Stopping relay chatter

Thread Starter

gte

Joined Sep 18, 2009
357
I have an rpm controlled switch that triggers the ground side of a 30 amp relay, which in turn controls a solenoid.

My problem is that when I get into a higher gear, the rpms oscillate a little while accelerating and this in turn oscillates the relay and the solenoid which triggers a check engine light.

The RPM switch is store bought, the solenoid is OEM, so the only thing I can think of doing is adding a little hysteresis to the rpm switch output (ground signal) before it connects to the relay's ground coil trigger.

How can I do this? Do I need a circuit or will a simple capacitor do?


Here is a very crudely drawn diagram


 

SgtWookie

Joined Jul 17, 2007
22,230
Well, there are some mistakes in your schematic.
You indicate in your text that the RPM switch triggers the ground side of the relay. That means both sides of the relay's coil are connected to ground when the RPM switch closes.

One side of the solenoid is connected to +12v, and the the other to the relays' NO 87 terminal. However, the 86 (COM) terminal is also connected to +12v instead of ground. I think you may have meant to connect the 86 terminal to ground?

The hysteresis needs to be built in to the RPM switch.

I suppose one could use a timer to introduce a fixed delay after the initial signal, but that would be crude, and not a very good fix.

Perhaps you could look at the root cause; why is engine RPM fluctuating? Is it due to excessive drivetrain windup, or is it due to whatever happens when the solenoid engages?
 

Thread Starter

gte

Joined Sep 18, 2009
357
Well, there are some mistakes in your schematic.
You indicate in your text that the RPM switch triggers the ground side of the relay. That means both sides of the relay's coil are connected to ground when the RPM switch closes.
I'm confused, pin 86 of the relay (coil trigger) gets B+ whenever the ignition is on

Pin 85 (negative side of the coil) gets a ground signal whenever a certain rpm is reached. When the coil is energized, pin 30 (which is always grounded) completes the circuit to pin 87, which grounds the solenoid.

Is that not what I drew? Because I've went over it a couple of times, and I think that's what I drew? The rpm switch and relay share a common ground through the chassis of the vehicle, just like it shares a common B+ with the solenoid and relay, which could have been drawn a little different in the diagram I guess?

The engine rpm fluctuates because the car does not accelerate very fast at that rpm/gear and because it appears that the rpm switch doesn't have the proper hysteresis built in ... not because anything is wrong with the car mechanically. I think that if there was a way to build in a cushion so that if the relay oscillates, the coil stays energized, I'll be golden.

Can't a simple capacitor that could discharge when needed do this? Or are you saying that a circuit has to be added?

One side of the solenoid is connected to +12v, and the the other to the relays' NO 87 terminal. However, the 86 (COM) terminal is also connected to +12v instead of ground. I think you may have meant to connect the 86 terminal to ground?

The hysteresis needs to be built in to the RPM switch.

I suppose one could use a timer to introduce a fixed delay after the initial signal, but that would be crude, and not a very good fix.

Perhaps you could look at the root cause; why is engine RPM fluctuating? Is it due to excessive drivetrain windup, or is it due to whatever happens when the solenoid engages?
 

SgtWookie

Joined Jul 17, 2007
22,230
I'm confused, pin 86 of the relay (coil trigger) gets B+ whenever the ignition is on

Pin 85 (negative side of the coil) gets a ground signal whenever a certain rpm is reached. When the coil is energized, pin 30 (which is always grounded) completes the circuit to pin 87, which grounds the solenoid.

Is that not what I drew? Because I've went over it a couple of times, and I think that's what I drew? The rpm switch and relay share a common ground through the chassis of the vehicle, just like it shares a common B+ with the solenoid and relay, which could have been drawn a little different in the diagram I guess?
OK, I'm in error. I got confused by the way the schematic was drawn. Sorry about that.

The engine rpm fluctuates because the car does not accelerate very fast at that rpm/gear and because it appears that the rpm switch doesn't have the proper hysteresis built in ... not because anything is wrong with the car mechanically. I think that if there was a way to build in a cushion so that if the relay oscillates, the coil stays energized, I'll be golden.

Can't a simple capacitor that could discharge when needed do this? Or are you saying that a circuit has to be added?
Well, it would have to be a pretty large electrolytic capacitor, and it would cause a large peak load on the RPM switch when it first provided the ground.

What are the specifications on the relay? Coil resistance, pulling voltage, dropout voltage?
 

Thread Starter

gte

Joined Sep 18, 2009
357
Sorry, my drawing skills aren't the best :)

The relay is your standard bosch relay

How do I get the voltage specs you ask for? I can go out and measure the coil resistance if you'd like




OK, I'm in error. I got confused by the way the schematic was drawn. Sorry about that.



Well, it would have to be a pretty large electrolytic capacitor, and it would cause a large peak load on the RPM switch when it first provided the ground.

What are the specifications on the relay? Coil resistance, pulling voltage, dropout voltage?
 

SgtWookie

Joined Jul 17, 2007
22,230
The simple circuit given at the link below, could be easily implemented and solve your problem.

http://www.google.it/imgres?imgurl=...lInUDQ&sa=X&oi=image_result&resnum=6&ct=image
Alberto
That circuit originally came from Bill Bowden on his Hobby Circuits website.
His site has moved; now that circuit is located here: http://www.bowdenshobbycircuits.info/page2.htm#relay_i.gif

According to the circuit description, it's intent is to provide a delay between when 12v is first applied and the energizing of the relay's coils.

It would require a good bit of re-work before it would meet gte's requirements.
 

Thread Starter

gte

Joined Sep 18, 2009
357
:(

And happiness was short lived, haha



That circuit originally came from Bill Bowden on his Hobby Circuits website.
His site has moved; now that circuit is located here: http://www.bowdenshobbycircuits.info/page2.htm#relay_i.gif

According to the circuit description, it's intent is to provide a delay between when 12v is first applied and the energizing of the relay's coils.

It would require a good bit of re-work before it would meet gte's requirements.
 

SgtWookie

Joined Jul 17, 2007
22,230
OK, here's a different version that's a cousin to Bill Bowden's circuit. See the attached.

V2, R98 and Q98 are there only to simulate the output of your RPM switch, which at the moment I'm assuming is open-collector. If it is not an open-collector output, you will need to add a diode between the RPM switch output and this circuit.

R1 limits the maximum current that your RPM switch has to sink when it turns on.

R3 limits the current through the base of Q1. Q1 and Q2 form a Darlington pair, which source current to the relay's coil.

C1, R2 and R3 provide an RC time constant. When the RPM switch trips and starts sinking current, C1 is discharged through R1. This happens very quickly.

When the RPM switch opens up again, it takes about 2 seconds for C1 to charge via R2 and R3, until Q1 and Q2 turn off, thus cutting off the current to the relay coil.

D1 is across the relay's coil to absorb the reverse voltage spike that occurs when current through a coil is suddenly stopped.

R99 could represent the coil connections for your solenoid - but it would need to be across the NO terminals instead of the NC terminals. It will also need a diode connected across it, like the relay's coil.
 

Attachments

Thread Starter

gte

Joined Sep 18, 2009
357
Thanks!

I will try and build this (maybe this weekend?) as soon as a I get a chance!




OK, here's a different version that's a cousin to Bill Bowden's circuit. See the attached.

V2, R98 and Q98 are there only to simulate the output of your RPM switch, which at the moment I'm assuming is open-collector. If it is not an open-collector output, you will need to add a diode between the RPM switch output and this circuit.

R1 limits the maximum current that your RPM switch has to sink when it turns on.

R3 limits the current through the base of Q1. Q1 and Q2 form a Darlington pair, which source current to the relay's coil.

C1, R2 and R3 provide an RC time constant. When the RPM switch trips and starts sinking current, C1 is discharged through R1. This happens very quickly.

When the RPM switch opens up again, it takes about 2 seconds for C1 to charge via R2 and R3, until Q1 and Q2 turn off, thus cutting off the current to the relay coil.

D1 is across the relay's coil to absorb the reverse voltage spike that occurs when current through a coil is suddenly stopped.

R99 could represent the coil connections for your solenoid - but it would need to be across the NO terminals instead of the NC terminals. It will also need a diode connected across it, like the relay's coil.
 
Top