Help with buzzing relay !!! Electrical gurus come in!!!

[Philstang]

This is on my truck but please help if you can...

I recently replaced the clutch fan with an electric fan. My problem is when my fan has been running and then turns off (controlled by temp sensor), the relays begin buzzing like crazy, when the fan comes back on, it goes away...

Please help!!!

 

[gt90stang]

Philstang,

Ok, relays are inductors and store energy. You need to clamp the energy so it doesn't arc contacts. This can be done by putting a diode across the coil reverse biased so that it dissipates the coils charge when it is de-energized. It is also possible that the switch that controls the relay is messed up from arcing because of no clamping diode.

Good Luck, Don

 

[HISSIN50]

Don, as always, has some great advice. If you search, terms like flyback diode or freewheeling diode will be helpful. Like Don said, the fan goes from a power-drawing fan to a generator once power is cut off [while it still rotates].

If you read the fan retrofit articles (M. Smith, et al) you will see the diode listed in the schematics (with part number). Or Don can probably just tell you what to run (I'd have to look it up because I dont have Don's knowledge). A Bosch 75 amp relay supposedly holds up better to the backfeed abuse (as well as having a higher load rating) compared to a 30 amp SPDT (if you happen to run the latter).

Good luck.

 

[Philstang]

yea, I am using a diode across the relay coil, silver stripe on the + side like seen in all the tech articles, it is still doing it, do I need a bigger diode or more than one? I am using two 30 amp relays in parallel to handle the taurus fan.

I am also using a diode across the fan with the silver stripe on the + side and the other side on the - side...

should I get the 85 amp glow plug relay from napa that the 4x4 guys are using?

 

[jrichker]

The buzz from the relay is most likely the heat sensor you use to trigger it isn't turning off like it should. A relay will buzz when a voltage less than that required to fully magnetize it is applied to the coil. If the thermostat switch doesn’t open up completely and allows a small voltage or leakage current to flow, then the relay will buzz.

To test this out, use a voltmeter applied to the thermostat switch side of the relay coil and ground. You should see only a small float voltage, less than .5 volt when the fan isn't running. More than 1 volt indicates a problem in the making. Either the diode used to clamp the relay coil is bad or miswired, or the thermostat switch for the relay is bad.

 

[Philstang]

I am using a flex a lite variable speed controller that is only rated at 25 amps. So I am using the two 30 amp relays in parallel after the controller to handle the tauraus fan amp draw.

Instead of running the controller (+) out to the fan, I am using that to activate my relays at the 85 connection, and the (-) running out of the controller is going to the relays at the 86 connection and to the fan (-) as a 10 ga, I only used a 16-18 ga wire on the relay controls at 85 and 86 because I believed it to only be the relays control.

I have 10 ga. wire fused from the battery to connection 30 of the relay awaiting to be activated, and 10 ga. coming off of the 87 connector that feeds into the fan (+)... again, there are diodes across the 85 and 86 of the relays and diodes near the fan across the (+) and (-).

Everything seems to be working fine except for the buzzing coming from the relays after the fans cycle off or are turned off...

Is there anything obviously wrong in this more detailed description of my setup???

Thanks for the input so far :SNSign:

 

[Philstang]

I will test with a mm today again to verify it but it seemed like I was getting weird #'s from it last night... If I remember the control side 85, 86 was getting 6 volts when the fan was off...

I think the controller sends out constant (+) but uses (-) as the on/off...

 

[soylentgreen]

Philstang,

Ok, relays are inductors and store energy. You need to clamp the energy so it doesn't arc contacts. This can be done by putting a diode across the coil reverse biased so that it dissipates the coils charge when it is de-energized. It is also possible that the switch that controls the relay is messed up from arcing because of no clamping diode.

Good Luck, Don

Relay coils do indeed store energy, but putting a diode on the coil will not do anything to the arcing on the relay's contacts. The only thing it does is protect the driver circuit that powers the coil. If you're using a switch to operate the coil instead of solid-state devices, I wouldn't use the diode (the contacts of the switch may be allowed to build up non-conductive films).

However, putting a diode on the relay's coil DOES increase the release time (time between when the coil is de-energized and when the contacts actually open). This is in the order of about 5 or 10 milliseconds...but the increased release time may contribute to premature failure of the relay's contacts.

Many relay manufacturers include a resistor internally to suppress the flyback voltage of the coil. In most cases, this is adaquate.

 

[gt90stang]

soylentgreen,

This is true but a properly sized motor starter relay can handle arcing. However in tuff cases a clamp diode or snubber is required across the load. This will reduce the peak voltage across the contacts. Sorry, I was distracted on my first post and ended up protecting the driver instead of the driven/driving component....hate it when that happens.

You may want to go with a single properly rated relay to drive the fan and protect it with a clamp diode.

YMMV, Don

 

[soylentgreen]

However in tuff cases a clamp diode or snubber is required across the load.

Ah, that's different! Suppression on the load will help dissipate the energy in the load at shut-off. Just make sure you're using a snubber up to the task. If it's too wimpy, it will probably smoke. That's not good...

 

[Philstang]

The buzzing has kind of gone away but I believe I will rewire my setup.

My controller sends out constant + and so I will run a fused + straight from battery to fan +. I will use the relays to control the ground side because the fan controller either grounds/ungrounds to power the fan.

Fan controller + out will go to term 85, fan controller - out will go to term 86, when the controller sends out the -, my relay will be energized and my fan will be grounded.

I believe this should work better than my current wiring.

oh, and I have a 70 watt relay on the way to get rid of the dual 30 ampers...

 

[jrichker]

If you are on a limited budget, try this…

It uses a relay to turn the fan on and off, and uses the computer's temp sense to trigger the fan. It isn't adjustable in terms of temp to turn the fan on and off, but it is cheap. All you need is some 10 gauge wire for the fan power & ground, a 30 amp relay, 30 amp fuse, and some 18 gauge wire for the relay coil wiring.

See http://fordfuelinjection.com/files/A9L_E-Fan_How-to.pdf for the full tech writeup. You will need the Adobe Acrobat viewer which is also a free download – http://www.adobe.com/products/acrobat/readstep2.html

 

[Philstang]

well, I figured out the source of my ticking... the flex a lite controller has a 60% startup and a/c setting. The 60% was probably achieved by the quick off/on off the controller output so that the fan would move at a lower than full power setting. However, this rapid on/off was causing a rapid on/off of my relays resulting in the perceived buzzing... time for a cheap thermostatic switch, I have spent too much already...

 

[jrichker]

well, I figured out the source of my ticking... the flex a lite controller has a 60% startup and a/c setting. The 60% was probably achieved by the quick off/on off the controller output so that the fan would move at a lower than full power setting. However, this rapid on/off was causing a rapid on/off of my relays resulting in the perceived buzzing... time for a cheap thermostatic switch, I have spent too much already...

Glad you found it. Maybe you might like to try the computer controlled solution I suggested. The additional cost is only a pin for the computer and the time to do the wiring since you already have the relays & other parts.