Leverhead
Forum Newbie
Posts: 37
Loc: New Mexico
Reg: 02-05-11
|
07-30-12 12:10 PM - Post#2253764
I've been researching on the forum about relays with diodes and I'm questioning the need for a coil clamping diode on a relay if that relay is switched by ground. In other words, if the relay coil is being fed a 12v source from the fuse block/ignition switch and the ECU is switching it to ground (closing the circuit) to turn the fans on, and then the fans are switched off by the ECU opening the ground circuit, doesn't the 'spike' normally reverse polarity and try to flow back towards the ignition switch...not the ECU? If so, then how does a diode protect the ECU from flyback? Also, how much voltage are we talking about in a 12v system?
I'm wondering because I have my spal fans/tyco relays set up to switch this way thru the HP ecu input/output connector (ground switched) and wondering if I need to swap out the relays for diode suppression types or jump the coils with a diode from radio shack or something.
Also the loose green fuel pump wire is switching a relay for the fuel pump instead of direct connect so I'm thinking a diode on that one. This one is direct 12v from the ECU so I can see a diode there. Nothing in the documentation about this which is interesting.
Thoughts?
|
|
65_Impala
Very Senior Member
Posts: 1607
Reg: 12-29-02
|
07-30-12 04:43 PM - Post#2253880
In response to Leverhead
In that case, the voltage spike goes into the ECU. However, the ECU should have protection for the relay driver circuit. If the manual doesn't mention needing a diode then it's likely OK without one. In the big picture, Any ECU pin should be properly protected against more than the typical noise and voltage spikes that can be expected. It still never hurts to be better protected by using a relay with the diode though.
|
Leverhead
Forum Newbie
Posts: 37
Loc: New Mexico
Reg: 02-05-11
|
07-30-12 05:34 PM - Post#2253915
In response to 65_Impala
Thanks for the input. But then how does the spike flow into the ECU if it is flowing backwards towards the Ign switch. The ECU power is direct to the battery. The fan trigger power is direct from the Ign switch.....only the ground side of the fan trigger wire (from the relay coil) is connected to the ECU harness. Still confused.....
|
mike beck
Contributor
Posts: 218
Reg: 05-20-10
|
07-30-12 07:03 PM - Post#2253980
In response to Leverhead
The energy of a collapsing field wants to go somewhere. Better to direct it then to let it wander-freely. By directing it, use a diode or a resistor to clamp it. A typical relay has a very small and short duration pulse coming out of it's coil, but why take the chance.
|
65_Impala
Very Senior Member
Posts: 1607
Reg: 12-29-02
|
07-31-12 03:17 PM - Post#2254250
In response to Leverhead
The spike isn't "flowing" anywhere or in any particular direction. A large negative voltage is generated across a coil when the power is suddenly interrupted. In your case, the voltage spike will be measured between the positive wire and the input of the ECU. Since one end of the coil is connected to the battery voltage which can't really change this means the ECU pin gets the voltage spike applied to it. It will be a voltage that rises way above the battery voltage level if it's not clamped by a diode or other surge supression device.
Most likely, the ECU has an internal diode that clamps the voltage to the battery voltage level instead of allowing it to damage the transistor inside.
|
Leverhead
Forum Newbie
Posts: 37
Loc: New Mexico
Reg: 02-05-11
|
07-31-12 05:25 PM - Post#2254288
In response to 65_Impala
Thanks again for taking the time to explain.
So what I'm getting is that since it is ground switched and the 12v source is never turned 'off', when the ground is 'opened' to turn off the fans and the field collapses, the reversed polarity spike cant't flow towards the 12v side because there is still voltage there, so it flows back towards the ECU ground?
Interesting since other text I have read on this says because the spike is reversed polarity and the spike is a higher voltage then 12v it will overwhelm the 12v and flow back towards the source not towards the ground side. This is confusing.
I may just replace my relays with diode supression types or add diodes as a failsafe.
|
mike beck
Contributor
Posts: 218
Reg: 05-20-10
|
07-31-12 07:17 PM - Post#2254315
In response to Leverhead
I use these everywhere, just throw it in, not polarity-sensitive.
http://www.newark.com/fairchild-semiconduct or/p6ke...
|
Leverhead
Forum Newbie
Posts: 37
Loc: New Mexico
Reg: 02-05-11
|
07-31-12 09:46 PM - Post#2254373
In response to mike beck
Thanks Mike. I have some 1N4001 (50v 1A) already. Seems like these are a popular choice for coil clamping diodes. Only problem is my fan relays are the sealed weatherpack type where the terminals arent accessable so I will probably have to go with internally suppressed type relays. Regards.
|
65_Impala
Very Senior Member
Posts: 1607
Reg: 12-29-02
|
08-01-12 03:09 PM - Post#2254606
In response to Leverhead
Lets look at it a different way. When the coil is energized it stores a certain amount of energy. This energy will try to leave the coil when you disconnect one of the coil leads. Since one lead isn't connected to anything, there is a large voltage produced across the coil as the energy "searches" for a way to get out of the coil. Eventually, the voltage goes high enough that the energy finds a path and current begins to flow.
This current flow requires a completed circuit or the energy must flow as current between the 2 pins of the coil. Now, relatively speaking, the positive wiring is heavier and connected to the battery and other electrical stuff so it's a relatively difficult wire to force a voltage change onto. The pin of the ECU isn't connected to anything directly (remember the circuit in the ECU is off) so it's a relatively easy wire to force a voltage change onto. So, as the energy "searches" for a way out and produces a high voltage across the coil, the ECU pin sees the brunt of the voltage spike since it's an easier circuit to force a voltage change onto.
The energy dissipating will circulate as current from the coil to the ECU pin and then through the ECU ciruit to positive or negative power ciruit where it then circulates back to the coil. OK, the actual direction of current flow varies according to the use of classic or electron current flow but that really doesn't matter.
The above isn't to say that the coil doesn't affect the positive wiring at all. The energy dissipating can cause a voltage transient on that circuit as well. All I'm saying is that the ECU pin will see more of the voltage spike. What will happen is the ECU pin will rise until the protection inside the ECU operates and then current begins to flow. The current making the circuit from coil - ECU - positive wiring - coil can then cause the voltage transient on the positive wiring ciruit. The wire as well as every other electrical part such as switches and connectors and the battery has resistance so any current flowing will cause a voltage across the wire and other parts.
A voltage spike doesn't flow. It is just a voltage developed between 2 terminals.
Edited by 65_Impala on 08-01-12 03:15 PM. Reason for edit: No reason given.
|
|
Gain extra benefits by becoming a Supporting Member 
ThomasG
