When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Ok, I'm really confused with this, so I need someone to explain it to me.
As far as my understanding of electronics extends, a diode is essentially a one-way check valve for electricity. It's purpose is to prevent backfeeding of voltage.
I have two LED's in my instrument cluster for the cooling fans. Because I'm a control freak and like to know when things happen, so I can make sure everything is working properly. They are powered by the common gauge power feed (12V ignition power) and they have no ground until either the PCM grounds the fan relay coil, or I throw the manual override switch, grounding the relay coil. The LED ground is spliced into that common ground wire.
When the ignition is on, everything works the way I intend it to, and as I just described above. When the ignition is off, voltage is flowing across the coil and down the ground wire and illuminating the LED's with reverse voltage somehow. These relays are wired according to the 2001 Camaro LS1 diagrams that I have, with both the 86 pin and the 30 pin of the relay being 12V constant from the battery.
Figuring that there was something special about LED's as opposed to another type of diode, I soldered in a blocking diode between each LED and their control switch to prevent the reverse voltage. All this did was serve to dim down the LED's when the ignition is off. They continue to illuminate even if I disconnect the power wire, so they have no complete circuit at all.
I'm fairly good with electricity, but I'm not an electrical engineer, so what am I doing wrong here? I can't switch the fans to ignition power without a massive rewiring of the fuse and relay box, which I'd really prefer to avoid if possible.
If you examine a diode and it's properties what you'll see is that in the "forward biased" direction of current flow the diode's resistance is quite low . Not 0 ohms mind you (0 ohms being a direct connection, a straight wire) but low , plenty low enough resistance to allow some healthy current flow . Now , in the "reverse biased" direction of current flow the diode's resistance is quite high . Not infinite ohms (no connection whatsoever , an open switch for example) but instead some high number of ohms in the kilo ohm or meg ohm realm , not enough of a connection to pass any appreciable current , , but , maybe just enough current flowing back through the circuit to light the led (which uses a TINY amount of power to light) due to this reverse leakage of current through the diode your attempting to use as a blocking diode .
I hope that made some sense of why the LED would light even with the additional diode in the circuit attempting to extinguish it . No diode is 0 ohm forward nor infinite ohms backward ....
You say if you disconnect the LED power it still lights? Can you post the diagram you are using and your additions to it? Orange is right about diode leakage but there still needs to be a complete circuit for that to happen. What LEDs are you using? How are they mounted?
Ok, So I can understand the leakage part, how it is still getting voltage even with a diode in line. What I don't understand is how it is lighting up without a complete circuit. I just drew my additions into the GM Diagram, as it was faster than working from scratch, and I followed the diagram in wiring up the relays.
My additions are pretty much at the bottom.
As to how the LED's are mounted, they are mounted into a vinyl covered aluminum sheet that I fabricated to replace the original gauge cluster/trim. That sheet is screwed to the original gauge trim mounting points in the dash. I need to verify again, but I believe when this first started I checked for continuity between the metal diode case and the ground lead and there was no connection. I will attempt to get out to the car tonight and re-verify this.
This is the only way that I can think that they would ground, as I'm using a common ground bus for the whole instrument panel and other LED's such as the High Beam indicator and turn signals ground through there. If there is continuity from the ground lead to the diode case on any of those LED's that lead to the ground bus that could complete the circuit. The aluminum panel itself is not grounded.
I don't believe this is the case, as my brake LED functions correctly and goes out when I release the parking brake and deny it ground. But then it also only illuminates when the ignition is on...
The two LED's in question are the ones over the temp gauge that are not illuminated in this picture, as the ignition is on and I'm assuming that the current flowing in the right direction is drowning out what is coming in reverse.
The diodes in that circuit are accomplishing exactly NOTHING.
The problem with your circuit goes deeper than just what you're complaining about. The REAL thing wrong with it is, they aren't indicating whether the fans are running; they're only indicating whether the switch is flipped or the ECM is calling for power... NOT whether power is being fed to the motors. IOW if the relay or its connectors goes bad, the lights will light even if the fan is not getting power.
Fix the logic and the other problem will take care of itself.
The LEDs need to be wired with one of them's anode on wire 409 and cathode on wire 532; and the other with anode on 504 and cathode on 150. I would suggest being VERY CAREFUL about how you make those connections, such that you disturb the factory wiring as little as possible; for example, I would NOT cut the big fat wires and use butt splices, and I would NOT use Scotchlocks. I would strip a half-inch or so of insulation off of each WITHOUT CUTTING ANYTHING, wrap the wires to the LEDs around them, solder, and cover with self-sealing electrical tape. https://www.mcmaster.com/#7643A75 or https://www.mcmaster.com/#7643A71, your preference as to whether you want the guide line or not, and in whatever color (I'd recommend black or grey).
Wired like this, they will light dim when the fans are in low speed (the 2 motors in series and each only getting 7V or whatever); and each will light bright when its fan is getting the full 14V and therefore is in high speed. The logic there of course is, dk grn grounded but not dk blu, both fans on low speed; dk blu grounded but not dk grn, right fan only on high speed; both wires grounded, both fans high. The LEDs will light dim or bright correspondingly if connected as described.
I would also STRONGLY recommend putting a temp switch in the circuit somewhere. I think my preference there would be, the switch in the stock location, in the pass side head; the lead going to the cathode of 2 diodes; the anode of one diode going to the dk grn wire and the anode of the other going to the dk blu one. That would turn both fans on high when the switch operates. I'd suggest finding a temp switch that's higher than any normal operating temp, but still below the "danger" zone; a stock 235°ish one might work OK but also might make the fans come on too much. If you could find one at around 250° or so, that would probably be a good solid "safety" backup, that would only come on at the utterest need.
No I'm not really an electrical engineer but I used to play one on TV. (hahaha... I was the chief engineer for radio & TV stations for decades but I'm really a mathematician and physicist)
Last edited by sofakingdom; 06-20-2018 at 07:38 AM.
06-19-2018, 07:48 PM
What your running into is , "leakage" . 
