if known.

kuoh · #5 01-24-2003, 09:16 AM

I think you're a little mixed up on the theory of capacitor operation. Capacitors pass AC and block DC. If you were to place a capacitor between the op amp and the relay, the relay would never turn on. Filtering caps should be wired between the power and ground to shunt the AC ripple to ground. I forget the pinouts of the LM324, but from your schematic, I would place an 18V electrolytic between the LM324's power supply pin and ground. If there is no ripple going in, there shouldn't be any coming out, unless the signal from the climate control module is not a simple on/off.

If Trevor is right about the 1.3W power dissapation of the LM324, then that gives you about 100mA of current to play with. RS dosen't list the coil specs on the relay you linked, but it might be pushing the limits. You might try the micro relay instead. It has a 1A output, which should be more than enough to trigger any standard 12V automotive relay.

KuoH

lee · #6 01-24-2003, 10:13 AM

Quote:

Originally posted by kuoh
I think you're a little mixed up on the theory of capacitor operation.
KuoH

Just a little huh? You are so right and I seriously wonder were my brain goes sometimes, cause I sure can't seem to find it when I need it.

thanks

Trevor · #7 01-25-2003, 02:56 AM

Now that you have given all the facts it is apparant that everything appears OK as far as your circuitry is concerned and my previous advice was not requiired. I have been unable to check with Radio Shack as I am getting " Web server down or busy ". However the type of relay you describe would typically be approx. 200-300 ohms for use on 12v. DC. so that you should be Ok in respect of current ratings.

It would appear that the relay is performing normally and is responding to a DC signal which is pulsating for a short period at turn on. You could try a capacitor directly across the relay coil in spite of this not being the most effective choice. Something in the order of 470- 1000 uF wiil probably be required and this will result in considerable inrush current at turn on. This can be limited by placing a resistor of few ohms in series with the capacitor. I would like to see the OP amp saddled with this rather than the unknown signal as it is an inexpensive item and this follows your original thinking and reason for including the OP amp. in the first place. If we knew more about the signal source and its limits things would be much simpler.

Just why is the signal not clean that is the question. If it is, the OP amp is the suspect item. Can you get hold of a 'scope, or someone who has one, to check the voltage at the input and at the relay coil ? Or as I mentioned previously does a meter give any indication of pulsating DC at turn on ? You are not involved with AC you can be sure.

lee · #8 01-25-2003, 07:16 AM

I don't know where to get an OScope, but I will look around. It would be a nice thing to have anyway. My cheapo Digital VOM shows no fluctuation, but then the bounce is very fast and a second or less in duration, so.....

Somewhere, I saw a diagram that showed a symbol for the ECU role in all this. The Climate Control Module, sends battery level to the ECU to a transistor, which in turn then provides ground to the A/C relay unless it senses WOT. I know so little about electronics in practical terms that I cooked up this added relay and OpAmp setup to mimic what I thought occurred. I thhink I will try the Cap/resistor across the relay coil, that makes sense.

Thanks for all the help guys

Beav · #9 01-25-2003, 11:18 AM

If this is to ground the clutch coil directly you might want to use a bit larger relay as I'm reasonably certain the coil needs more than a couple amps.

kuoh · #10 01-25-2003, 01:45 PM

I think he said that it was the ground to the relay driving the clutch coil, in effect a relay driving a relay driving the clutch coil.

KuoH

lee · #11 01-25-2003, 06:43 PM

Yes, the latter

oab_au · #12 01-26-2003, 03:22 PM

Hi Lee, You like to experiment, More power to you.

I reckon that you have an inductive reactance problem with the coil on the relay. Turning coils on and off produce a reactance to the flow of current. Like an ignition coil.

As your chattering only lasts for a sec or so, it would seem that the coil 's inductance is resonating with a capacitance some where to produce an 'inductor capacitor resonate circuit', that will run till the energy is dissipated.

The fix is any thing that will upset the resonate circuit. The way I see it, would be to put a 'flywheel diode' across the coil of the relay. This will short out the oscillations and also protect the op-amp against voltage spike, when the coil turns off to prevent damaging it.

The diode is say, 1 amp 100v connected backwards across the coil terminals, with the anode connect to terminal 86 and the cathode connected to 85. This will short out any voltage reversal from the reactance.

Also as KuoH says, to put some filtering caps on the supply to the op-amp 0.1 uf, 0.01 uf and 0.001uf

Harvey.

lee · #13 01-27-2003, 07:08 AM

thanks one and all