Ok so I wanted to install a little blinking led for a fake alarm. And I wanted it to go off when the truck is running and on when it stops so first I found a way to do it with a relay but it seemed a little complicated. Then I found a second option. All I needed to do was put the ground wire from the LED to a constant power source in the fuse box and the positive to a run only power source. Hooked it up and to my amazement it works. Turn the truck off and the light starts blinking, turn it on and it shuts off. PERFECT!

so here is my question, both of the fuse sources are obviously positive. How the hell does this circuit work lol? In the video I watched they called it a floating ground.

 

When the truck is on, both sides of the LED are at or very near the same potential = no light.

When the truck is off, no more hot on the switched side BUT there is still a path through the loads on that circuit to ground. For most any circuit on a vehicle, a single LED would have a tiny current flow and the largest voltage drop compared to another load combination in series with it to ground. So it sees the voltage drop of the other load(s) in the ground path as very small, near ground level. Constant hot on one side, near ground on the other = LED is powered.

 

yup. electricity needs to flow like a river. With power on both ends it just sits there and no flowing happens.

 

Just wanted to share... this is what I did, even though your explanations of why this thing is working is still a bit confusing lol.

 

Very cool! I never knew about a "floating ground". Thanks

 

If anyone cares the two circuits I used were #17 (turn signal/hazzards) for my constant power source and #10 (instrument cluster) for my run only power source. Again, NO ground.

 

It has a path to ground, just not in the way we usually think of it.

 

When the switch is closed, the LED has charging system voltage (or very close to it) applied to both sides. The LED does not illuminate because there is not sufficient voltage difference across the two legs.

When the switch is open, it becomes a different circuit. One leg of the LED is disconnected from battery voltage but still connected to ground through that switched circuit's load - in this case, the cluster. So, think of it as a larger load (the cluster taken as a whole) and a much smaller load (single LED) in series. The LED is very high resistance which makes it the dominant resistance to current flow. In a series situation, all loads (cluster and LED) must have the same current flow. The very low current through the LED means that there is not enough power to operate the cluster. The voltage drop across the high resistance load (LED) is very large compared to the voltage drop across the lower resistance load (cluster). The wire in between the LED and cluster is close to ground level.

So, when the switched circuit is disconnected from battery voltage, the LED sees battery voltage from the unswitched B+ source on one side and near ground on the other that is disconnected from the battery. This difference approaches the potential difference of an LED connected directly to B+ and true chassis ground. The lamp illuminates.

 

That makes sense! Thanks!

 

Just an update... whatever I did, DO NOT DO IT!!!!

I didn't drive my truck for 2 days and woke up to a dead battery. Tested at only 5 volts. Too make it more ironic, the only single thing in the truck that worked.... was the blinking light. Waiting for my battery to finish charging so I can verify what the problem actually is but it's a brand new battery and the only thing I did before this was install that damn blinky light. So there has to be a problem in this floating ground nonsense or how I achieved it.

I'll do an actual parasitic draw test in the morning once the battery is fully charged to verify the issue. Plus test the light itself etc. I may be wrong... could have left the dome or bedlights on or something but I doubt it. I'll give details tomorrow.

I knew I should have just put the stupid thing on a toggle switch!

 

Well i screwed the pooch for sure. Finally got the battery charged up and did a parasitic draw test. With my multimeter on 10 amp setting it read .56 amps. Way too high. I unplugged the dumb blinking light and it dropped to .22. Noticeable difference but still too high. Then I pulled the #33 fuse (instrument cluster battery feed b+) and it dropped to .06 which is within spec.

So i fudged up the instrument cluster circuit with this dumb light. Now the question is, how do I fix it! Ive already thrown the damn light away but the cluster is still causing a drain so something must be shorted out? Any ideas? I should mention the truck starts up and the instrument cluster is still 100% operational so that's even more confusing.